[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_helper.h>
#include <net/netfilter/nf_conntrack_labels.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;
};

/* Metadata mark for masked write to conntrack mark */
struct md_mark {
	u32 value;
	u32 mask;
};

/* Metadata label for masked write to conntrack label. */
struct md_label {
	struct ovs_key_ct_label value;
	struct ovs_key_ct_label mask;
};

/* Conntrack action context for execution. */
struct ovs_conntrack_info {
	struct nf_conntrack_helper *helper;
	struct nf_conntrack_zone zone;
	struct nf_conn *ct;
	u32 flags;
	u16 family;
	struct md_mark mark;
	struct md_label label;
};

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_get_label(const struct nf_conn *ct,
			     struct ovs_key_ct_label *label)
{
	struct nf_conn_labels *cl = ct ? nf_ct_labels_find(ct) : NULL;

	if (cl) {
		size_t len = cl->words * sizeof(long);

		if (len > OVS_CT_LABEL_LEN)
			len = OVS_CT_LABEL_LEN;
		else if (len < OVS_CT_LABEL_LEN)
			memset(label, 0, OVS_CT_LABEL_LEN);
		memcpy(label, cl->bits, len);
	} else {
		memset(label, 0, OVS_CT_LABEL_LEN);
	}
}

static void __ovs_ct_update_key(struct sw_flow_key *key, u8 state,
				const struct nf_conntrack_zone *zone,
				const struct nf_conn *ct)
{
	key->ct.state = state;
	key->ct.zone = zone->id;
	key->ct.mark = ct ? ct->mark : 0;
	ovs_ct_get_label(ct, &key->ct.label);
}

/* 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, ct);
}

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;

	if (IS_ENABLED(CONFIG_NF_CONNTRACK_MARK) &&
	    nla_put_u32(skb, OVS_KEY_ATTR_CT_MARK, key->ct.mark))
		return -EMSGSIZE;

	if (IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS) &&
	    nla_put(skb, OVS_KEY_ATTR_CT_LABEL, sizeof(key->ct.label),
		    &key->ct.label))
		return -EMSGSIZE;

	return 0;
}

static int ovs_ct_set_mark(struct sk_buff *skb, struct sw_flow_key *key,
			   u32 ct_mark, u32 mask)
{
	enum ip_conntrack_info ctinfo;
	struct nf_conn *ct;
	u32 new_mark;

	if (!IS_ENABLED(CONFIG_NF_CONNTRACK_MARK))
		return -ENOTSUPP;

	/* The connection could be invalid, in which case set_mark is no-op. */
	ct = nf_ct_get(skb, &ctinfo);
	if (!ct)
		return 0;

	new_mark = ct_mark | (ct->mark & ~(mask));
	if (ct->mark != new_mark) {
		ct->mark = new_mark;
		nf_conntrack_event_cache(IPCT_MARK, ct);
		key->ct.mark = new_mark;
	}

	return 0;
}

static int ovs_ct_set_label(struct sk_buff *skb, struct sw_flow_key *key,
			    const struct ovs_key_ct_label *label,
			    const struct ovs_key_ct_label *mask)
{
	enum ip_conntrack_info ctinfo;
	struct nf_conn_labels *cl;
	struct nf_conn *ct;
	int err;

	if (!IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS))
		return -ENOTSUPP;

	/* The connection could be invalid, in which case set_label is no-op.*/
	ct = nf_ct_get(skb, &ctinfo);
	if (!ct)
		return 0;

	cl = nf_ct_labels_find(ct);
	if (!cl) {
		nf_ct_labels_ext_add(ct);
		cl = nf_ct_labels_find(ct);
	}
	if (!cl || cl->words * sizeof(long) < OVS_CT_LABEL_LEN)
		return -ENOSPC;

	err = nf_connlabels_replace(ct, (u32 *)label, (u32 *)mask,
				    OVS_CT_LABEL_LEN / sizeof(u32));
	if (err)
		return err;

	ovs_ct_get_label(ct, &key->ct.label);
	return 0;
}

/* 'skb' should already be pulled to nh_ofs. */
static int ovs_ct_helper(struct sk_buff *skb, u16 proto)
{
	const struct nf_conntrack_helper *helper;
	const struct nf_conn_help *help;
	enum ip_conntrack_info ctinfo;
	unsigned int protoff;
	struct nf_conn *ct;

	ct = nf_ct_get(skb, &ctinfo);
	if (!ct || ctinfo == IP_CT_RELATED_REPLY)
		return NF_ACCEPT;

	help = nfct_help(ct);
	if (!help)
		return NF_ACCEPT;

	helper = rcu_dereference(help->helper);
	if (!helper)
		return NF_ACCEPT;

	switch (proto) {
	case NFPROTO_IPV4:
		protoff = ip_hdrlen(skb);
		break;
	case NFPROTO_IPV6: {
		u8 nexthdr = ipv6_hdr(skb)->nexthdr;
		__be16 frag_off;

		protoff = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr),
					   &nexthdr, &frag_off);
		if (protoff < 0 || (frag_off & htons(~0x7)) != 0) {
			pr_debug("proto header not found\n");
			return NF_ACCEPT;
		}
		break;
	}
	default:
		WARN_ONCE(1, "helper invoked on non-IP family!");
		return NF_DROP;
	}

	return helper->help(skb, protoff, ct, ctinfo);
}

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;
	if (info->helper) {
		struct nf_conn_help *help;

		help = nf_ct_ext_find(ct, NF_CT_EXT_HELPER);
		if (help && rcu_access_pointer(help->helper) != info->helper)
			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;

		if (ovs_ct_helper(skb, info->family) != NF_ACCEPT) {
			WARN_ONCE(1, "helper rejected packet");
			return -EINVAL;
		}
	}

	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, exp->master);
	} 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;
}

static bool label_nonzero(const struct ovs_key_ct_label *label)
{
	size_t i;

	for (i = 0; i < sizeof(*label); i++)
		if (label->ct_label[i])
			return true;

	return false;
}

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);
	if (err)
		goto err;

	if (info->mark.mask) {
		err = ovs_ct_set_mark(skb, key, info->mark.value,
				      info->mark.mask);
		if (err)
			goto err;
	}
	if (label_nonzero(&info->label.mask))
		err = ovs_ct_set_label(skb, key, &info->label.value,
				       &info->label.mask);
err:
	skb_push(skb, nh_ofs);
	return err;
}

static int ovs_ct_add_helper(struct ovs_conntrack_info *info, const char *name,
			     const struct sw_flow_key *key, bool log)
{
	struct nf_conntrack_helper *helper;
	struct nf_conn_help *help;

	helper = nf_conntrack_helper_try_module_get(name, info->family,
						    key->ip.proto);
	if (!helper) {
		OVS_NLERR(log, "Unknown helper \"%s\"", name);
		return -EINVAL;
	}

	help = nf_ct_helper_ext_add(info->ct, helper, GFP_KERNEL);
	if (!help) {
		module_put(helper->me);
		return -ENOMEM;
	}

	rcu_assign_pointer(help->helper, helper);
	info->helper = helper;
	return 0;
}

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) },
	[OVS_CT_ATTR_MARK]	= { .minlen = sizeof(struct md_mark),
				    .maxlen = sizeof(struct md_mark) },
	[OVS_CT_ATTR_LABEL]	= { .minlen = sizeof(struct md_label),
				    .maxlen = sizeof(struct md_label) },
	[OVS_CT_ATTR_HELPER]	= { .minlen = 1,
				    .maxlen = NF_CT_HELPER_NAME_LEN }
};

static int parse_ct(const struct nlattr *attr, struct ovs_conntrack_info *info,
		    const char **helper, 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
#ifdef CONFIG_NF_CONNTRACK_MARK
		case OVS_CT_ATTR_MARK: {
			struct md_mark *mark = nla_data(a);

			info->mark = *mark;
			break;
		}
#endif
#ifdef CONFIG_NF_CONNTRACK_LABELS
		case OVS_CT_ATTR_LABEL: {
			struct md_label *label = nla_data(a);

			info->label = *label;
			break;
		}
#endif
		case OVS_CT_ATTR_HELPER:
			*helper = nla_data(a);
			if (!memchr(*helper, '\0', nla_len(a))) {
				OVS_NLERR(log, "Invalid conntrack helper");
				return -EINVAL;
			}
			break;
		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(struct net *net, 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;
	if (IS_ENABLED(CONFIG_NF_CONNTRACK_MARK) &&
	    attr == OVS_KEY_ATTR_CT_MARK)
		return true;
	if (IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS) &&
	    attr == OVS_KEY_ATTR_CT_LABEL) {
		struct ovs_net *ovs_net = net_generic(net, ovs_net_id);

		return ovs_net->xt_label;
	}

	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;
	const char *helper = NULL;
	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, &helper, 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;
	}
	if (helper) {
		err = ovs_ct_add_helper(&ct_info, helper, key, log);
		if (err)
			goto err_free_ct;
	}

	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;
	if (IS_ENABLED(CONFIG_NF_CONNTRACK_MARK) &&
	    nla_put(skb, OVS_CT_ATTR_MARK, sizeof(ct_info->mark),
		    &ct_info->mark))
		return -EMSGSIZE;
	if (IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS) &&
	    nla_put(skb, OVS_CT_ATTR_LABEL, sizeof(ct_info->label),
		    &ct_info->label))
		return -EMSGSIZE;
	if (ct_info->helper) {
		if (nla_put_string(skb, OVS_CT_ATTR_HELPER,
				   ct_info->helper->name))
			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->helper)
		module_put(ct_info->helper->me);
	if (ct_info->ct)
		nf_ct_put(ct_info->ct);
}

void ovs_ct_init(struct net *net)
{
	unsigned int n_bits = sizeof(struct ovs_key_ct_label) * BITS_PER_BYTE;
	struct ovs_net *ovs_net = net_generic(net, ovs_net_id);

	if (nf_connlabels_get(net, n_bits)) {
		ovs_net->xt_label = false;
		OVS_NLERR(true, "Failed to set connlabel length");
	} else {
		ovs_net->xt_label = true;
	}
}

void ovs_ct_exit(struct net *net)
{
	struct ovs_net *ovs_net = net_generic(net, ovs_net_id);

	if (ovs_net->xt_label)
		nf_connlabels_put(net);
}
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>
cae3a262	18	#include <net/netfilter/nf_conntrack_helper.h>
c2ac6673	19	#include <net/netfilter/nf_conntrack_labels.h>
7f8a436e JS	20	#include <net/netfilter/nf_conntrack_zones.h>
	21	#include <net/netfilter/ipv6/nf_defrag_ipv6.h>
	22
	23	#include "datapath.h"
	24	#include "conntrack.h"
	25	#include "flow.h"
	26	#include "flow_netlink.h"
	27
	28	struct ovs_ct_len_tbl {
	29	size_t maxlen;
	30	size_t minlen;
	31	};
	32
182e3042 JS	33	/* Metadata mark for masked write to conntrack mark */
	34	struct md_mark {
	35	u32 value;
	36	u32 mask;
	37	};
	38
c2ac6673 JS	39	/* Metadata label for masked write to conntrack label. */
	40	struct md_label {
	41	struct ovs_key_ct_label value;
	42	struct ovs_key_ct_label mask;
	43	};
	44
7f8a436e JS	45	/* Conntrack action context for execution. */
7f8a436e JS	46	struct ovs_conntrack_info {
cae3a262	47	struct nf_conntrack_helper *helper;
7f8a436e JS	48	struct nf_conntrack_zone zone;
	49	struct nf_conn *ct;
	50	u32 flags;
	51	u16 family;
182e3042	52	struct md_mark mark;
c2ac6673	53	struct md_label label;
7f8a436e JS	54	};
	55
	56	static u16 key_to_nfproto(const struct sw_flow_key *key)
	57	{
	58	switch (ntohs(key->eth.type)) {
	59	case ETH_P_IP:
	60	return NFPROTO_IPV4;
	61	case ETH_P_IPV6:
	62	return NFPROTO_IPV6;
	63	default:
	64	return NFPROTO_UNSPEC;
	65	}
	66	}
	67
	68	/* Map SKB connection state into the values used by flow definition. */
	69	static u8 ovs_ct_get_state(enum ip_conntrack_info ctinfo)
	70	{
	71	u8 ct_state = OVS_CS_F_TRACKED;
	72
	73	switch (ctinfo) {
	74	case IP_CT_ESTABLISHED_REPLY:
	75	case IP_CT_RELATED_REPLY:
	76	case IP_CT_NEW_REPLY:
	77	ct_state \|= OVS_CS_F_REPLY_DIR;
	78	break;
	79	default:
	80	break;
	81	}
	82
	83	switch (ctinfo) {
	84	case IP_CT_ESTABLISHED:
	85	case IP_CT_ESTABLISHED_REPLY:
	86	ct_state \|= OVS_CS_F_ESTABLISHED;
	87	break;
	88	case IP_CT_RELATED:
	89	case IP_CT_RELATED_REPLY:
	90	ct_state \|= OVS_CS_F_RELATED;
	91	break;
	92	case IP_CT_NEW:
	93	case IP_CT_NEW_REPLY:
	94	ct_state \|= OVS_CS_F_NEW;
	95	break;
	96	default:
	97	break;
	98	}
	99
	100	return ct_state;
	101	}
	102
c2ac6673 JS	103	static void ovs_ct_get_label(const struct nf_conn *ct,
	104	struct ovs_key_ct_label *label)
	105	{
	106	struct nf_conn_labels *cl = ct ? nf_ct_labels_find(ct) : NULL;
	107
	108	if (cl) {
	109	size_t len = cl->words * sizeof(long);
	110
	111	if (len > OVS_CT_LABEL_LEN)
	112	len = OVS_CT_LABEL_LEN;
	113	else if (len < OVS_CT_LABEL_LEN)
	114	memset(label, 0, OVS_CT_LABEL_LEN);
	115	memcpy(label, cl->bits, len);
	116	} else {
	117	memset(label, 0, OVS_CT_LABEL_LEN);
	118	}
	119	}
	120
7f8a436e	121	static void __ovs_ct_update_key(struct sw_flow_key *key, u8 state,
182e3042 JS	122	const struct nf_conntrack_zone *zone,
182e3042 JS	123	const struct nf_conn *ct)
7f8a436e JS	124	{
	125	key->ct.state = state;
	126	key->ct.zone = zone->id;
182e3042	127	key->ct.mark = ct ? ct->mark : 0;
c2ac6673	128	ovs_ct_get_label(ct, &key->ct.label);
7f8a436e JS	129	}
	130
	131	/* Update 'key' based on skb->nfct. If 'post_ct' is true, then OVS has
	132	* previously sent the packet to conntrack via the ct action.
	133	*/
	134	static void ovs_ct_update_key(const struct sk_buff *skb,
	135	struct sw_flow_key *key, bool post_ct)
	136	{
	137	const struct nf_conntrack_zone *zone = &nf_ct_zone_dflt;
	138	enum ip_conntrack_info ctinfo;
	139	struct nf_conn *ct;
	140	u8 state = 0;
	141
	142	ct = nf_ct_get(skb, &ctinfo);
	143	if (ct) {
	144	state = ovs_ct_get_state(ctinfo);
	145	if (ct->master)
	146	state \|= OVS_CS_F_RELATED;
	147	zone = nf_ct_zone(ct);
	148	} else if (post_ct) {
	149	state = OVS_CS_F_TRACKED \| OVS_CS_F_INVALID;
	150	}
182e3042	151	__ovs_ct_update_key(key, state, zone, ct);
7f8a436e JS	152	}
	153
	154	void ovs_ct_fill_key(const struct sk_buff skb, struct sw_flow_key key)
	155	{
	156	ovs_ct_update_key(skb, key, false);
	157	}
	158
	159	int ovs_ct_put_key(const struct sw_flow_key key, struct sk_buff skb)
	160	{
	161	if (nla_put_u8(skb, OVS_KEY_ATTR_CT_STATE, key->ct.state))
	162	return -EMSGSIZE;
	163
	164	if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) &&
	165	nla_put_u16(skb, OVS_KEY_ATTR_CT_ZONE, key->ct.zone))
	166	return -EMSGSIZE;
	167
182e3042 JS	168	if (IS_ENABLED(CONFIG_NF_CONNTRACK_MARK) &&
	169	nla_put_u32(skb, OVS_KEY_ATTR_CT_MARK, key->ct.mark))
	170	return -EMSGSIZE;
	171
9723e6ab	172	if (IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS) &&
c2ac6673 JS	173	nla_put(skb, OVS_KEY_ATTR_CT_LABEL, sizeof(key->ct.label),
	174	&key->ct.label))
	175	return -EMSGSIZE;
	176
182e3042 JS	177	return 0;
	178	}
	179
	180	static int ovs_ct_set_mark(struct sk_buff skb, struct sw_flow_key key,
	181	u32 ct_mark, u32 mask)
	182	{
	183	enum ip_conntrack_info ctinfo;
	184	struct nf_conn *ct;
	185	u32 new_mark;
	186
	187	if (!IS_ENABLED(CONFIG_NF_CONNTRACK_MARK))
	188	return -ENOTSUPP;
	189
	190	/* The connection could be invalid, in which case set_mark is no-op. */
	191	ct = nf_ct_get(skb, &ctinfo);
	192	if (!ct)
	193	return 0;
	194
	195	new_mark = ct_mark \| (ct->mark & ~(mask));
	196	if (ct->mark != new_mark) {
	197	ct->mark = new_mark;
	198	nf_conntrack_event_cache(IPCT_MARK, ct);
	199	key->ct.mark = new_mark;
	200	}
	201
7f8a436e JS	202	return 0;
	203	}
	204
c2ac6673 JS	205	static int ovs_ct_set_label(struct sk_buff skb, struct sw_flow_key key,
	206	const struct ovs_key_ct_label *label,
	207	const struct ovs_key_ct_label *mask)
	208	{
	209	enum ip_conntrack_info ctinfo;
	210	struct nf_conn_labels *cl;
	211	struct nf_conn *ct;
	212	int err;
	213
	214	if (!IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS))
	215	return -ENOTSUPP;
	216
	217	/* The connection could be invalid, in which case set_label is no-op.*/
	218	ct = nf_ct_get(skb, &ctinfo);
	219	if (!ct)
	220	return 0;
	221
	222	cl = nf_ct_labels_find(ct);
	223	if (!cl) {
	224	nf_ct_labels_ext_add(ct);
	225	cl = nf_ct_labels_find(ct);
	226	}
	227	if (!cl \|\| cl->words * sizeof(long) < OVS_CT_LABEL_LEN)
	228	return -ENOSPC;
	229
	230	err = nf_connlabels_replace(ct, (u32 )label, (u32 )mask,
	231	OVS_CT_LABEL_LEN / sizeof(u32));
	232	if (err)
	233	return err;
	234
	235	ovs_ct_get_label(ct, &key->ct.label);
	236	return 0;
	237	}
	238
cae3a262 JS	239	/* 'skb' should already be pulled to nh_ofs. */
	240	static int ovs_ct_helper(struct sk_buff *skb, u16 proto)
	241	{
	242	const struct nf_conntrack_helper *helper;
	243	const struct nf_conn_help *help;
	244	enum ip_conntrack_info ctinfo;
	245	unsigned int protoff;
	246	struct nf_conn *ct;
	247
	248	ct = nf_ct_get(skb, &ctinfo);
	249	if (!ct \|\| ctinfo == IP_CT_RELATED_REPLY)
	250	return NF_ACCEPT;
	251
	252	help = nfct_help(ct);
	253	if (!help)
	254	return NF_ACCEPT;
	255
	256	helper = rcu_dereference(help->helper);
	257	if (!helper)
	258	return NF_ACCEPT;
	259
	260	switch (proto) {
	261	case NFPROTO_IPV4:
	262	protoff = ip_hdrlen(skb);
	263	break;
	264	case NFPROTO_IPV6: {
	265	u8 nexthdr = ipv6_hdr(skb)->nexthdr;
	266	__be16 frag_off;
	267
	268	protoff = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr),
	269	&nexthdr, &frag_off);
	270	if (protoff < 0 \|\| (frag_off & htons(~0x7)) != 0) {
	271	pr_debug("proto header not found\n");
	272	return NF_ACCEPT;
	273	}
	274	break;
	275	}
	276	default:
	277	WARN_ONCE(1, "helper invoked on non-IP family!");
	278	return NF_DROP;
	279	}
	280
	281	return helper->help(skb, protoff, ct, ctinfo);
	282	}
	283
7f8a436e JS	284	static int handle_fragments(struct net net, struct sw_flow_key key,
	285	u16 zone, struct sk_buff *skb)
	286	{
	287	struct ovs_skb_cb ovs_cb = *OVS_CB(skb);
	288
	289	if (key->eth.type == htons(ETH_P_IP)) {
	290	enum ip_defrag_users user = IP_DEFRAG_CONNTRACK_IN + zone;
	291	int err;
	292
	293	memset(IPCB(skb), 0, sizeof(struct inet_skb_parm));
	294	err = ip_defrag(skb, user);
	295	if (err)
	296	return err;
	297
	298	ovs_cb.mru = IPCB(skb)->frag_max_size;
	299	} else if (key->eth.type == htons(ETH_P_IPV6)) {
	300	#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6)
	301	enum ip6_defrag_users user = IP6_DEFRAG_CONNTRACK_IN + zone;
	302	struct sk_buff *reasm;
	303
	304	memset(IP6CB(skb), 0, sizeof(struct inet6_skb_parm));
	305	reasm = nf_ct_frag6_gather(skb, user);
	306	if (!reasm)
	307	return -EINPROGRESS;
	308
	309	if (skb == reasm)
	310	return -EINVAL;
	311
	312	key->ip.proto = ipv6_hdr(reasm)->nexthdr;
	313	skb_morph(skb, reasm);
	314	consume_skb(reasm);
	315	ovs_cb.mru = IP6CB(skb)->frag_max_size;
	316	#else
	317	return -EPFNOSUPPORT;
	318	#endif
	319	} else {
	320	return -EPFNOSUPPORT;
	321	}
	322
	323	key->ip.frag = OVS_FRAG_TYPE_NONE;
	324	skb_clear_hash(skb);
	325	skb->ignore_df = 1;
	326	*OVS_CB(skb) = ovs_cb;
	327
	328	return 0;
	329	}
	330
	331	static struct nf_conntrack_expect *
	332	ovs_ct_expect_find(struct net net, const struct nf_conntrack_zone zone,
	333	u16 proto, const struct sk_buff *skb)
	334	{
	335	struct nf_conntrack_tuple tuple;
	336
	337	if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), proto, &tuple))
	338	return NULL;
	339	return __nf_ct_expect_find(net, zone, &tuple);
	340	}
	341
	342	/* Determine whether skb->nfct is equal to the result of conntrack lookup. */
	343	static bool skb_nfct_cached(const struct net net, const struct sk_buff skb,
	344	const struct ovs_conntrack_info *info)
	345	{
	346	enum ip_conntrack_info ctinfo;
	347	struct nf_conn *ct;
348
349	ct = nf_ct_get(skb, &ctinfo);
350	if (!ct)
351	return false;
352	if (!net_eq(net, read_pnet(&ct->ct_net)))
353	return false;
354	if (!nf_ct_zone_equal_any(info->ct, nf_ct_zone(ct)))
355	return false;
cae3a262 JS	356	if (info->helper) {
	357	struct nf_conn_help *help;
	358
	359	help = nf_ct_ext_find(ct, NF_CT_EXT_HELPER);
	360	if (help && rcu_access_pointer(help->helper) != info->helper)
	361	return false;
	362	}
7f8a436e JS	363
	364	return true;
	365	}
	366
	367	static int __ovs_ct_lookup(struct net net, const struct sw_flow_key key,
	368	const struct ovs_conntrack_info *info,
	369	struct sk_buff *skb)
	370	{
	371	/* If we are recirculating packets to match on conntrack fields and
	372	* committing with a separate conntrack action, then we don't need to
	373	* actually run the packet through conntrack twice unless it's for a
	374	* different zone.
	375	*/
	376	if (!skb_nfct_cached(net, skb, info)) {
	377	struct nf_conn *tmpl = info->ct;
	378
	379	/* Associate skb with specified zone. */
	380	if (tmpl) {
	381	if (skb->nfct)
	382	nf_conntrack_put(skb->nfct);
	383	nf_conntrack_get(&tmpl->ct_general);
	384	skb->nfct = &tmpl->ct_general;
	385	skb->nfctinfo = IP_CT_NEW;
	386	}
	387
	388	if (nf_conntrack_in(net, info->family, NF_INET_PRE_ROUTING,
	389	skb) != NF_ACCEPT)
	390	return -ENOENT;
cae3a262 JS	391
	392	if (ovs_ct_helper(skb, info->family) != NF_ACCEPT) {
	393	WARN_ONCE(1, "helper rejected packet");
	394	return -EINVAL;
	395	}
7f8a436e JS	396	}
	397
	398	return 0;
	399	}
	400
	401	/* Lookup connection and read fields into key. */
	402	static int ovs_ct_lookup(struct net net, struct sw_flow_key key,
	403	const struct ovs_conntrack_info *info,
	404	struct sk_buff *skb)
	405	{
	406	struct nf_conntrack_expect *exp;
	407
	408	exp = ovs_ct_expect_find(net, &info->zone, info->family, skb);
	409	if (exp) {
	410	u8 state;
	411
	412	state = OVS_CS_F_TRACKED \| OVS_CS_F_NEW \| OVS_CS_F_RELATED;
182e3042	413	__ovs_ct_update_key(key, state, &info->zone, exp->master);
7f8a436e JS	414	} else {
	415	int err;
	416
	417	err = __ovs_ct_lookup(net, key, info, skb);
	418	if (err)
	419	return err;
	420
	421	ovs_ct_update_key(skb, key, true);
	422	}
	423
	424	return 0;
	425	}
	426
	427	/* Lookup connection and confirm if unconfirmed. */
	428	static int ovs_ct_commit(struct net net, struct sw_flow_key key,
	429	const struct ovs_conntrack_info *info,
	430	struct sk_buff *skb)
	431	{
	432	u8 state;
	433	int err;
	434
	435	state = key->ct.state;
	436	if (key->ct.zone == info->zone.id &&
	437	((state & OVS_CS_F_TRACKED) && !(state & OVS_CS_F_NEW))) {
	438	/* Previous lookup has shown that this connection is already
	439	* tracked and committed. Skip committing.
	440	*/
	441	return 0;
	442	}
	443
	444	err = __ovs_ct_lookup(net, key, info, skb);
	445	if (err)
	446	return err;
	447	if (nf_conntrack_confirm(skb) != NF_ACCEPT)
	448	return -EINVAL;
	449
	450	ovs_ct_update_key(skb, key, true);
	451
	452	return 0;
	453	}
	454
c2ac6673 JS	455	static bool label_nonzero(const struct ovs_key_ct_label *label)
	456	{
	457	size_t i;
	458
	459	for (i = 0; i < sizeof(*label); i++)
	460	if (label->ct_label[i])
	461	return true;
	462
	463	return false;
	464	}
	465
7f8a436e JS	466	int ovs_ct_execute(struct net net, struct sk_buff skb,
	467	struct sw_flow_key *key,
	468	const struct ovs_conntrack_info *info)
	469	{
	470	int nh_ofs;
	471	int err;
	472
	473	/* The conntrack module expects to be working at L3. */
	474	nh_ofs = skb_network_offset(skb);
	475	skb_pull(skb, nh_ofs);
	476
	477	if (key->ip.frag != OVS_FRAG_TYPE_NONE) {
	478	err = handle_fragments(net, key, info->zone.id, skb);
	479	if (err)
	480	return err;
	481	}
	482
	483	if (info->flags & OVS_CT_F_COMMIT)
	484	err = ovs_ct_commit(net, key, info, skb);
	485	else
	486	err = ovs_ct_lookup(net, key, info, skb);
182e3042 JS	487	if (err)
182e3042 JS	488	goto err;
7f8a436e	489
c2ac6673	490	if (info->mark.mask) {
182e3042 JS	491	err = ovs_ct_set_mark(skb, key, info->mark.value,
182e3042 JS	492	info->mark.mask);
c2ac6673 JS	493	if (err)
	494	goto err;
	495	}
	496	if (label_nonzero(&info->label.mask))
	497	err = ovs_ct_set_label(skb, key, &info->label.value,
	498	&info->label.mask);
182e3042	499	err:
7f8a436e JS	500	skb_push(skb, nh_ofs);
	501	return err;
	502	}
	503
cae3a262 JS	504	static int ovs_ct_add_helper(struct ovs_conntrack_info info, const char name,
	505	const struct sw_flow_key *key, bool log)
	506	{
	507	struct nf_conntrack_helper *helper;
	508	struct nf_conn_help *help;
	509
	510	helper = nf_conntrack_helper_try_module_get(name, info->family,
	511	key->ip.proto);
	512	if (!helper) {
	513	OVS_NLERR(log, "Unknown helper \"%s\"", name);
	514	return -EINVAL;
	515	}
	516
	517	help = nf_ct_helper_ext_add(info->ct, helper, GFP_KERNEL);
	518	if (!help) {
	519	module_put(helper->me);
	520	return -ENOMEM;
	521	}
	522
	523	rcu_assign_pointer(help->helper, helper);
	524	info->helper = helper;
	525	return 0;
	526	}
	527
7f8a436e JS	528	static const struct ovs_ct_len_tbl ovs_ct_attr_lens[OVS_CT_ATTR_MAX + 1] = {
	529	[OVS_CT_ATTR_FLAGS] = { .minlen = sizeof(u32),
	530	.maxlen = sizeof(u32) },
	531	[OVS_CT_ATTR_ZONE] = { .minlen = sizeof(u16),
	532	.maxlen = sizeof(u16) },
182e3042 JS	533	[OVS_CT_ATTR_MARK] = { .minlen = sizeof(struct md_mark),
182e3042 JS	534	.maxlen = sizeof(struct md_mark) },
c2ac6673 JS	535	[OVS_CT_ATTR_LABEL] = { .minlen = sizeof(struct md_label),
c2ac6673 JS	536	.maxlen = sizeof(struct md_label) },
cae3a262 JS	537	[OVS_CT_ATTR_HELPER] = { .minlen = 1,
cae3a262 JS	538	.maxlen = NF_CT_HELPER_NAME_LEN }
7f8a436e JS	539	};
	540
	541	static int parse_ct(const struct nlattr attr, struct ovs_conntrack_info info,
cae3a262	542	const char **helper, bool log)
7f8a436e JS	543	{
	544	struct nlattr *a;
	545	int rem;
	546
	547	nla_for_each_nested(a, attr, rem) {
	548	int type = nla_type(a);
	549	int maxlen = ovs_ct_attr_lens[type].maxlen;
	550	int minlen = ovs_ct_attr_lens[type].minlen;
	551
	552	if (type > OVS_CT_ATTR_MAX) {
	553	OVS_NLERR(log,
	554	"Unknown conntrack attr (type=%d, max=%d)",
	555	type, OVS_CT_ATTR_MAX);
	556	return -EINVAL;
	557	}
	558	if (nla_len(a) < minlen \|\| nla_len(a) > maxlen) {
	559	OVS_NLERR(log,
	560	"Conntrack attr type has unexpected length (type=%d, length=%d, expected=%d)",
	561	type, nla_len(a), maxlen);
	562	return -EINVAL;
	563	}
	564
	565	switch (type) {
	566	case OVS_CT_ATTR_FLAGS:
	567	info->flags = nla_get_u32(a);
	568	break;
	569	#ifdef CONFIG_NF_CONNTRACK_ZONES
	570	case OVS_CT_ATTR_ZONE:
	571	info->zone.id = nla_get_u16(a);
	572	break;
182e3042 JS	573	#endif
	574	#ifdef CONFIG_NF_CONNTRACK_MARK
	575	case OVS_CT_ATTR_MARK: {
	576	struct md_mark *mark = nla_data(a);
	577
	578	info->mark = *mark;
	579	break;
	580	}
c2ac6673 JS	581	#endif
	582	#ifdef CONFIG_NF_CONNTRACK_LABELS
	583	case OVS_CT_ATTR_LABEL: {
	584	struct md_label *label = nla_data(a);
	585
	586	info->label = *label;
	587	break;
	588	}
7f8a436e	589	#endif
cae3a262 JS	590	case OVS_CT_ATTR_HELPER:
	591	*helper = nla_data(a);
	592	if (!memchr(*helper, '\0', nla_len(a))) {
	593	OVS_NLERR(log, "Invalid conntrack helper");
	594	return -EINVAL;
	595	}
	596	break;
7f8a436e JS	597	default:
	598	OVS_NLERR(log, "Unknown conntrack attr (%d)",
	599	type);
	600	return -EINVAL;
	601	}
	602	}
	603
	604	if (rem > 0) {
	605	OVS_NLERR(log, "Conntrack attr has %d unknown bytes", rem);
	606	return -EINVAL;
	607	}
	608
	609	return 0;
	610	}
	611
c2ac6673	612	bool ovs_ct_verify(struct net *net, enum ovs_key_attr attr)
7f8a436e JS	613	{
	614	if (attr == OVS_KEY_ATTR_CT_STATE)
	615	return true;
	616	if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) &&
	617	attr == OVS_KEY_ATTR_CT_ZONE)
	618	return true;
182e3042 JS	619	if (IS_ENABLED(CONFIG_NF_CONNTRACK_MARK) &&
	620	attr == OVS_KEY_ATTR_CT_MARK)
	621	return true;
c2ac6673 JS	622	if (IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS) &&
	623	attr == OVS_KEY_ATTR_CT_LABEL) {
	624	struct ovs_net *ovs_net = net_generic(net, ovs_net_id);
	625
	626	return ovs_net->xt_label;
	627	}
7f8a436e JS	628
	629	return false;
	630	}
	631
	632	int ovs_ct_copy_action(struct net net, const struct nlattr attr,
	633	const struct sw_flow_key *key,
	634	struct sw_flow_actions **sfa, bool log)
	635	{
	636	struct ovs_conntrack_info ct_info;
cae3a262	637	const char *helper = NULL;
7f8a436e JS	638	u16 family;
	639	int err;
	640
	641	family = key_to_nfproto(key);
	642	if (family == NFPROTO_UNSPEC) {
	643	OVS_NLERR(log, "ct family unspecified");
	644	return -EINVAL;
	645	}
	646
	647	memset(&ct_info, 0, sizeof(ct_info));
	648	ct_info.family = family;
	649
	650	nf_ct_zone_init(&ct_info.zone, NF_CT_DEFAULT_ZONE_ID,
	651	NF_CT_DEFAULT_ZONE_DIR, 0);
	652
cae3a262	653	err = parse_ct(attr, &ct_info, &helper, log);
7f8a436e JS	654	if (err)
	655	return err;
	656
	657	/* Set up template for tracking connections in specific zones. */
	658	ct_info.ct = nf_ct_tmpl_alloc(net, &ct_info.zone, GFP_KERNEL);
	659	if (!ct_info.ct) {
	660	OVS_NLERR(log, "Failed to allocate conntrack template");
	661	return -ENOMEM;
	662	}
cae3a262 JS	663	if (helper) {
	664	err = ovs_ct_add_helper(&ct_info, helper, key, log);
	665	if (err)
	666	goto err_free_ct;
	667	}
7f8a436e JS	668
	669	err = ovs_nla_add_action(sfa, OVS_ACTION_ATTR_CT, &ct_info,
	670	sizeof(ct_info), log);
	671	if (err)
	672	goto err_free_ct;
	673
	674	__set_bit(IPS_CONFIRMED_BIT, &ct_info.ct->status);
	675	nf_conntrack_get(&ct_info.ct->ct_general);
	676	return 0;
	677	err_free_ct:
	678	nf_conntrack_free(ct_info.ct);
	679	return err;
	680	}
	681
	682	int ovs_ct_action_to_attr(const struct ovs_conntrack_info *ct_info,
	683	struct sk_buff *skb)
	684	{
	685	struct nlattr *start;
	686
	687	start = nla_nest_start(skb, OVS_ACTION_ATTR_CT);
	688	if (!start)
	689	return -EMSGSIZE;
	690
	691	if (nla_put_u32(skb, OVS_CT_ATTR_FLAGS, ct_info->flags))
	692	return -EMSGSIZE;
	693	if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) &&
	694	nla_put_u16(skb, OVS_CT_ATTR_ZONE, ct_info->zone.id))
	695	return -EMSGSIZE;
182e3042 JS	696	if (IS_ENABLED(CONFIG_NF_CONNTRACK_MARK) &&
	697	nla_put(skb, OVS_CT_ATTR_MARK, sizeof(ct_info->mark),
	698	&ct_info->mark))
	699	return -EMSGSIZE;
c2ac6673 JS	700	if (IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS) &&
	701	nla_put(skb, OVS_CT_ATTR_LABEL, sizeof(ct_info->label),
	702	&ct_info->label))
	703	return -EMSGSIZE;
cae3a262 JS	704	if (ct_info->helper) {
	705	if (nla_put_string(skb, OVS_CT_ATTR_HELPER,
	706	ct_info->helper->name))
	707	return -EMSGSIZE;
	708	}
7f8a436e JS	709
	710	nla_nest_end(skb, start);
	711
	712	return 0;
	713	}
	714
	715	void ovs_ct_free_action(const struct nlattr *a)
	716	{
	717	struct ovs_conntrack_info *ct_info = nla_data(a);
	718
cae3a262 JS	719	if (ct_info->helper)
cae3a262 JS	720	module_put(ct_info->helper->me);
7f8a436e JS	721	if (ct_info->ct)
	722	nf_ct_put(ct_info->ct);
	723	}
c2ac6673 JS	724
	725	void ovs_ct_init(struct net *net)
	726	{
	727	unsigned int n_bits = sizeof(struct ovs_key_ct_label) * BITS_PER_BYTE;
	728	struct ovs_net *ovs_net = net_generic(net, ovs_net_id);
	729
	730	if (nf_connlabels_get(net, n_bits)) {
	731	ovs_net->xt_label = false;
	732	OVS_NLERR(true, "Failed to set connlabel length");
	733	} else {
	734	ovs_net->xt_label = true;
	735	}
	736	}
	737
	738	void ovs_ct_exit(struct net *net)
	739	{
	740	struct ovs_net *ovs_net = net_generic(net, ovs_net_id);
	741
	742	if (ovs_net->xt_label)
	743	nf_connlabels_put(net);
	744	}