[linux-2.6-block.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>
#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
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>
2314e879	16	#include <net/netfilter/nf_conntrack_helper.h>
a3c90f7a PNA	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
	53	static bool nft_flow_offload_skip(struct sk_buff *skb)
	54	{
	55	struct ip_options *opt = &(IPCB(skb)->opt);
	56
	57	if (unlikely(opt->optlen))
	58	return true;
	59	if (skb_sec_path(skb))
	60	return true;
	61
	62	return false;
	63	}
	64
	65	static void nft_flow_offload_eval(const struct nft_expr *expr,
	66	struct nft_regs *regs,
	67	const struct nft_pktinfo *pkt)
	68	{
	69	struct nft_flow_offload *priv = nft_expr_priv(expr);
	70	struct nf_flowtable *flowtable = &priv->flowtable->data;
2314e879	71	const struct nf_conn_help *help;
a3c90f7a PNA	72	enum ip_conntrack_info ctinfo;
	73	struct nf_flow_route route;
	74	struct flow_offload *flow;
	75	enum ip_conntrack_dir dir;
	76	struct nf_conn *ct;
	77	int ret;
	78
	79	if (nft_flow_offload_skip(pkt->skb))
	80	goto out;
	81
	82	ct = nf_ct_get(pkt->skb, &ctinfo);
	83	if (!ct)
	84	goto out;
	85
	86	switch (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum) {
	87	case IPPROTO_TCP:
	88	case IPPROTO_UDP:
	89	break;
	90	default:
	91	goto out;
	92	}
	93
2314e879 HY	94	help = nfct_help(ct);
2314e879 HY	95	if (help)
a3c90f7a PNA	96	goto out;
a3c90f7a PNA	97
270a8a29	98	if (!nf_ct_is_confirmed(ct))
a3c90f7a PNA	99	goto out;
	100
	101	if (test_and_set_bit(IPS_OFFLOAD_BIT, &ct->status))
	102	goto out;
	103
	104	dir = CTINFO2DIR(ctinfo);
	105	if (nft_flow_route(pkt, ct, &route, dir) < 0)
	106	goto err_flow_route;
	107
	108	flow = flow_offload_alloc(ct, &route);
	109	if (!flow)
	110	goto err_flow_alloc;
	111
	112	ret = flow_offload_add(flowtable, flow);
	113	if (ret < 0)
	114	goto err_flow_add;
	115
26a302af	116	dst_release(route.tuple[!dir].dst);
a3c90f7a PNA	117	return;
	118
	119	err_flow_add:
	120	flow_offload_free(flow);
	121	err_flow_alloc:
	122	dst_release(route.tuple[!dir].dst);
	123	err_flow_route:
	124	clear_bit(IPS_OFFLOAD_BIT, &ct->status);
	125	out:
	126	regs->verdict.code = NFT_BREAK;
	127	}
	128
	129	static int nft_flow_offload_validate(const struct nft_ctx *ctx,
	130	const struct nft_expr *expr,
	131	const struct nft_data **data)
	132	{
	133	unsigned int hook_mask = (1 << NF_INET_FORWARD);
	134
	135	return nft_chain_validate_hooks(ctx->chain, hook_mask);
	136	}
	137
	138	static int nft_flow_offload_init(const struct nft_ctx *ctx,
	139	const struct nft_expr *expr,
	140	const struct nlattr * const tb[])
	141	{
	142	struct nft_flow_offload *priv = nft_expr_priv(expr);
	143	u8 genmask = nft_genmask_next(ctx->net);
	144	struct nft_flowtable *flowtable;
	145
	146	if (!tb[NFTA_FLOW_TABLE_NAME])
	147	return -EINVAL;
	148
cac20fcd PNA	149	flowtable = nft_flowtable_lookup(ctx->table, tb[NFTA_FLOW_TABLE_NAME],
cac20fcd PNA	150	genmask);
a3c90f7a PNA	151	if (IS_ERR(flowtable))
	152	return PTR_ERR(flowtable);
	153
	154	priv->flowtable = flowtable;
	155	flowtable->use++;
	156
36596dad	157	return nf_ct_netns_get(ctx->net, ctx->family);
a3c90f7a PNA	158	}
	159
	160	static void nft_flow_offload_destroy(const struct nft_ctx *ctx,
	161	const struct nft_expr *expr)
	162	{
	163	struct nft_flow_offload *priv = nft_expr_priv(expr);
	164
	165	priv->flowtable->use--;
36596dad	166	nf_ct_netns_put(ctx->net, ctx->family);
a3c90f7a PNA	167	}
	168
	169	static int nft_flow_offload_dump(struct sk_buff skb, const struct nft_expr expr)
	170	{
	171	struct nft_flow_offload *priv = nft_expr_priv(expr);
	172
	173	if (nla_put_string(skb, NFTA_FLOW_TABLE_NAME, priv->flowtable->name))
	174	goto nla_put_failure;
	175
	176	return 0;
	177
	178	nla_put_failure:
	179	return -1;
	180	}
	181
	182	static struct nft_expr_type nft_flow_offload_type;
	183	static const struct nft_expr_ops nft_flow_offload_ops = {
	184	.type = &nft_flow_offload_type,
	185	.size = NFT_EXPR_SIZE(sizeof(struct nft_flow_offload)),
	186	.eval = nft_flow_offload_eval,
	187	.init = nft_flow_offload_init,
	188	.destroy = nft_flow_offload_destroy,
	189	.validate = nft_flow_offload_validate,
	190	.dump = nft_flow_offload_dump,
	191	};
	192
	193	static struct nft_expr_type nft_flow_offload_type __read_mostly = {
	194	.name = "flow_offload",
	195	.ops = &nft_flow_offload_ops,
	196	.maxattr = NFTA_FLOW_MAX,
	197	.owner = THIS_MODULE,
	198	};
	199
a3c90f7a PNA	200	static int flow_offload_netdev_event(struct notifier_block *this,
	201	unsigned long event, void *ptr)
	202	{
	203	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
	204
	205	if (event != NETDEV_DOWN)
	206	return NOTIFY_DONE;
	207
5f1be84a	208	nf_flow_table_cleanup(dev);
a3c90f7a PNA	209
	210	return NOTIFY_DONE;
	211	}
	212
	213	static struct notifier_block flow_offload_netdev_notifier = {
	214	.notifier_call = flow_offload_netdev_event,
	215	};
	216
	217	static int __init nft_flow_offload_module_init(void)
	218	{
	219	int err;
	220
584eab29 TY	221	err = register_netdevice_notifier(&flow_offload_netdev_notifier);
	222	if (err)
	223	goto err;
a3c90f7a PNA	224
	225	err = nft_register_expr(&nft_flow_offload_type);
	226	if (err < 0)
	227	goto register_expr;
	228
	229	return 0;
	230
	231	register_expr:
	232	unregister_netdevice_notifier(&flow_offload_netdev_notifier);
584eab29	233	err:
a3c90f7a PNA	234	return err;
	235	}
	236
	237	static void __exit nft_flow_offload_module_exit(void)
	238	{
a3c90f7a PNA	239	nft_unregister_expr(&nft_flow_offload_type);
a3c90f7a PNA	240	unregister_netdevice_notifier(&flow_offload_netdev_notifier);
a3c90f7a PNA	241	}
	242
	243	module_init(nft_flow_offload_module_init);
	244	module_exit(nft_flow_offload_module_exit);
	245
	246	MODULE_LICENSE("GPL");
	247	MODULE_AUTHOR("Pablo Neira Ayuso <pablo@netfilter.org>");
	248	MODULE_ALIAS_NFT_EXPR("flow_offload");