[thirdparty/kernel/stable.git] / net / sched / act_connmark.c

/*
 * net/sched/act_connmark.c  netfilter connmark retriever action
 * skb mark is over-written
 *
 * Copyright (c) 2011 Felix Fietkau <nbd@openwrt.org>
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
*/

#include <linux/module.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/skbuff.h>
#include <linux/rtnetlink.h>
#include <linux/pkt_cls.h>
#include <linux/ip.h>
#include <linux/ipv6.h>
#include <net/netlink.h>
#include <net/pkt_sched.h>
#include <net/act_api.h>
#include <net/pkt_cls.h>
#include <uapi/linux/tc_act/tc_connmark.h>
#include <net/tc_act/tc_connmark.h>

#include <net/netfilter/nf_conntrack.h>
#include <net/netfilter/nf_conntrack_core.h>
#include <net/netfilter/nf_conntrack_zones.h>

static unsigned int connmark_net_id;
static struct tc_action_ops act_connmark_ops;

static int tcf_connmark_act(struct sk_buff *skb, const struct tc_action *a,
			    struct tcf_result *res)
{
	const struct nf_conntrack_tuple_hash *thash;
	struct nf_conntrack_tuple tuple;
	enum ip_conntrack_info ctinfo;
	struct tcf_connmark_info *ca = to_connmark(a);
	struct nf_conntrack_zone zone;
	struct nf_conn *c;
	int proto;

	spin_lock(&ca->tcf_lock);
	tcf_lastuse_update(&ca->tcf_tm);
	bstats_update(&ca->tcf_bstats, skb);

	if (skb->protocol == htons(ETH_P_IP)) {
		if (skb->len < sizeof(struct iphdr))
			goto out;

		proto = NFPROTO_IPV4;
	} else if (skb->protocol == htons(ETH_P_IPV6)) {
		if (skb->len < sizeof(struct ipv6hdr))
			goto out;

		proto = NFPROTO_IPV6;
	} else {
		goto out;
	}

	c = nf_ct_get(skb, &ctinfo);
	if (c) {
		skb->mark = c->mark;
		/* using overlimits stats to count how many packets marked */
		ca->tcf_qstats.overlimits++;
		goto out;
	}

	if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb),
			       proto, ca->net, &tuple))
		goto out;

	zone.id = ca->zone;
	zone.dir = NF_CT_DEFAULT_ZONE_DIR;

	thash = nf_conntrack_find_get(ca->net, &zone, &tuple);
	if (!thash)
		goto out;

	c = nf_ct_tuplehash_to_ctrack(thash);
	/* using overlimits stats to count how many packets marked */
	ca->tcf_qstats.overlimits++;
	skb->mark = c->mark;
	nf_ct_put(c);

out:
	spin_unlock(&ca->tcf_lock);
	return ca->tcf_action;
}

static const struct nla_policy connmark_policy[TCA_CONNMARK_MAX + 1] = {
	[TCA_CONNMARK_PARMS] = { .len = sizeof(struct tc_connmark) },
};

static int tcf_connmark_init(struct net *net, struct nlattr *nla,
			     struct nlattr *est, struct tc_action **a,
			     int ovr, int bind, bool rtnl_held,
			     struct tcf_proto *tp,
			     struct netlink_ext_ack *extack)
{
	struct tc_action_net *tn = net_generic(net, connmark_net_id);
	struct nlattr *tb[TCA_CONNMARK_MAX + 1];
	struct tcf_chain *goto_ch = NULL;
	struct tcf_connmark_info *ci;
	struct tc_connmark *parm;
	int ret = 0, err;

	if (!nla)
		return -EINVAL;

	ret = nla_parse_nested(tb, TCA_CONNMARK_MAX, nla, connmark_policy,
			       NULL);
	if (ret < 0)
		return ret;

	if (!tb[TCA_CONNMARK_PARMS])
		return -EINVAL;

	parm = nla_data(tb[TCA_CONNMARK_PARMS]);

	ret = tcf_idr_check_alloc(tn, &parm->index, a, bind);
	if (!ret) {
		ret = tcf_idr_create(tn, parm->index, est, a,
				     &act_connmark_ops, bind, false);
		if (ret) {
			tcf_idr_cleanup(tn, parm->index);
			return ret;
		}

		ci = to_connmark(*a);
		err = tcf_action_check_ctrlact(parm->action, tp, &goto_ch,
					       extack);
		if (err < 0)
			goto release_idr;
		tcf_action_set_ctrlact(*a, parm->action, goto_ch);
		ci->net = net;
		ci->zone = parm->zone;

		tcf_idr_insert(tn, *a);
		ret = ACT_P_CREATED;
	} else if (ret > 0) {
		ci = to_connmark(*a);
		if (bind)
			return 0;
		if (!ovr) {
			tcf_idr_release(*a, bind);
			return -EEXIST;
		}
		err = tcf_action_check_ctrlact(parm->action, tp, &goto_ch,
					       extack);
		if (err < 0)
			goto release_idr;
		/* replacing action and zone */
		spin_lock_bh(&ci->tcf_lock);
		goto_ch = tcf_action_set_ctrlact(*a, parm->action, goto_ch);
		ci->zone = parm->zone;
		spin_unlock_bh(&ci->tcf_lock);
		if (goto_ch)
			tcf_chain_put_by_act(goto_ch);
		ret = 0;
	}

	return ret;
release_idr:
	tcf_idr_release(*a, bind);
	return err;
}

static inline int tcf_connmark_dump(struct sk_buff *skb, struct tc_action *a,
				    int bind, int ref)
{
	unsigned char *b = skb_tail_pointer(skb);
	struct tcf_connmark_info *ci = to_connmark(a);
	struct tc_connmark opt = {
		.index   = ci->tcf_index,
		.refcnt  = refcount_read(&ci->tcf_refcnt) - ref,
		.bindcnt = atomic_read(&ci->tcf_bindcnt) - bind,
	};
	struct tcf_t t;

	spin_lock_bh(&ci->tcf_lock);
	opt.action = ci->tcf_action;
	opt.zone = ci->zone;
	if (nla_put(skb, TCA_CONNMARK_PARMS, sizeof(opt), &opt))
		goto nla_put_failure;

	tcf_tm_dump(&t, &ci->tcf_tm);
	if (nla_put_64bit(skb, TCA_CONNMARK_TM, sizeof(t), &t,
			  TCA_CONNMARK_PAD))
		goto nla_put_failure;
	spin_unlock_bh(&ci->tcf_lock);

	return skb->len;

nla_put_failure:
	spin_unlock_bh(&ci->tcf_lock);
	nlmsg_trim(skb, b);
	return -1;
}

static int tcf_connmark_walker(struct net *net, struct sk_buff *skb,
			       struct netlink_callback *cb, int type,
			       const struct tc_action_ops *ops,
			       struct netlink_ext_ack *extack)
{
	struct tc_action_net *tn = net_generic(net, connmark_net_id);

	return tcf_generic_walker(tn, skb, cb, type, ops, extack);
}

static int tcf_connmark_search(struct net *net, struct tc_action **a, u32 index)
{
	struct tc_action_net *tn = net_generic(net, connmark_net_id);

	return tcf_idr_search(tn, a, index);
}

static struct tc_action_ops act_connmark_ops = {
	.kind		=	"connmark",
	.id		=	TCA_ID_CONNMARK,
	.owner		=	THIS_MODULE,
	.act		=	tcf_connmark_act,
	.dump		=	tcf_connmark_dump,
	.init		=	tcf_connmark_init,
	.walk		=	tcf_connmark_walker,
	.lookup		=	tcf_connmark_search,
	.size		=	sizeof(struct tcf_connmark_info),
};

static __net_init int connmark_init_net(struct net *net)
{
	struct tc_action_net *tn = net_generic(net, connmark_net_id);

	return tc_action_net_init(tn, &act_connmark_ops);
}

static void __net_exit connmark_exit_net(struct list_head *net_list)
{
	tc_action_net_exit(net_list, connmark_net_id);
}

static struct pernet_operations connmark_net_ops = {
	.init = connmark_init_net,
	.exit_batch = connmark_exit_net,
	.id   = &connmark_net_id,
	.size = sizeof(struct tc_action_net),
};

static int __init connmark_init_module(void)
{
	return tcf_register_action(&act_connmark_ops, &connmark_net_ops);
}

static void __exit connmark_cleanup_module(void)
{
	tcf_unregister_action(&act_connmark_ops, &connmark_net_ops);
}

module_init(connmark_init_module);
module_exit(connmark_cleanup_module);
MODULE_AUTHOR("Felix Fietkau <nbd@openwrt.org>");
MODULE_DESCRIPTION("Connection tracking mark restoring");
MODULE_LICENSE("GPL");
Commit	Line	Data
22a5dc0e FF	1	/*
	2	* net/sched/act_connmark.c netfilter connmark retriever action
	3	* skb mark is over-written
	4	*
	5	* Copyright (c) 2011 Felix Fietkau <nbd@openwrt.org>
	6	*
	7	* This program is free software; you can redistribute it and/or modify
	8	* it under the terms of the GNU General Public License as published by
	9	* the Free Software Foundation; either version 2 of the License, or
	10	* (at your option) any later version.
	11	*/
	12
	13	#include <linux/module.h>
	14	#include <linux/init.h>
	15	#include <linux/kernel.h>
	16	#include <linux/skbuff.h>
	17	#include <linux/rtnetlink.h>
	18	#include <linux/pkt_cls.h>
	19	#include <linux/ip.h>
	20	#include <linux/ipv6.h>
	21	#include <net/netlink.h>
	22	#include <net/pkt_sched.h>
	23	#include <net/act_api.h>
c53075ea	24	#include <net/pkt_cls.h>
22a5dc0e FF	25	#include <uapi/linux/tc_act/tc_connmark.h>
	26	#include <net/tc_act/tc_connmark.h>
	27
	28	#include <net/netfilter/nf_conntrack.h>
	29	#include <net/netfilter/nf_conntrack_core.h>
	30	#include <net/netfilter/nf_conntrack_zones.h>
	31
c7d03a00	32	static unsigned int connmark_net_id;
a85a970a	33	static struct tc_action_ops act_connmark_ops;
ddf97ccd	34
962ad1f9 JHS	35	static int tcf_connmark_act(struct sk_buff skb, const struct tc_action a,
962ad1f9 JHS	36	struct tcf_result *res)
22a5dc0e FF	37	{
	38	const struct nf_conntrack_tuple_hash *thash;
	39	struct nf_conntrack_tuple tuple;
	40	enum ip_conntrack_info ctinfo;
a85a970a	41	struct tcf_connmark_info *ca = to_connmark(a);
308ac914	42	struct nf_conntrack_zone zone;
22a5dc0e FF	43	struct nf_conn *c;
	44	int proto;
	45
	46	spin_lock(&ca->tcf_lock);
9c4a4e48	47	tcf_lastuse_update(&ca->tcf_tm);
22a5dc0e FF	48	bstats_update(&ca->tcf_bstats, skb);
	49
	50	if (skb->protocol == htons(ETH_P_IP)) {
	51	if (skb->len < sizeof(struct iphdr))
	52	goto out;
	53
	54	proto = NFPROTO_IPV4;
	55	} else if (skb->protocol == htons(ETH_P_IPV6)) {
	56	if (skb->len < sizeof(struct ipv6hdr))
	57	goto out;
	58
	59	proto = NFPROTO_IPV6;
	60	} else {
	61	goto out;
	62	}
	63
	64	c = nf_ct_get(skb, &ctinfo);
	65	if (c) {
	66	skb->mark = c->mark;
	67	/* using overlimits stats to count how many packets marked */
	68	ca->tcf_qstats.overlimits++;
22a5dc0e FF	69	goto out;
	70	}
	71
	72	if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb),
a31f1adc	73	proto, ca->net, &tuple))
22a5dc0e FF	74	goto out;
22a5dc0e FF	75
308ac914	76	zone.id = ca->zone;
deedb590	77	zone.dir = NF_CT_DEFAULT_ZONE_DIR;
308ac914	78
a4ffe319	79	thash = nf_conntrack_find_get(ca->net, &zone, &tuple);
22a5dc0e FF	80	if (!thash)
	81	goto out;
	82
	83	c = nf_ct_tuplehash_to_ctrack(thash);
	84	/* using overlimits stats to count how many packets marked */
	85	ca->tcf_qstats.overlimits++;
	86	skb->mark = c->mark;
	87	nf_ct_put(c);
	88
	89	out:
22a5dc0e FF	90	spin_unlock(&ca->tcf_lock);
	91	return ca->tcf_action;
	92	}
	93
	94	static const struct nla_policy connmark_policy[TCA_CONNMARK_MAX + 1] = {
	95	[TCA_CONNMARK_PARMS] = { .len = sizeof(struct tc_connmark) },
	96	};
	97
	98	static int tcf_connmark_init(struct net net, struct nlattr nla,
a85a970a	99	struct nlattr est, struct tc_action *a,
789871bb	100	int ovr, int bind, bool rtnl_held,
85d0966f	101	struct tcf_proto *tp,
589dad6d	102	struct netlink_ext_ack *extack)
22a5dc0e	103	{
ddf97ccd	104	struct tc_action_net *tn = net_generic(net, connmark_net_id);
22a5dc0e	105	struct nlattr *tb[TCA_CONNMARK_MAX + 1];
c53075ea	106	struct tcf_chain *goto_ch = NULL;
22a5dc0e FF	107	struct tcf_connmark_info *ci;
22a5dc0e FF	108	struct tc_connmark *parm;
c53075ea	109	int ret = 0, err;
22a5dc0e FF	110
	111	if (!nla)
	112	return -EINVAL;
	113
fceb6435 JB	114	ret = nla_parse_nested(tb, TCA_CONNMARK_MAX, nla, connmark_policy,
fceb6435 JB	115	NULL);
22a5dc0e FF	116	if (ret < 0)
	117	return ret;
	118
52491c76 EN	119	if (!tb[TCA_CONNMARK_PARMS])
	120	return -EINVAL;
	121
22a5dc0e FF	122	parm = nla_data(tb[TCA_CONNMARK_PARMS]);
22a5dc0e FF	123
0190c1d4 VB	124	ret = tcf_idr_check_alloc(tn, &parm->index, a, bind);
0190c1d4 VB	125	if (!ret) {
65a206c0 CM	126	ret = tcf_idr_create(tn, parm->index, est, a,
65a206c0 CM	127	&act_connmark_ops, bind, false);
0190c1d4 VB	128	if (ret) {
0190c1d4 VB	129	tcf_idr_cleanup(tn, parm->index);
22a5dc0e	130	return ret;
0190c1d4	131	}
22a5dc0e	132
a85a970a	133	ci = to_connmark(*a);
c53075ea DC	134	err = tcf_action_check_ctrlact(parm->action, tp, &goto_ch,
	135	extack);
	136	if (err < 0)
	137	goto release_idr;
	138	tcf_action_set_ctrlact(*a, parm->action, goto_ch);
a4ffe319	139	ci->net = net;
22a5dc0e FF	140	ci->zone = parm->zone;
22a5dc0e FF	141
65a206c0	142	tcf_idr_insert(tn, *a);
22a5dc0e	143	ret = ACT_P_CREATED;
0190c1d4	144	} else if (ret > 0) {
a85a970a	145	ci = to_connmark(*a);
22a5dc0e FF	146	if (bind)
22a5dc0e FF	147	return 0;
4e8ddd7f VB	148	if (!ovr) {
4e8ddd7f VB	149	tcf_idr_release(*a, bind);
22a5dc0e	150	return -EEXIST;
4e8ddd7f	151	}
c53075ea DC	152	err = tcf_action_check_ctrlact(parm->action, tp, &goto_ch,
	153	extack);
	154	if (err < 0)
	155	goto release_idr;
22a5dc0e	156	/* replacing action and zone */
506a03aa	157	spin_lock_bh(&ci->tcf_lock);
c53075ea	158	goto_ch = tcf_action_set_ctrlact(*a, parm->action, goto_ch);
22a5dc0e	159	ci->zone = parm->zone;
506a03aa	160	spin_unlock_bh(&ci->tcf_lock);
c53075ea DC	161	if (goto_ch)
c53075ea DC	162	tcf_chain_put_by_act(goto_ch);
0190c1d4	163	ret = 0;
22a5dc0e FF	164	}
	165
	166	return ret;
c53075ea DC	167	release_idr:
	168	tcf_idr_release(*a, bind);
	169	return err;
22a5dc0e FF	170	}
	171
	172	static inline int tcf_connmark_dump(struct sk_buff skb, struct tc_action a,
	173	int bind, int ref)
	174	{
	175	unsigned char *b = skb_tail_pointer(skb);
a85a970a	176	struct tcf_connmark_info *ci = to_connmark(a);
22a5dc0e FF	177	struct tc_connmark opt = {
22a5dc0e FF	178	.index = ci->tcf_index,
036bb443 VB	179	.refcnt = refcount_read(&ci->tcf_refcnt) - ref,
036bb443 VB	180	.bindcnt = atomic_read(&ci->tcf_bindcnt) - bind,
22a5dc0e FF	181	};
	182	struct tcf_t t;
	183
506a03aa CW	184	spin_lock_bh(&ci->tcf_lock);
	185	opt.action = ci->tcf_action;
	186	opt.zone = ci->zone;
22a5dc0e FF	187	if (nla_put(skb, TCA_CONNMARK_PARMS, sizeof(opt), &opt))
	188	goto nla_put_failure;
	189
48d8ee16	190	tcf_tm_dump(&t, &ci->tcf_tm);
9854518e ND	191	if (nla_put_64bit(skb, TCA_CONNMARK_TM, sizeof(t), &t,
9854518e ND	192	TCA_CONNMARK_PAD))
22a5dc0e	193	goto nla_put_failure;
506a03aa	194	spin_unlock_bh(&ci->tcf_lock);
22a5dc0e FF	195
22a5dc0e FF	196	return skb->len;
506a03aa	197
22a5dc0e	198	nla_put_failure:
506a03aa	199	spin_unlock_bh(&ci->tcf_lock);
22a5dc0e FF	200	nlmsg_trim(skb, b);
	201	return -1;
	202	}
	203
ddf97ccd WC	204	static int tcf_connmark_walker(struct net net, struct sk_buff skb,
ddf97ccd WC	205	struct netlink_callback *cb, int type,
41780105 AA	206	const struct tc_action_ops *ops,
41780105 AA	207	struct netlink_ext_ack *extack)
ddf97ccd WC	208	{
	209	struct tc_action_net *tn = net_generic(net, connmark_net_id);
	210
b3620145	211	return tcf_generic_walker(tn, skb, cb, type, ops, extack);
ddf97ccd WC	212	}
ddf97ccd WC	213
f061b48c	214	static int tcf_connmark_search(struct net net, struct tc_action *a, u32 index)
ddf97ccd WC	215	{
	216	struct tc_action_net *tn = net_generic(net, connmark_net_id);
	217
65a206c0	218	return tcf_idr_search(tn, a, index);
ddf97ccd WC	219	}
ddf97ccd WC	220
22a5dc0e FF	221	static struct tc_action_ops act_connmark_ops = {
22a5dc0e FF	222	.kind = "connmark",
eddd2cf1	223	.id = TCA_ID_CONNMARK,
22a5dc0e	224	.owner = THIS_MODULE,
962ad1f9	225	.act = tcf_connmark_act,
22a5dc0e FF	226	.dump = tcf_connmark_dump,
22a5dc0e FF	227	.init = tcf_connmark_init,
ddf97ccd WC	228	.walk = tcf_connmark_walker,
ddf97ccd WC	229	.lookup = tcf_connmark_search,
a85a970a	230	.size = sizeof(struct tcf_connmark_info),
ddf97ccd WC	231	};
	232
	233	static __net_init int connmark_init_net(struct net *net)
	234	{
	235	struct tc_action_net *tn = net_generic(net, connmark_net_id);
	236
c7e460ce	237	return tc_action_net_init(tn, &act_connmark_ops);
ddf97ccd WC	238	}
ddf97ccd WC	239
039af9c6	240	static void __net_exit connmark_exit_net(struct list_head *net_list)
ddf97ccd	241	{
039af9c6	242	tc_action_net_exit(net_list, connmark_net_id);
ddf97ccd WC	243	}
	244
	245	static struct pernet_operations connmark_net_ops = {
	246	.init = connmark_init_net,
039af9c6	247	.exit_batch = connmark_exit_net,
ddf97ccd WC	248	.id = &connmark_net_id,
ddf97ccd WC	249	.size = sizeof(struct tc_action_net),
22a5dc0e FF	250	};
	251
	252	static int __init connmark_init_module(void)
	253	{
ddf97ccd	254	return tcf_register_action(&act_connmark_ops, &connmark_net_ops);
22a5dc0e FF	255	}
	256
	257	static void __exit connmark_cleanup_module(void)
	258	{
ddf97ccd	259	tcf_unregister_action(&act_connmark_ops, &connmark_net_ops);
22a5dc0e FF	260	}
	261
	262	module_init(connmark_init_module);
	263	module_exit(connmark_cleanup_module);
	264	MODULE_AUTHOR("Felix Fietkau <nbd@openwrt.org>");
	265	MODULE_DESCRIPTION("Connection tracking mark restoring");
	266	MODULE_LICENSE("GPL");