[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_conntrack_common.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 <net/netfilter/nf_conntrack_extend.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;
	struct tcphdr _tcph, *tcph = NULL;
	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, 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:
		tcph = skb_header_pointer(pkt->skb, pkt->xt.thoff,
					  sizeof(_tcph), &_tcph);
		if (unlikely(!tcph || tcph->fin || tcph->rst))
			goto out;
		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 (tcph) {
		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 const struct nla_policy nft_flow_offload_policy[NFTA_FLOW_MAX + 1] = {
	[NFTA_FLOW_TABLE_NAME]	= { .type = NLA_STRING,
				    .len = NFT_NAME_MAXLEN - 1 },
};

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,
	.policy		= nft_flow_offload_policy,
	.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>
40d102cd	9	#include <linux/netfilter/nf_conntrack_common.h>
a3c90f7a PNA	10	#include <linux/netfilter/nf_tables.h>
	11	#include <net/ip.h> /* for ipv4 options. */
	12	#include <net/netfilter/nf_tables.h>
	13	#include <net/netfilter/nf_tables_core.h>
	14	#include <net/netfilter/nf_conntrack_core.h>
40d102cd	15	#include <net/netfilter/nf_conntrack_extend.h>
a3c90f7a PNA	16	#include <net/netfilter/nf_flow_table.h>
	17
	18	struct nft_flow_offload {
	19	struct nft_flowtable *flowtable;
	20	};
	21
	22	static int nft_flow_route(const struct nft_pktinfo *pkt,
	23	const struct nf_conn *ct,
	24	struct nf_flow_route *route,
	25	enum ip_conntrack_dir dir)
	26	{
	27	struct dst_entry *this_dst = skb_dst(pkt->skb);
	28	struct dst_entry *other_dst = NULL;
	29	struct flowi fl;
	30
	31	memset(&fl, 0, sizeof(fl));
	32	switch (nft_pf(pkt)) {
	33	case NFPROTO_IPV4:
a799aea0	34	fl.u.ip4.daddr = ct->tuplehash[dir].tuple.src.u3.ip;
10f4e765	35	fl.u.ip4.flowi4_oif = nft_in(pkt)->ifindex;
a3c90f7a PNA	36	break;
a3c90f7a PNA	37	case NFPROTO_IPV6:
a799aea0	38	fl.u.ip6.daddr = ct->tuplehash[dir].tuple.src.u3.in6;
10f4e765	39	fl.u.ip6.flowi6_oif = nft_in(pkt)->ifindex;
a3c90f7a PNA	40	break;
	41	}
	42
	43	nf_route(nft_net(pkt), &other_dst, &fl, false, nft_pf(pkt));
	44	if (!other_dst)
	45	return -ENOENT;
	46
	47	route->tuple[dir].dst = this_dst;
a3c90f7a	48	route->tuple[!dir].dst = other_dst;
a3c90f7a PNA	49
	50	return 0;
	51	}
	52
69aeb538	53	static bool nft_flow_offload_skip(struct sk_buff *skb, int family)
a3c90f7a	54	{
a3c90f7a PNA	55	if (skb_sec_path(skb))
	56	return true;
	57
69aeb538 FW	58	if (family == NFPROTO_IPV4) {
	59	const struct ip_options *opt;
	60
	61	opt = &(IPCB(skb)->opt);
	62
	63	if (unlikely(opt->optlen))
	64	return true;
	65	}
	66
a3c90f7a PNA	67	return false;
	68	}
	69
	70	static void nft_flow_offload_eval(const struct nft_expr *expr,
	71	struct nft_regs *regs,
	72	const struct nft_pktinfo *pkt)
	73	{
	74	struct nft_flow_offload *priv = nft_expr_priv(expr);
	75	struct nf_flowtable *flowtable = &priv->flowtable->data;
dfe42be1	76	struct tcphdr _tcph, *tcph = NULL;
a3c90f7a PNA	77	enum ip_conntrack_info ctinfo;
	78	struct nf_flow_route route;
	79	struct flow_offload *flow;
	80	enum ip_conntrack_dir dir;
	81	struct nf_conn *ct;
	82	int ret;
	83
69aeb538	84	if (nft_flow_offload_skip(pkt->skb, nft_pf(pkt)))
a3c90f7a PNA	85	goto out;
	86
	87	ct = nf_ct_get(pkt->skb, &ctinfo);
	88	if (!ct)
	89	goto out;
	90
	91	switch (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum) {
	92	case IPPROTO_TCP:
dfe42be1 PNA	93	tcph = skb_header_pointer(pkt->skb, pkt->xt.thoff,
	94	sizeof(_tcph), &_tcph);
	95	if (unlikely(!tcph \|\| tcph->fin \|\| tcph->rst))
	96	goto out;
8437a620	97	break;
a3c90f7a PNA	98	case IPPROTO_UDP:
	99	break;
	100	default:
	101	goto out;
	102	}
	103
91a9048f FW	104	if (nf_ct_ext_exist(ct, NF_CT_EXT_HELPER) \|\|
91a9048f FW	105	ct->status & IPS_SEQ_ADJUST)
a3c90f7a PNA	106	goto out;
a3c90f7a PNA	107
270a8a29	108	if (!nf_ct_is_confirmed(ct))
a3c90f7a PNA	109	goto out;
	110
	111	if (test_and_set_bit(IPS_OFFLOAD_BIT, &ct->status))
	112	goto out;
	113
	114	dir = CTINFO2DIR(ctinfo);
	115	if (nft_flow_route(pkt, ct, &route, dir) < 0)
	116	goto err_flow_route;
	117
	118	flow = flow_offload_alloc(ct, &route);
	119	if (!flow)
	120	goto err_flow_alloc;
	121
dfe42be1	122	if (tcph) {
8437a620 FW	123	ct->proto.tcp.seen[0].flags \|= IP_CT_TCP_FLAG_BE_LIBERAL;
	124	ct->proto.tcp.seen[1].flags \|= IP_CT_TCP_FLAG_BE_LIBERAL;
	125	}
	126
a3c90f7a PNA	127	ret = flow_offload_add(flowtable, flow);
	128	if (ret < 0)
	129	goto err_flow_add;
	130
26a302af	131	dst_release(route.tuple[!dir].dst);
a3c90f7a PNA	132	return;
	133
	134	err_flow_add:
	135	flow_offload_free(flow);
	136	err_flow_alloc:
	137	dst_release(route.tuple[!dir].dst);
	138	err_flow_route:
	139	clear_bit(IPS_OFFLOAD_BIT, &ct->status);
	140	out:
	141	regs->verdict.code = NFT_BREAK;
	142	}
	143
	144	static int nft_flow_offload_validate(const struct nft_ctx *ctx,
	145	const struct nft_expr *expr,
	146	const struct nft_data **data)
	147	{
	148	unsigned int hook_mask = (1 << NF_INET_FORWARD);
	149
	150	return nft_chain_validate_hooks(ctx->chain, hook_mask);
	151	}
	152
14c41586 PNA	153	static const struct nla_policy nft_flow_offload_policy[NFTA_FLOW_MAX + 1] = {
	154	[NFTA_FLOW_TABLE_NAME] = { .type = NLA_STRING,
	155	.len = NFT_NAME_MAXLEN - 1 },
	156	};
	157
a3c90f7a PNA	158	static int nft_flow_offload_init(const struct nft_ctx *ctx,
	159	const struct nft_expr *expr,
	160	const struct nlattr * const tb[])
	161	{
	162	struct nft_flow_offload *priv = nft_expr_priv(expr);
	163	u8 genmask = nft_genmask_next(ctx->net);
	164	struct nft_flowtable *flowtable;
	165
	166	if (!tb[NFTA_FLOW_TABLE_NAME])
	167	return -EINVAL;
	168
cac20fcd PNA	169	flowtable = nft_flowtable_lookup(ctx->table, tb[NFTA_FLOW_TABLE_NAME],
cac20fcd PNA	170	genmask);
a3c90f7a PNA	171	if (IS_ERR(flowtable))
	172	return PTR_ERR(flowtable);
	173
	174	priv->flowtable = flowtable;
	175	flowtable->use++;
	176
36596dad	177	return nf_ct_netns_get(ctx->net, ctx->family);
a3c90f7a PNA	178	}
	179
	180	static void nft_flow_offload_destroy(const struct nft_ctx *ctx,
	181	const struct nft_expr *expr)
	182	{
	183	struct nft_flow_offload *priv = nft_expr_priv(expr);
	184
	185	priv->flowtable->use--;
36596dad	186	nf_ct_netns_put(ctx->net, ctx->family);
a3c90f7a PNA	187	}
	188
	189	static int nft_flow_offload_dump(struct sk_buff skb, const struct nft_expr expr)
	190	{
	191	struct nft_flow_offload *priv = nft_expr_priv(expr);
	192
	193	if (nla_put_string(skb, NFTA_FLOW_TABLE_NAME, priv->flowtable->name))
	194	goto nla_put_failure;
	195
	196	return 0;
	197
	198	nla_put_failure:
	199	return -1;
	200	}
	201
	202	static struct nft_expr_type nft_flow_offload_type;
	203	static const struct nft_expr_ops nft_flow_offload_ops = {
	204	.type = &nft_flow_offload_type,
	205	.size = NFT_EXPR_SIZE(sizeof(struct nft_flow_offload)),
	206	.eval = nft_flow_offload_eval,
	207	.init = nft_flow_offload_init,
	208	.destroy = nft_flow_offload_destroy,
	209	.validate = nft_flow_offload_validate,
	210	.dump = nft_flow_offload_dump,
	211	};
	212
	213	static struct nft_expr_type nft_flow_offload_type __read_mostly = {
	214	.name = "flow_offload",
	215	.ops = &nft_flow_offload_ops,
14c41586	216	.policy = nft_flow_offload_policy,
a3c90f7a PNA	217	.maxattr = NFTA_FLOW_MAX,
	218	.owner = THIS_MODULE,
	219	};
	220
a3c90f7a PNA	221	static int flow_offload_netdev_event(struct notifier_block *this,
	222	unsigned long event, void *ptr)
	223	{
	224	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
	225
	226	if (event != NETDEV_DOWN)
	227	return NOTIFY_DONE;
	228
5f1be84a	229	nf_flow_table_cleanup(dev);
a3c90f7a PNA	230
	231	return NOTIFY_DONE;
	232	}
	233
	234	static struct notifier_block flow_offload_netdev_notifier = {
	235	.notifier_call = flow_offload_netdev_event,
	236	};
	237
	238	static int __init nft_flow_offload_module_init(void)
	239	{
	240	int err;
	241
584eab29 TY	242	err = register_netdevice_notifier(&flow_offload_netdev_notifier);
	243	if (err)
	244	goto err;
a3c90f7a PNA	245
	246	err = nft_register_expr(&nft_flow_offload_type);
	247	if (err < 0)
	248	goto register_expr;
	249
	250	return 0;
	251
	252	register_expr:
	253	unregister_netdevice_notifier(&flow_offload_netdev_notifier);
584eab29	254	err:
a3c90f7a PNA	255	return err;
	256	}
	257
	258	static void __exit nft_flow_offload_module_exit(void)
	259	{
a3c90f7a PNA	260	nft_unregister_expr(&nft_flow_offload_type);
a3c90f7a PNA	261	unregister_netdevice_notifier(&flow_offload_netdev_notifier);
a3c90f7a PNA	262	}
	263
	264	module_init(nft_flow_offload_module_init);
	265	module_exit(nft_flow_offload_module_exit);
	266
	267	MODULE_LICENSE("GPL");
	268	MODULE_AUTHOR("Pablo Neira Ayuso <pablo@netfilter.org>");
	269	MODULE_ALIAS_NFT_EXPR("flow_offload");