[mirror_ubuntu-jammy-kernel.git] / 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>
#include <net/netfilter/nf_conntrack_helper.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)
{
	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;
	const struct nf_conn_help *help;
	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;
	}

	help = nfct_help(ct);
	if (help)
		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;

	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
a3c90f7a PNA	1	#include <linux/kernel.h>
	2	#include <linux/module.h>
	3	#include <linux/init.h>
	4	#include <linux/netlink.h>
	5	#include <linux/netfilter.h>
	6	#include <linux/workqueue.h>
	7	#include <linux/spinlock.h>
	8	#include <linux/netfilter/nf_tables.h>
	9	#include <net/ip.h> /* for ipv4 options. */
	10	#include <net/netfilter/nf_tables.h>
	11	#include <net/netfilter/nf_tables_core.h>
	12	#include <net/netfilter/nf_conntrack_core.h>
	13	#include <linux/netfilter/nf_conntrack_common.h>
	14	#include <net/netfilter/nf_flow_table.h>
2314e879	15	#include <net/netfilter/nf_conntrack_helper.h>
a3c90f7a PNA	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
	52	static bool nft_flow_offload_skip(struct sk_buff *skb)
	53	{
	54	struct ip_options *opt = &(IPCB(skb)->opt);
	55
	56	if (unlikely(opt->optlen))
	57	return true;
	58	if (skb_sec_path(skb))
	59	return true;
	60
	61	return false;
	62	}
	63
	64	static void nft_flow_offload_eval(const struct nft_expr *expr,
	65	struct nft_regs *regs,
	66	const struct nft_pktinfo *pkt)
	67	{
	68	struct nft_flow_offload *priv = nft_expr_priv(expr);
	69	struct nf_flowtable *flowtable = &priv->flowtable->data;
2314e879	70	const struct nf_conn_help *help;
a3c90f7a PNA	71	enum ip_conntrack_info ctinfo;
	72	struct nf_flow_route route;
	73	struct flow_offload *flow;
	74	enum ip_conntrack_dir dir;
	75	struct nf_conn *ct;
	76	int ret;
	77
	78	if (nft_flow_offload_skip(pkt->skb))
	79	goto out;
	80
	81	ct = nf_ct_get(pkt->skb, &ctinfo);
	82	if (!ct)
	83	goto out;
	84
	85	switch (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum) {
	86	case IPPROTO_TCP:
	87	case IPPROTO_UDP:
	88	break;
	89	default:
	90	goto out;
	91	}
	92
2314e879 HY	93	help = nfct_help(ct);
2314e879 HY	94	if (help)
a3c90f7a PNA	95	goto out;
a3c90f7a PNA	96
270a8a29	97	if (!nf_ct_is_confirmed(ct))
a3c90f7a PNA	98	goto out;
	99
	100	if (test_and_set_bit(IPS_OFFLOAD_BIT, &ct->status))
	101	goto out;
	102
	103	dir = CTINFO2DIR(ctinfo);
	104	if (nft_flow_route(pkt, ct, &route, dir) < 0)
	105	goto err_flow_route;
	106
	107	flow = flow_offload_alloc(ct, &route);
	108	if (!flow)
	109	goto err_flow_alloc;
	110
	111	ret = flow_offload_add(flowtable, flow);
	112	if (ret < 0)
	113	goto err_flow_add;
	114
26a302af	115	dst_release(route.tuple[!dir].dst);
a3c90f7a PNA	116	return;
	117
	118	err_flow_add:
	119	flow_offload_free(flow);
	120	err_flow_alloc:
	121	dst_release(route.tuple[!dir].dst);
	122	err_flow_route:
	123	clear_bit(IPS_OFFLOAD_BIT, &ct->status);
	124	out:
	125	regs->verdict.code = NFT_BREAK;
	126	}
	127
	128	static int nft_flow_offload_validate(const struct nft_ctx *ctx,
	129	const struct nft_expr *expr,
	130	const struct nft_data **data)
	131	{
	132	unsigned int hook_mask = (1 << NF_INET_FORWARD);
	133
	134	return nft_chain_validate_hooks(ctx->chain, hook_mask);
	135	}
	136
	137	static int nft_flow_offload_init(const struct nft_ctx *ctx,
	138	const struct nft_expr *expr,
	139	const struct nlattr * const tb[])
	140	{
	141	struct nft_flow_offload *priv = nft_expr_priv(expr);
	142	u8 genmask = nft_genmask_next(ctx->net);
	143	struct nft_flowtable *flowtable;
	144
	145	if (!tb[NFTA_FLOW_TABLE_NAME])
	146	return -EINVAL;
	147
cac20fcd PNA	148	flowtable = nft_flowtable_lookup(ctx->table, tb[NFTA_FLOW_TABLE_NAME],
cac20fcd PNA	149	genmask);
a3c90f7a PNA	150	if (IS_ERR(flowtable))
	151	return PTR_ERR(flowtable);
	152
	153	priv->flowtable = flowtable;
	154	flowtable->use++;
	155
36596dad	156	return nf_ct_netns_get(ctx->net, ctx->family);
a3c90f7a PNA	157	}
	158
	159	static void nft_flow_offload_destroy(const struct nft_ctx *ctx,
	160	const struct nft_expr *expr)
	161	{
	162	struct nft_flow_offload *priv = nft_expr_priv(expr);
	163
	164	priv->flowtable->use--;
36596dad	165	nf_ct_netns_put(ctx->net, ctx->family);
a3c90f7a PNA	166	}
	167
	168	static int nft_flow_offload_dump(struct sk_buff skb, const struct nft_expr expr)
	169	{
	170	struct nft_flow_offload *priv = nft_expr_priv(expr);
	171
	172	if (nla_put_string(skb, NFTA_FLOW_TABLE_NAME, priv->flowtable->name))
	173	goto nla_put_failure;
	174
	175	return 0;
	176
	177	nla_put_failure:
	178	return -1;
	179	}
	180
	181	static struct nft_expr_type nft_flow_offload_type;
	182	static const struct nft_expr_ops nft_flow_offload_ops = {
	183	.type = &nft_flow_offload_type,
	184	.size = NFT_EXPR_SIZE(sizeof(struct nft_flow_offload)),
	185	.eval = nft_flow_offload_eval,
	186	.init = nft_flow_offload_init,
	187	.destroy = nft_flow_offload_destroy,
	188	.validate = nft_flow_offload_validate,
	189	.dump = nft_flow_offload_dump,
	190	};
	191
	192	static struct nft_expr_type nft_flow_offload_type __read_mostly = {
	193	.name = "flow_offload",
	194	.ops = &nft_flow_offload_ops,
	195	.maxattr = NFTA_FLOW_MAX,
	196	.owner = THIS_MODULE,
	197	};
	198
a3c90f7a PNA	199	static int flow_offload_netdev_event(struct notifier_block *this,
	200	unsigned long event, void *ptr)
	201	{
	202	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
	203
	204	if (event != NETDEV_DOWN)
	205	return NOTIFY_DONE;
	206
5f1be84a	207	nf_flow_table_cleanup(dev);
a3c90f7a PNA	208
	209	return NOTIFY_DONE;
	210	}
	211
	212	static struct notifier_block flow_offload_netdev_notifier = {
	213	.notifier_call = flow_offload_netdev_event,
	214	};
	215
	216	static int __init nft_flow_offload_module_init(void)
	217	{
	218	int err;
	219
584eab29 TY	220	err = register_netdevice_notifier(&flow_offload_netdev_notifier);
	221	if (err)
	222	goto err;
a3c90f7a PNA	223
	224	err = nft_register_expr(&nft_flow_offload_type);
	225	if (err < 0)
	226	goto register_expr;
	227
	228	return 0;
	229
	230	register_expr:
	231	unregister_netdevice_notifier(&flow_offload_netdev_notifier);
584eab29	232	err:
a3c90f7a PNA	233	return err;
	234	}
	235
	236	static void __exit nft_flow_offload_module_exit(void)
	237	{
a3c90f7a PNA	238	nft_unregister_expr(&nft_flow_offload_type);
a3c90f7a PNA	239	unregister_netdevice_notifier(&flow_offload_netdev_notifier);
a3c90f7a PNA	240	}
	241
	242	module_init(nft_flow_offload_module_init);
	243	module_exit(nft_flow_offload_module_exit);
	244
	245	MODULE_LICENSE("GPL");
	246	MODULE_AUTHOR("Pablo Neira Ayuso <pablo@netfilter.org>");
	247	MODULE_ALIAS_NFT_EXPR("flow_offload");