[deliverable/linux.git] / net / openvswitch / conntrack.c

/*
 * Copyright (c) 2015 Nicira, Inc.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of version 2 of the GNU General Public
 * License as published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful, but
 * WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
 * General Public License for more details.
 */

#include <linux/module.h>
#include <linux/openvswitch.h>
#include <net/ip.h>
#include <net/netfilter/nf_conntrack_core.h>
#include <net/netfilter/nf_conntrack_zones.h>
#include <net/netfilter/ipv6/nf_defrag_ipv6.h>

#include "datapath.h"
#include "conntrack.h"
#include "flow.h"
#include "flow_netlink.h"

struct ovs_ct_len_tbl {
	size_t maxlen;
	size_t minlen;
};

/* Conntrack action context for execution. */
struct ovs_conntrack_info {
	struct nf_conntrack_zone zone;
	struct nf_conn *ct;
	u32 flags;
	u16 family;
};

static u16 key_to_nfproto(const struct sw_flow_key *key)
{
	switch (ntohs(key->eth.type)) {
	case ETH_P_IP:
		return NFPROTO_IPV4;
	case ETH_P_IPV6:
		return NFPROTO_IPV6;
	default:
		return NFPROTO_UNSPEC;
	}
}

/* Map SKB connection state into the values used by flow definition. */
static u8 ovs_ct_get_state(enum ip_conntrack_info ctinfo)
{
	u8 ct_state = OVS_CS_F_TRACKED;

	switch (ctinfo) {
	case IP_CT_ESTABLISHED_REPLY:
	case IP_CT_RELATED_REPLY:
	case IP_CT_NEW_REPLY:
		ct_state |= OVS_CS_F_REPLY_DIR;
		break;
	default:
		break;
	}

	switch (ctinfo) {
	case IP_CT_ESTABLISHED:
	case IP_CT_ESTABLISHED_REPLY:
		ct_state |= OVS_CS_F_ESTABLISHED;
		break;
	case IP_CT_RELATED:
	case IP_CT_RELATED_REPLY:
		ct_state |= OVS_CS_F_RELATED;
		break;
	case IP_CT_NEW:
	case IP_CT_NEW_REPLY:
		ct_state |= OVS_CS_F_NEW;
		break;
	default:
		break;
	}

	return ct_state;
}

static void __ovs_ct_update_key(struct sw_flow_key *key, u8 state,
				const struct nf_conntrack_zone *zone)
{
	key->ct.state = state;
	key->ct.zone = zone->id;
}

/* Update 'key' based on skb->nfct. If 'post_ct' is true, then OVS has
 * previously sent the packet to conntrack via the ct action.
 */
static void ovs_ct_update_key(const struct sk_buff *skb,
			      struct sw_flow_key *key, bool post_ct)
{
	const struct nf_conntrack_zone *zone = &nf_ct_zone_dflt;
	enum ip_conntrack_info ctinfo;
	struct nf_conn *ct;
	u8 state = 0;

	ct = nf_ct_get(skb, &ctinfo);
	if (ct) {
		state = ovs_ct_get_state(ctinfo);
		if (ct->master)
			state |= OVS_CS_F_RELATED;
		zone = nf_ct_zone(ct);
	} else if (post_ct) {
		state = OVS_CS_F_TRACKED | OVS_CS_F_INVALID;
	}
	__ovs_ct_update_key(key, state, zone);
}

void ovs_ct_fill_key(const struct sk_buff *skb, struct sw_flow_key *key)
{
	ovs_ct_update_key(skb, key, false);
}

int ovs_ct_put_key(const struct sw_flow_key *key, struct sk_buff *skb)
{
	if (nla_put_u8(skb, OVS_KEY_ATTR_CT_STATE, key->ct.state))
		return -EMSGSIZE;

	if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) &&
	    nla_put_u16(skb, OVS_KEY_ATTR_CT_ZONE, key->ct.zone))
		return -EMSGSIZE;

	return 0;
}

static int handle_fragments(struct net *net, struct sw_flow_key *key,
			    u16 zone, struct sk_buff *skb)
{
	struct ovs_skb_cb ovs_cb = *OVS_CB(skb);

	if (key->eth.type == htons(ETH_P_IP)) {
		enum ip_defrag_users user = IP_DEFRAG_CONNTRACK_IN + zone;
		int err;

		memset(IPCB(skb), 0, sizeof(struct inet_skb_parm));
		err = ip_defrag(skb, user);
		if (err)
			return err;

		ovs_cb.mru = IPCB(skb)->frag_max_size;
	} else if (key->eth.type == htons(ETH_P_IPV6)) {
#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6)
		enum ip6_defrag_users user = IP6_DEFRAG_CONNTRACK_IN + zone;
		struct sk_buff *reasm;

		memset(IP6CB(skb), 0, sizeof(struct inet6_skb_parm));
		reasm = nf_ct_frag6_gather(skb, user);
		if (!reasm)
			return -EINPROGRESS;

		if (skb == reasm)
			return -EINVAL;

		key->ip.proto = ipv6_hdr(reasm)->nexthdr;
		skb_morph(skb, reasm);
		consume_skb(reasm);
		ovs_cb.mru = IP6CB(skb)->frag_max_size;
#else
		return -EPFNOSUPPORT;
#endif
	} else {
		return -EPFNOSUPPORT;
	}

	key->ip.frag = OVS_FRAG_TYPE_NONE;
	skb_clear_hash(skb);
	skb->ignore_df = 1;
	*OVS_CB(skb) = ovs_cb;

	return 0;
}

static struct nf_conntrack_expect *
ovs_ct_expect_find(struct net *net, const struct nf_conntrack_zone *zone,
		   u16 proto, const struct sk_buff *skb)
{
	struct nf_conntrack_tuple tuple;

	if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), proto, &tuple))
		return NULL;
	return __nf_ct_expect_find(net, zone, &tuple);
}

/* Determine whether skb->nfct is equal to the result of conntrack lookup. */
static bool skb_nfct_cached(const struct net *net, const struct sk_buff *skb,
			    const struct ovs_conntrack_info *info)
{
	enum ip_conntrack_info ctinfo;
	struct nf_conn *ct;

	ct = nf_ct_get(skb, &ctinfo);
	if (!ct)
		return false;
	if (!net_eq(net, read_pnet(&ct->ct_net)))
		return false;
	if (!nf_ct_zone_equal_any(info->ct, nf_ct_zone(ct)))
		return false;

	return true;
}

static int __ovs_ct_lookup(struct net *net, const struct sw_flow_key *key,
			   const struct ovs_conntrack_info *info,
			   struct sk_buff *skb)
{
	/* If we are recirculating packets to match on conntrack fields and
	 * committing with a separate conntrack action,  then we don't need to
	 * actually run the packet through conntrack twice unless it's for a
	 * different zone.
	 */
	if (!skb_nfct_cached(net, skb, info)) {
		struct nf_conn *tmpl = info->ct;

		/* Associate skb with specified zone. */
		if (tmpl) {
			if (skb->nfct)
				nf_conntrack_put(skb->nfct);
			nf_conntrack_get(&tmpl->ct_general);
			skb->nfct = &tmpl->ct_general;
			skb->nfctinfo = IP_CT_NEW;
		}

		if (nf_conntrack_in(net, info->family, NF_INET_PRE_ROUTING,
				    skb) != NF_ACCEPT)
			return -ENOENT;
	}

	return 0;
}

/* Lookup connection and read fields into key. */
static int ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
			 const struct ovs_conntrack_info *info,
			 struct sk_buff *skb)
{
	struct nf_conntrack_expect *exp;

	exp = ovs_ct_expect_find(net, &info->zone, info->family, skb);
	if (exp) {
		u8 state;

		state = OVS_CS_F_TRACKED | OVS_CS_F_NEW | OVS_CS_F_RELATED;
		__ovs_ct_update_key(key, state, &info->zone);
	} else {
		int err;

		err = __ovs_ct_lookup(net, key, info, skb);
		if (err)
			return err;

		ovs_ct_update_key(skb, key, true);
	}

	return 0;
}

/* Lookup connection and confirm if unconfirmed. */
static int ovs_ct_commit(struct net *net, struct sw_flow_key *key,
			 const struct ovs_conntrack_info *info,
			 struct sk_buff *skb)
{
	u8 state;
	int err;

	state = key->ct.state;
	if (key->ct.zone == info->zone.id &&
	    ((state & OVS_CS_F_TRACKED) && !(state & OVS_CS_F_NEW))) {
		/* Previous lookup has shown that this connection is already
		 * tracked and committed. Skip committing.
		 */
		return 0;
	}

	err = __ovs_ct_lookup(net, key, info, skb);
	if (err)
		return err;
	if (nf_conntrack_confirm(skb) != NF_ACCEPT)
		return -EINVAL;

	ovs_ct_update_key(skb, key, true);

	return 0;
}

int ovs_ct_execute(struct net *net, struct sk_buff *skb,
		   struct sw_flow_key *key,
		   const struct ovs_conntrack_info *info)
{
	int nh_ofs;
	int err;

	/* The conntrack module expects to be working at L3. */
	nh_ofs = skb_network_offset(skb);
	skb_pull(skb, nh_ofs);

	if (key->ip.frag != OVS_FRAG_TYPE_NONE) {
		err = handle_fragments(net, key, info->zone.id, skb);
		if (err)
			return err;
	}

	if (info->flags & OVS_CT_F_COMMIT)
		err = ovs_ct_commit(net, key, info, skb);
	else
		err = ovs_ct_lookup(net, key, info, skb);

	skb_push(skb, nh_ofs);
	return err;
}

static const struct ovs_ct_len_tbl ovs_ct_attr_lens[OVS_CT_ATTR_MAX + 1] = {
	[OVS_CT_ATTR_FLAGS]	= { .minlen = sizeof(u32),
				    .maxlen = sizeof(u32) },
	[OVS_CT_ATTR_ZONE]	= { .minlen = sizeof(u16),
				    .maxlen = sizeof(u16) },
};

static int parse_ct(const struct nlattr *attr, struct ovs_conntrack_info *info,
		    bool log)
{
	struct nlattr *a;
	int rem;

	nla_for_each_nested(a, attr, rem) {
		int type = nla_type(a);
		int maxlen = ovs_ct_attr_lens[type].maxlen;
		int minlen = ovs_ct_attr_lens[type].minlen;

		if (type > OVS_CT_ATTR_MAX) {
			OVS_NLERR(log,
				  "Unknown conntrack attr (type=%d, max=%d)",
				  type, OVS_CT_ATTR_MAX);
			return -EINVAL;
		}
		if (nla_len(a) < minlen || nla_len(a) > maxlen) {
			OVS_NLERR(log,
				  "Conntrack attr type has unexpected length (type=%d, length=%d, expected=%d)",
				  type, nla_len(a), maxlen);
			return -EINVAL;
		}

		switch (type) {
		case OVS_CT_ATTR_FLAGS:
			info->flags = nla_get_u32(a);
			break;
#ifdef CONFIG_NF_CONNTRACK_ZONES
		case OVS_CT_ATTR_ZONE:
			info->zone.id = nla_get_u16(a);
			break;
#endif
		default:
			OVS_NLERR(log, "Unknown conntrack attr (%d)",
				  type);
			return -EINVAL;
		}
	}

	if (rem > 0) {
		OVS_NLERR(log, "Conntrack attr has %d unknown bytes", rem);
		return -EINVAL;
	}

	return 0;
}

bool ovs_ct_verify(enum ovs_key_attr attr)
{
	if (attr == OVS_KEY_ATTR_CT_STATE)
		return true;
	if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) &&
	    attr == OVS_KEY_ATTR_CT_ZONE)
		return true;

	return false;
}

int ovs_ct_copy_action(struct net *net, const struct nlattr *attr,
		       const struct sw_flow_key *key,
		       struct sw_flow_actions **sfa,  bool log)
{
	struct ovs_conntrack_info ct_info;
	u16 family;
	int err;

	family = key_to_nfproto(key);
	if (family == NFPROTO_UNSPEC) {
		OVS_NLERR(log, "ct family unspecified");
		return -EINVAL;
	}

	memset(&ct_info, 0, sizeof(ct_info));
	ct_info.family = family;

	nf_ct_zone_init(&ct_info.zone, NF_CT_DEFAULT_ZONE_ID,
			NF_CT_DEFAULT_ZONE_DIR, 0);

	err = parse_ct(attr, &ct_info, log);
	if (err)
		return err;

	/* Set up template for tracking connections in specific zones. */
	ct_info.ct = nf_ct_tmpl_alloc(net, &ct_info.zone, GFP_KERNEL);
	if (!ct_info.ct) {
		OVS_NLERR(log, "Failed to allocate conntrack template");
		return -ENOMEM;
	}

	err = ovs_nla_add_action(sfa, OVS_ACTION_ATTR_CT, &ct_info,
				 sizeof(ct_info), log);
	if (err)
		goto err_free_ct;

	__set_bit(IPS_CONFIRMED_BIT, &ct_info.ct->status);
	nf_conntrack_get(&ct_info.ct->ct_general);
	return 0;
err_free_ct:
	nf_conntrack_free(ct_info.ct);
	return err;
}

int ovs_ct_action_to_attr(const struct ovs_conntrack_info *ct_info,
			  struct sk_buff *skb)
{
	struct nlattr *start;

	start = nla_nest_start(skb, OVS_ACTION_ATTR_CT);
	if (!start)
		return -EMSGSIZE;

	if (nla_put_u32(skb, OVS_CT_ATTR_FLAGS, ct_info->flags))
		return -EMSGSIZE;
	if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) &&
	    nla_put_u16(skb, OVS_CT_ATTR_ZONE, ct_info->zone.id))
		return -EMSGSIZE;

	nla_nest_end(skb, start);

	return 0;
}

void ovs_ct_free_action(const struct nlattr *a)
{
	struct ovs_conntrack_info *ct_info = nla_data(a);

	if (ct_info->ct)
		nf_ct_put(ct_info->ct);
}
Commit	Line	Data
7f8a436e JS	1	/*
	2	* Copyright (c) 2015 Nicira, Inc.
	3	*
	4	* This program is free software; you can redistribute it and/or
	5	* modify it under the terms of version 2 of the GNU General Public
	6	* License as published by the Free Software Foundation.
	7	*
	8	* This program is distributed in the hope that it will be useful, but
	9	* WITHOUT ANY WARRANTY; without even the implied warranty of
	10	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
	11	* General Public License for more details.
	12	*/
	13
	14	#include <linux/module.h>
	15	#include <linux/openvswitch.h>
	16	#include <net/ip.h>
	17	#include <net/netfilter/nf_conntrack_core.h>
	18	#include <net/netfilter/nf_conntrack_zones.h>
	19	#include <net/netfilter/ipv6/nf_defrag_ipv6.h>
	20
	21	#include "datapath.h"
	22	#include "conntrack.h"
	23	#include "flow.h"
	24	#include "flow_netlink.h"
	25
	26	struct ovs_ct_len_tbl {
	27	size_t maxlen;
	28	size_t minlen;
	29	};
	30
	31	/* Conntrack action context for execution. */
	32	struct ovs_conntrack_info {
	33	struct nf_conntrack_zone zone;
	34	struct nf_conn *ct;
	35	u32 flags;
	36	u16 family;
	37	};
	38
	39	static u16 key_to_nfproto(const struct sw_flow_key *key)
	40	{
	41	switch (ntohs(key->eth.type)) {
	42	case ETH_P_IP:
	43	return NFPROTO_IPV4;
	44	case ETH_P_IPV6:
	45	return NFPROTO_IPV6;
	46	default:
	47	return NFPROTO_UNSPEC;
	48	}
	49	}
	50
	51	/* Map SKB connection state into the values used by flow definition. */
	52	static u8 ovs_ct_get_state(enum ip_conntrack_info ctinfo)
	53	{
	54	u8 ct_state = OVS_CS_F_TRACKED;
	55
	56	switch (ctinfo) {
	57	case IP_CT_ESTABLISHED_REPLY:
	58	case IP_CT_RELATED_REPLY:
	59	case IP_CT_NEW_REPLY:
	60	ct_state \|= OVS_CS_F_REPLY_DIR;
	61	break;
	62	default:
	63	break;
	64	}
65
66	switch (ctinfo) {
67	case IP_CT_ESTABLISHED:
68	case IP_CT_ESTABLISHED_REPLY:
69	ct_state \|= OVS_CS_F_ESTABLISHED;
70	break;
71	case IP_CT_RELATED:
72	case IP_CT_RELATED_REPLY:
73	ct_state \|= OVS_CS_F_RELATED;
74	break;
75	case IP_CT_NEW:
76	case IP_CT_NEW_REPLY:
77	ct_state \|= OVS_CS_F_NEW;
78	break;
79	default:
80	break;
81	}
82
83	return ct_state;
84	}
85
86	static void __ovs_ct_update_key(struct sw_flow_key *key, u8 state,
87	const struct nf_conntrack_zone *zone)
88	{
89	key->ct.state = state;
90	key->ct.zone = zone->id;
91	}
92
93	/* Update 'key' based on skb->nfct. If 'post_ct' is true, then OVS has
94	* previously sent the packet to conntrack via the ct action.
95	*/
96	static void ovs_ct_update_key(const struct sk_buff *skb,
97	struct sw_flow_key *key, bool post_ct)
98	{
99	const struct nf_conntrack_zone *zone = &nf_ct_zone_dflt;
100	enum ip_conntrack_info ctinfo;
101	struct nf_conn *ct;
102	u8 state = 0;
103
104	ct = nf_ct_get(skb, &ctinfo);
105	if (ct) {
106	state = ovs_ct_get_state(ctinfo);
107	if (ct->master)
108	state \|= OVS_CS_F_RELATED;
109	zone = nf_ct_zone(ct);
110	} else if (post_ct) {
111	state = OVS_CS_F_TRACKED \| OVS_CS_F_INVALID;
112	}
113	__ovs_ct_update_key(key, state, zone);
114	}
115
116	void ovs_ct_fill_key(const struct sk_buff skb, struct sw_flow_key key)
117	{
118	ovs_ct_update_key(skb, key, false);
119	}
120
121	int ovs_ct_put_key(const struct sw_flow_key key, struct sk_buff skb)
122	{
123	if (nla_put_u8(skb, OVS_KEY_ATTR_CT_STATE, key->ct.state))
124	return -EMSGSIZE;
125
126	if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) &&
127	nla_put_u16(skb, OVS_KEY_ATTR_CT_ZONE, key->ct.zone))
128	return -EMSGSIZE;
129
130	return 0;
131	}
132
133	static int handle_fragments(struct net net, struct sw_flow_key key,
134	u16 zone, struct sk_buff *skb)
135	{
136	struct ovs_skb_cb ovs_cb = *OVS_CB(skb);
137
138	if (key->eth.type == htons(ETH_P_IP)) {
139	enum ip_defrag_users user = IP_DEFRAG_CONNTRACK_IN + zone;
140	int err;
141
142	memset(IPCB(skb), 0, sizeof(struct inet_skb_parm));
143	err = ip_defrag(skb, user);
144	if (err)
145	return err;
146
147	ovs_cb.mru = IPCB(skb)->frag_max_size;
148	} else if (key->eth.type == htons(ETH_P_IPV6)) {
149	#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6)
150	enum ip6_defrag_users user = IP6_DEFRAG_CONNTRACK_IN + zone;
151	struct sk_buff *reasm;
152
153	memset(IP6CB(skb), 0, sizeof(struct inet6_skb_parm));
154	reasm = nf_ct_frag6_gather(skb, user);
155	if (!reasm)
156	return -EINPROGRESS;
157
158	if (skb == reasm)
159	return -EINVAL;
160
161	key->ip.proto = ipv6_hdr(reasm)->nexthdr;
162	skb_morph(skb, reasm);
163	consume_skb(reasm);
164	ovs_cb.mru = IP6CB(skb)->frag_max_size;
165	#else
166	return -EPFNOSUPPORT;
167	#endif
168	} else {
169	return -EPFNOSUPPORT;
170	}
171
172	key->ip.frag = OVS_FRAG_TYPE_NONE;
173	skb_clear_hash(skb);
174	skb->ignore_df = 1;
175	*OVS_CB(skb) = ovs_cb;
176
177	return 0;
178	}
179
180	static struct nf_conntrack_expect *
181	ovs_ct_expect_find(struct net net, const struct nf_conntrack_zone zone,
182	u16 proto, const struct sk_buff *skb)
183	{
184	struct nf_conntrack_tuple tuple;
185
186	if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), proto, &tuple))
187	return NULL;
188	return __nf_ct_expect_find(net, zone, &tuple);
189	}
190
191	/* Determine whether skb->nfct is equal to the result of conntrack lookup. */
192	static bool skb_nfct_cached(const struct net net, const struct sk_buff skb,
193	const struct ovs_conntrack_info *info)
194	{
195	enum ip_conntrack_info ctinfo;
196	struct nf_conn *ct;
197
198	ct = nf_ct_get(skb, &ctinfo);
199	if (!ct)
200	return false;
201	if (!net_eq(net, read_pnet(&ct->ct_net)))
202	return false;
203	if (!nf_ct_zone_equal_any(info->ct, nf_ct_zone(ct)))
204	return false;
205
206	return true;
207	}
208
209	static int __ovs_ct_lookup(struct net net, const struct sw_flow_key key,
210	const struct ovs_conntrack_info *info,
211	struct sk_buff *skb)
212	{
213	/* If we are recirculating packets to match on conntrack fields and
214	* committing with a separate conntrack action, then we don't need to
215	* actually run the packet through conntrack twice unless it's for a
216	* different zone.
217	*/
218	if (!skb_nfct_cached(net, skb, info)) {
219	struct nf_conn *tmpl = info->ct;
220
221	/* Associate skb with specified zone. */
222	if (tmpl) {
223	if (skb->nfct)
224	nf_conntrack_put(skb->nfct);
225	nf_conntrack_get(&tmpl->ct_general);
226	skb->nfct = &tmpl->ct_general;
227	skb->nfctinfo = IP_CT_NEW;
228	}
229
230	if (nf_conntrack_in(net, info->family, NF_INET_PRE_ROUTING,
231	skb) != NF_ACCEPT)
232	return -ENOENT;
233	}
234
235	return 0;
236	}
237
238	/* Lookup connection and read fields into key. */
239	static int ovs_ct_lookup(struct net net, struct sw_flow_key key,
240	const struct ovs_conntrack_info *info,
241	struct sk_buff *skb)
242	{
243	struct nf_conntrack_expect *exp;
244
245	exp = ovs_ct_expect_find(net, &info->zone, info->family, skb);
246	if (exp) {
247	u8 state;
248
249	state = OVS_CS_F_TRACKED \| OVS_CS_F_NEW \| OVS_CS_F_RELATED;
250	__ovs_ct_update_key(key, state, &info->zone);
251	} else {
252	int err;
253
254	err = __ovs_ct_lookup(net, key, info, skb);
255	if (err)
256	return err;
257
258	ovs_ct_update_key(skb, key, true);
259	}
260
261	return 0;
262	}
263
264	/* Lookup connection and confirm if unconfirmed. */
265	static int ovs_ct_commit(struct net net, struct sw_flow_key key,
266	const struct ovs_conntrack_info *info,
267	struct sk_buff *skb)
268	{
269	u8 state;
270	int err;
271
272	state = key->ct.state;
273	if (key->ct.zone == info->zone.id &&
274	((state & OVS_CS_F_TRACKED) && !(state & OVS_CS_F_NEW))) {
275	/* Previous lookup has shown that this connection is already
276	* tracked and committed. Skip committing.
277	*/
278	return 0;
279	}
280
281	err = __ovs_ct_lookup(net, key, info, skb);
282	if (err)
283	return err;
284	if (nf_conntrack_confirm(skb) != NF_ACCEPT)
285	return -EINVAL;
286
287	ovs_ct_update_key(skb, key, true);
288
289	return 0;
290	}
291
292	int ovs_ct_execute(struct net net, struct sk_buff skb,
293	struct sw_flow_key *key,
294	const struct ovs_conntrack_info *info)
295	{
296	int nh_ofs;
297	int err;
298
299	/* The conntrack module expects to be working at L3. */
300	nh_ofs = skb_network_offset(skb);
301	skb_pull(skb, nh_ofs);
302
303	if (key->ip.frag != OVS_FRAG_TYPE_NONE) {
304	err = handle_fragments(net, key, info->zone.id, skb);
305	if (err)
306	return err;
307	}
308
309	if (info->flags & OVS_CT_F_COMMIT)
310	err = ovs_ct_commit(net, key, info, skb);
311	else
312	err = ovs_ct_lookup(net, key, info, skb);
313
314	skb_push(skb, nh_ofs);
315	return err;
316	}
317
318	static const struct ovs_ct_len_tbl ovs_ct_attr_lens[OVS_CT_ATTR_MAX + 1] = {
319	[OVS_CT_ATTR_FLAGS] = { .minlen = sizeof(u32),
320	.maxlen = sizeof(u32) },
321	[OVS_CT_ATTR_ZONE] = { .minlen = sizeof(u16),
322	.maxlen = sizeof(u16) },
323	};
324
325	static int parse_ct(const struct nlattr attr, struct ovs_conntrack_info info,
326	bool log)
327	{
328	struct nlattr *a;
329	int rem;
330
331	nla_for_each_nested(a, attr, rem) {
332	int type = nla_type(a);
333	int maxlen = ovs_ct_attr_lens[type].maxlen;
334	int minlen = ovs_ct_attr_lens[type].minlen;
335
336	if (type > OVS_CT_ATTR_MAX) {
337	OVS_NLERR(log,
338	"Unknown conntrack attr (type=%d, max=%d)",
339	type, OVS_CT_ATTR_MAX);
340	return -EINVAL;
341	}
342	if (nla_len(a) < minlen \|\| nla_len(a) > maxlen) {
343	OVS_NLERR(log,
344	"Conntrack attr type has unexpected length (type=%d, length=%d, expected=%d)",
345	type, nla_len(a), maxlen);
346	return -EINVAL;
347	}
348
349	switch (type) {
350	case OVS_CT_ATTR_FLAGS:
351	info->flags = nla_get_u32(a);
352	break;
353	#ifdef CONFIG_NF_CONNTRACK_ZONES
354	case OVS_CT_ATTR_ZONE:
355	info->zone.id = nla_get_u16(a);
356	break;
357	#endif
358	default:
359	OVS_NLERR(log, "Unknown conntrack attr (%d)",
360	type);
361	return -EINVAL;
362	}
363	}
364
365	if (rem > 0) {
366	OVS_NLERR(log, "Conntrack attr has %d unknown bytes", rem);
367	return -EINVAL;
368	}
369
370	return 0;
371	}
372
373	bool ovs_ct_verify(enum ovs_key_attr attr)
374	{
375	if (attr == OVS_KEY_ATTR_CT_STATE)
376	return true;
377	if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) &&
378	attr == OVS_KEY_ATTR_CT_ZONE)
379	return true;
380
381	return false;
382	}
383
384	int ovs_ct_copy_action(struct net net, const struct nlattr attr,
385	const struct sw_flow_key *key,
386	struct sw_flow_actions **sfa, bool log)
387	{
388	struct ovs_conntrack_info ct_info;
389	u16 family;
390	int err;
391
392	family = key_to_nfproto(key);
393	if (family == NFPROTO_UNSPEC) {
394	OVS_NLERR(log, "ct family unspecified");
395	return -EINVAL;
396	}
397
398	memset(&ct_info, 0, sizeof(ct_info));
399	ct_info.family = family;
400
401	nf_ct_zone_init(&ct_info.zone, NF_CT_DEFAULT_ZONE_ID,
402	NF_CT_DEFAULT_ZONE_DIR, 0);
403
404	err = parse_ct(attr, &ct_info, log);
405	if (err)
406	return err;
407
408	/* Set up template for tracking connections in specific zones. */
409	ct_info.ct = nf_ct_tmpl_alloc(net, &ct_info.zone, GFP_KERNEL);
410	if (!ct_info.ct) {
411	OVS_NLERR(log, "Failed to allocate conntrack template");
412	return -ENOMEM;
413	}
414
415	err = ovs_nla_add_action(sfa, OVS_ACTION_ATTR_CT, &ct_info,
416	sizeof(ct_info), log);
417	if (err)
418	goto err_free_ct;
419
420	__set_bit(IPS_CONFIRMED_BIT, &ct_info.ct->status);
421	nf_conntrack_get(&ct_info.ct->ct_general);
422	return 0;
423	err_free_ct:
424	nf_conntrack_free(ct_info.ct);
425	return err;
426	}
427
428	int ovs_ct_action_to_attr(const struct ovs_conntrack_info *ct_info,
429	struct sk_buff *skb)
430	{
431	struct nlattr *start;
432
433	start = nla_nest_start(skb, OVS_ACTION_ATTR_CT);
434	if (!start)
435	return -EMSGSIZE;
436
437	if (nla_put_u32(skb, OVS_CT_ATTR_FLAGS, ct_info->flags))
438	return -EMSGSIZE;
439	if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) &&
440	nla_put_u16(skb, OVS_CT_ATTR_ZONE, ct_info->zone.id))
441	return -EMSGSIZE;
442
443	nla_nest_end(skb, start);
444
445	return 0;
446	}
447
448	void ovs_ct_free_action(const struct nlattr *a)
449	{
450	struct ovs_conntrack_info *ct_info = nla_data(a);
451
452	if (ct_info->ct)
453	nf_ct_put(ct_info->ct);
454	}