[mirror_ubuntu-jammy-kernel.git] / net / netfilter / nft_flow_offload.c

// SPDX-License-Identifier: GPL-2.0-only
#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.src.u3.ip;
		fl.u.ip4.flowi4_oif = nft_in(pkt)->ifindex;
		break;
	case NFPROTO_IPV6:
		fl.u.ip6.daddr = ct->tuplehash[dir].tuple.src.u3.in6;
		fl.u.ip6.flowi6_oif = nft_in(pkt)->ifindex;
		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].dst		= other_dst;

	return 0;
}

static bool nft_flow_offload_skip(struct sk_buff *skb, int family)
{
	if (skb_sec_path(skb))
		return true;

	if (family == NFPROTO_IPV4) {
		const struct ip_options *opt;

		opt = &(IPCB(skb)->opt);

		if (unlikely(opt->optlen))
			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;
	bool is_tcp = false;
	struct nf_conn *ct;
	int ret;

	if (nft_flow_offload_skip(pkt->skb, nft_pf(pkt)))
		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:
		is_tcp = true;
		break;
	case IPPROTO_UDP:
		break;
	default:
		goto out;
	}

	if (nf_ct_ext_exist(ct, NF_CT_EXT_HELPER) ||
	    ct->status & IPS_SEQ_ADJUST)
		goto out;

	if (!nf_ct_is_confirmed(ct))
		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;

	if (is_tcp) {
		ct->proto.tcp.seen[0].flags |= IP_CT_TCP_FLAG_BE_LIBERAL;
		ct->proto.tcp.seen[1].flags |= IP_CT_TCP_FLAG_BE_LIBERAL;
	}

	ret = flow_offload_add(flowtable, flow);
	if (ret < 0)
		goto err_flow_add;

	dst_release(route.tuple[!dir].dst);
	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 = nft_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->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->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 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;

	nf_flow_table_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;

	err = register_netdevice_notifier(&flow_offload_netdev_notifier);
	if (err)
		goto err;

	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);
err:
	return err;
}

static void __exit nft_flow_offload_module_exit(void)
{
	nft_unregister_expr(&nft_flow_offload_type);
	unregister_netdevice_notifier(&flow_offload_netdev_notifier);
}

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");
Commit	Line	Data
09c434b8	1	// SPDX-License-Identifier: GPL-2.0-only
a3c90f7a PNA	2	#include <linux/kernel.h>
	3	#include <linux/module.h>
	4	#include <linux/init.h>
	5	#include <linux/netlink.h>
	6	#include <linux/netfilter.h>
	7	#include <linux/workqueue.h>
	8	#include <linux/spinlock.h>
	9	#include <linux/netfilter/nf_tables.h>
	10	#include <net/ip.h> /* for ipv4 options. */
	11	#include <net/netfilter/nf_tables.h>
	12	#include <net/netfilter/nf_tables_core.h>
	13	#include <net/netfilter/nf_conntrack_core.h>
	14	#include <linux/netfilter/nf_conntrack_common.h>
	15	#include <net/netfilter/nf_flow_table.h>
	16
	17	struct nft_flow_offload {
	18	struct nft_flowtable *flowtable;
	19	};
	20
	21	static int nft_flow_route(const struct nft_pktinfo *pkt,
	22	const struct nf_conn *ct,
	23	struct nf_flow_route *route,
	24	enum ip_conntrack_dir dir)
	25	{
	26	struct dst_entry *this_dst = skb_dst(pkt->skb);
	27	struct dst_entry *other_dst = NULL;
	28	struct flowi fl;
	29
	30	memset(&fl, 0, sizeof(fl));
	31	switch (nft_pf(pkt)) {
	32	case NFPROTO_IPV4:
a799aea0	33	fl.u.ip4.daddr = ct->tuplehash[dir].tuple.src.u3.ip;
10f4e765	34	fl.u.ip4.flowi4_oif = nft_in(pkt)->ifindex;
a3c90f7a PNA	35	break;
a3c90f7a PNA	36	case NFPROTO_IPV6:
a799aea0	37	fl.u.ip6.daddr = ct->tuplehash[dir].tuple.src.u3.in6;
10f4e765	38	fl.u.ip6.flowi6_oif = nft_in(pkt)->ifindex;
a3c90f7a PNA	39	break;
	40	}
	41
	42	nf_route(nft_net(pkt), &other_dst, &fl, false, nft_pf(pkt));
	43	if (!other_dst)
	44	return -ENOENT;
	45
	46	route->tuple[dir].dst = this_dst;
a3c90f7a	47	route->tuple[!dir].dst = other_dst;
a3c90f7a PNA	48
	49	return 0;
	50	}
	51
69aeb538	52	static bool nft_flow_offload_skip(struct sk_buff *skb, int family)
a3c90f7a	53	{
a3c90f7a PNA	54	if (skb_sec_path(skb))
	55	return true;
	56
69aeb538 FW	57	if (family == NFPROTO_IPV4) {
	58	const struct ip_options *opt;
	59
	60	opt = &(IPCB(skb)->opt);
	61
	62	if (unlikely(opt->optlen))
	63	return true;
	64	}
	65
a3c90f7a PNA	66	return false;
	67	}
	68
	69	static void nft_flow_offload_eval(const struct nft_expr *expr,
	70	struct nft_regs *regs,
	71	const struct nft_pktinfo *pkt)
	72	{
	73	struct nft_flow_offload *priv = nft_expr_priv(expr);
	74	struct nf_flowtable *flowtable = &priv->flowtable->data;
	75	enum ip_conntrack_info ctinfo;
	76	struct nf_flow_route route;
	77	struct flow_offload *flow;
	78	enum ip_conntrack_dir dir;
8437a620	79	bool is_tcp = false;
a3c90f7a PNA	80	struct nf_conn *ct;
	81	int ret;
	82
69aeb538	83	if (nft_flow_offload_skip(pkt->skb, nft_pf(pkt)))
a3c90f7a PNA	84	goto out;
	85
	86	ct = nf_ct_get(pkt->skb, &ctinfo);
	87	if (!ct)
	88	goto out;
	89
	90	switch (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum) {
	91	case IPPROTO_TCP:
8437a620 FW	92	is_tcp = true;
8437a620 FW	93	break;
a3c90f7a PNA	94	case IPPROTO_UDP:
	95	break;
	96	default:
	97	goto out;
	98	}
	99
91a9048f FW	100	if (nf_ct_ext_exist(ct, NF_CT_EXT_HELPER) \|\|
91a9048f FW	101	ct->status & IPS_SEQ_ADJUST)
a3c90f7a PNA	102	goto out;
a3c90f7a PNA	103
270a8a29	104	if (!nf_ct_is_confirmed(ct))
a3c90f7a PNA	105	goto out;
	106
	107	if (test_and_set_bit(IPS_OFFLOAD_BIT, &ct->status))
	108	goto out;
	109
	110	dir = CTINFO2DIR(ctinfo);
	111	if (nft_flow_route(pkt, ct, &route, dir) < 0)
	112	goto err_flow_route;
	113
	114	flow = flow_offload_alloc(ct, &route);
	115	if (!flow)
	116	goto err_flow_alloc;
	117
8437a620 FW	118	if (is_tcp) {
	119	ct->proto.tcp.seen[0].flags \|= IP_CT_TCP_FLAG_BE_LIBERAL;
	120	ct->proto.tcp.seen[1].flags \|= IP_CT_TCP_FLAG_BE_LIBERAL;
	121	}
	122
a3c90f7a PNA	123	ret = flow_offload_add(flowtable, flow);
	124	if (ret < 0)
	125	goto err_flow_add;
	126
26a302af	127	dst_release(route.tuple[!dir].dst);
a3c90f7a PNA	128	return;
	129
	130	err_flow_add:
	131	flow_offload_free(flow);
	132	err_flow_alloc:
	133	dst_release(route.tuple[!dir].dst);
	134	err_flow_route:
	135	clear_bit(IPS_OFFLOAD_BIT, &ct->status);
	136	out:
	137	regs->verdict.code = NFT_BREAK;
	138	}
	139
	140	static int nft_flow_offload_validate(const struct nft_ctx *ctx,
	141	const struct nft_expr *expr,
	142	const struct nft_data **data)
	143	{
	144	unsigned int hook_mask = (1 << NF_INET_FORWARD);
	145
	146	return nft_chain_validate_hooks(ctx->chain, hook_mask);
	147	}
	148
	149	static int nft_flow_offload_init(const struct nft_ctx *ctx,
	150	const struct nft_expr *expr,
	151	const struct nlattr * const tb[])
	152	{
	153	struct nft_flow_offload *priv = nft_expr_priv(expr);
	154	u8 genmask = nft_genmask_next(ctx->net);
	155	struct nft_flowtable *flowtable;
	156
	157	if (!tb[NFTA_FLOW_TABLE_NAME])
	158	return -EINVAL;
	159
cac20fcd PNA	160	flowtable = nft_flowtable_lookup(ctx->table, tb[NFTA_FLOW_TABLE_NAME],
cac20fcd PNA	161	genmask);
a3c90f7a PNA	162	if (IS_ERR(flowtable))
	163	return PTR_ERR(flowtable);
	164
	165	priv->flowtable = flowtable;
	166	flowtable->use++;
	167
36596dad	168	return nf_ct_netns_get(ctx->net, ctx->family);
a3c90f7a PNA	169	}
	170
	171	static void nft_flow_offload_destroy(const struct nft_ctx *ctx,
	172	const struct nft_expr *expr)
	173	{
	174	struct nft_flow_offload *priv = nft_expr_priv(expr);
	175
	176	priv->flowtable->use--;
36596dad	177	nf_ct_netns_put(ctx->net, ctx->family);
a3c90f7a PNA	178	}
	179
	180	static int nft_flow_offload_dump(struct sk_buff skb, const struct nft_expr expr)
	181	{
	182	struct nft_flow_offload *priv = nft_expr_priv(expr);
	183
	184	if (nla_put_string(skb, NFTA_FLOW_TABLE_NAME, priv->flowtable->name))
	185	goto nla_put_failure;
	186
	187	return 0;
	188
	189	nla_put_failure:
	190	return -1;
	191	}
	192
	193	static struct nft_expr_type nft_flow_offload_type;
	194	static const struct nft_expr_ops nft_flow_offload_ops = {
	195	.type = &nft_flow_offload_type,
	196	.size = NFT_EXPR_SIZE(sizeof(struct nft_flow_offload)),
	197	.eval = nft_flow_offload_eval,
	198	.init = nft_flow_offload_init,
	199	.destroy = nft_flow_offload_destroy,
	200	.validate = nft_flow_offload_validate,
	201	.dump = nft_flow_offload_dump,
	202	};
	203
	204	static struct nft_expr_type nft_flow_offload_type __read_mostly = {
	205	.name = "flow_offload",
	206	.ops = &nft_flow_offload_ops,
	207	.maxattr = NFTA_FLOW_MAX,
	208	.owner = THIS_MODULE,
	209	};
	210
a3c90f7a PNA	211	static int flow_offload_netdev_event(struct notifier_block *this,
	212	unsigned long event, void *ptr)
	213	{
	214	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
	215
	216	if (event != NETDEV_DOWN)
	217	return NOTIFY_DONE;
	218
5f1be84a	219	nf_flow_table_cleanup(dev);
a3c90f7a PNA	220
	221	return NOTIFY_DONE;
	222	}
	223
	224	static struct notifier_block flow_offload_netdev_notifier = {
	225	.notifier_call = flow_offload_netdev_event,
	226	};
	227
	228	static int __init nft_flow_offload_module_init(void)
	229	{
	230	int err;
	231
584eab29 TY	232	err = register_netdevice_notifier(&flow_offload_netdev_notifier);
	233	if (err)
	234	goto err;
a3c90f7a PNA	235
	236	err = nft_register_expr(&nft_flow_offload_type);
	237	if (err < 0)
	238	goto register_expr;
	239
	240	return 0;
	241
	242	register_expr:
	243	unregister_netdevice_notifier(&flow_offload_netdev_notifier);
584eab29	244	err:
a3c90f7a PNA	245	return err;
	246	}
	247
	248	static void __exit nft_flow_offload_module_exit(void)
	249	{
a3c90f7a PNA	250	nft_unregister_expr(&nft_flow_offload_type);
a3c90f7a PNA	251	unregister_netdevice_notifier(&flow_offload_netdev_notifier);
a3c90f7a PNA	252	}
	253
	254	module_init(nft_flow_offload_module_init);
	255	module_exit(nft_flow_offload_module_exit);
	256
	257	MODULE_LICENSE("GPL");
	258	MODULE_AUTHOR("Pablo Neira Ayuso <pablo@netfilter.org>");
	259	MODULE_ALIAS_NFT_EXPR("flow_offload");