[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_labels {
	struct ovs_key_ct_labels value;
	struct ovs_key_ct_labels 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_labels labels;
};

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 u32 ovs_ct_get_mark(const struct nf_conn *ct)
{
#if IS_ENABLED(CONFIG_NF_CONNTRACK_MARK)
	return ct ? ct->mark : 0;
#else
	return 0;
#endif
}

static void ovs_ct_get_labels(const struct nf_conn *ct,
			      struct ovs_key_ct_labels *labels)
{
	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_LABELS_LEN)
			len = OVS_CT_LABELS_LEN;
		else if (len < OVS_CT_LABELS_LEN)
			memset(labels, 0, OVS_CT_LABELS_LEN);
		memcpy(labels, cl->bits, len);
	} else {
		memset(labels, 0, OVS_CT_LABELS_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 = ovs_ct_get_mark(ct);
	ovs_ct_get_labels(ct, &key->ct.labels);
}

/* 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_LABELS, sizeof(key->ct.labels),
		    &key->ct.labels))
		return -EMSGSIZE;

	return 0;
}

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


	/* 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;
#else
	return -ENOTSUPP;
#endif
}

static int ovs_ct_set_labels(struct sk_buff *skb, struct sw_flow_key *key,
			     const struct ovs_key_ct_labels *labels,
			     const struct ovs_key_ct_labels *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_LABELS_LEN)
		return -ENOSPC;

	err = nf_connlabels_replace(ct, (u32 *)labels, (u32 *)mask,
				    OVS_CT_LABELS_LEN / sizeof(u32));
	if (err)
		return err;

	ovs_ct_get_labels(ct, &key->ct.labels);
	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;
		int ofs;

		ofs = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), &nexthdr,
				       &frag_off);
		if (ofs < 0 || (frag_off & htons(~0x7)) != 0) {
			pr_debug("proto header not found\n");
			return NF_ACCEPT;
		}
		protoff = ofs;
		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 labels_nonzero(const struct ovs_key_ct_labels *labels)
{
	size_t i;

	for (i = 0; i < sizeof(*labels); i++)
		if (labels->ct_labels[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 (labels_nonzero(&info->labels.mask))
		err = ovs_ct_set_labels(skb, key, &info->labels.value,
					&info->labels.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_LABELS]	= { .minlen = sizeof(struct md_labels),
				    .maxlen = sizeof(struct md_labels) },
	[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_LABELS: {
			struct md_labels *labels = nla_data(a);

			info->labels = *labels;
			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_LABELS) {
		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_LABELS, sizeof(ct_info->labels),
		    &ct_info->labels))
		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_labels) * 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	39	/* Metadata label for masked write to conntrack label. */
33db4125 JS	40	struct md_labels {
	41	struct ovs_key_ct_labels value;
	42	struct ovs_key_ct_labels mask;
c2ac6673 JS	43	};
c2ac6673 JS	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;
33db4125	53	struct md_labels labels;
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
0d5cdef8 JS	103	static u32 ovs_ct_get_mark(const struct nf_conn *ct)
	104	{
	105	#if IS_ENABLED(CONFIG_NF_CONNTRACK_MARK)
	106	return ct ? ct->mark : 0;
	107	#else
	108	return 0;
	109	#endif
	110	}
	111
33db4125 JS	112	static void ovs_ct_get_labels(const struct nf_conn *ct,
33db4125 JS	113	struct ovs_key_ct_labels *labels)
c2ac6673 JS	114	{
	115	struct nf_conn_labels *cl = ct ? nf_ct_labels_find(ct) : NULL;
	116
	117	if (cl) {
	118	size_t len = cl->words * sizeof(long);
	119
33db4125 JS	120	if (len > OVS_CT_LABELS_LEN)
	121	len = OVS_CT_LABELS_LEN;
	122	else if (len < OVS_CT_LABELS_LEN)
	123	memset(labels, 0, OVS_CT_LABELS_LEN);
	124	memcpy(labels, cl->bits, len);
c2ac6673	125	} else {
33db4125	126	memset(labels, 0, OVS_CT_LABELS_LEN);
c2ac6673 JS	127	}
	128	}
	129
7f8a436e	130	static void __ovs_ct_update_key(struct sw_flow_key *key, u8 state,
182e3042 JS	131	const struct nf_conntrack_zone *zone,
182e3042 JS	132	const struct nf_conn *ct)
7f8a436e JS	133	{
	134	key->ct.state = state;
	135	key->ct.zone = zone->id;
0d5cdef8	136	key->ct.mark = ovs_ct_get_mark(ct);
33db4125	137	ovs_ct_get_labels(ct, &key->ct.labels);
7f8a436e JS	138	}
	139
	140	/* Update 'key' based on skb->nfct. If 'post_ct' is true, then OVS has
	141	* previously sent the packet to conntrack via the ct action.
	142	*/
	143	static void ovs_ct_update_key(const struct sk_buff *skb,
	144	struct sw_flow_key *key, bool post_ct)
	145	{
	146	const struct nf_conntrack_zone *zone = &nf_ct_zone_dflt;
	147	enum ip_conntrack_info ctinfo;
	148	struct nf_conn *ct;
	149	u8 state = 0;
	150
	151	ct = nf_ct_get(skb, &ctinfo);
	152	if (ct) {
	153	state = ovs_ct_get_state(ctinfo);
	154	if (ct->master)
	155	state \|= OVS_CS_F_RELATED;
	156	zone = nf_ct_zone(ct);
	157	} else if (post_ct) {
	158	state = OVS_CS_F_TRACKED \| OVS_CS_F_INVALID;
	159	}
182e3042	160	__ovs_ct_update_key(key, state, zone, ct);
7f8a436e JS	161	}
	162
	163	void ovs_ct_fill_key(const struct sk_buff skb, struct sw_flow_key key)
	164	{
	165	ovs_ct_update_key(skb, key, false);
	166	}
	167
	168	int ovs_ct_put_key(const struct sw_flow_key key, struct sk_buff skb)
	169	{
	170	if (nla_put_u8(skb, OVS_KEY_ATTR_CT_STATE, key->ct.state))
	171	return -EMSGSIZE;
	172
	173	if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) &&
	174	nla_put_u16(skb, OVS_KEY_ATTR_CT_ZONE, key->ct.zone))
	175	return -EMSGSIZE;
	176
182e3042 JS	177	if (IS_ENABLED(CONFIG_NF_CONNTRACK_MARK) &&
	178	nla_put_u32(skb, OVS_KEY_ATTR_CT_MARK, key->ct.mark))
	179	return -EMSGSIZE;
	180
9723e6ab	181	if (IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS) &&
33db4125 JS	182	nla_put(skb, OVS_KEY_ATTR_CT_LABELS, sizeof(key->ct.labels),
33db4125 JS	183	&key->ct.labels))
c2ac6673 JS	184	return -EMSGSIZE;
c2ac6673 JS	185
182e3042 JS	186	return 0;
	187	}
	188
	189	static int ovs_ct_set_mark(struct sk_buff skb, struct sw_flow_key key,
	190	u32 ct_mark, u32 mask)
	191	{
0d5cdef8	192	#if IS_ENABLED(CONFIG_NF_CONNTRACK_MARK)
182e3042 JS	193	enum ip_conntrack_info ctinfo;
	194	struct nf_conn *ct;
	195	u32 new_mark;
	196
182e3042 JS	197
	198	/* The connection could be invalid, in which case set_mark is no-op. */
	199	ct = nf_ct_get(skb, &ctinfo);
	200	if (!ct)
	201	return 0;
	202
	203	new_mark = ct_mark \| (ct->mark & ~(mask));
	204	if (ct->mark != new_mark) {
	205	ct->mark = new_mark;
	206	nf_conntrack_event_cache(IPCT_MARK, ct);
	207	key->ct.mark = new_mark;
	208	}
	209
7f8a436e	210	return 0;
0d5cdef8 JS	211	#else
	212	return -ENOTSUPP;
	213	#endif
7f8a436e JS	214	}
7f8a436e JS	215
33db4125 JS	216	static int ovs_ct_set_labels(struct sk_buff skb, struct sw_flow_key key,
	217	const struct ovs_key_ct_labels *labels,
	218	const struct ovs_key_ct_labels *mask)
c2ac6673 JS	219	{
	220	enum ip_conntrack_info ctinfo;
	221	struct nf_conn_labels *cl;
	222	struct nf_conn *ct;
	223	int err;
	224
	225	if (!IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS))
	226	return -ENOTSUPP;
	227
	228	/* The connection could be invalid, in which case set_label is no-op.*/
	229	ct = nf_ct_get(skb, &ctinfo);
	230	if (!ct)
	231	return 0;
	232
	233	cl = nf_ct_labels_find(ct);
	234	if (!cl) {
	235	nf_ct_labels_ext_add(ct);
	236	cl = nf_ct_labels_find(ct);
	237	}
33db4125	238	if (!cl \|\| cl->words * sizeof(long) < OVS_CT_LABELS_LEN)
c2ac6673 JS	239	return -ENOSPC;
c2ac6673 JS	240
33db4125 JS	241	err = nf_connlabels_replace(ct, (u32 )labels, (u32 )mask,
33db4125 JS	242	OVS_CT_LABELS_LEN / sizeof(u32));
c2ac6673 JS	243	if (err)
	244	return err;
	245
33db4125	246	ovs_ct_get_labels(ct, &key->ct.labels);
c2ac6673 JS	247	return 0;
	248	}
	249
cae3a262 JS	250	/* 'skb' should already be pulled to nh_ofs. */
	251	static int ovs_ct_helper(struct sk_buff *skb, u16 proto)
	252	{
	253	const struct nf_conntrack_helper *helper;
	254	const struct nf_conn_help *help;
	255	enum ip_conntrack_info ctinfo;
	256	unsigned int protoff;
	257	struct nf_conn *ct;
	258
	259	ct = nf_ct_get(skb, &ctinfo);
	260	if (!ct \|\| ctinfo == IP_CT_RELATED_REPLY)
	261	return NF_ACCEPT;
	262
	263	help = nfct_help(ct);
	264	if (!help)
	265	return NF_ACCEPT;
	266
	267	helper = rcu_dereference(help->helper);
	268	if (!helper)
	269	return NF_ACCEPT;
	270
	271	switch (proto) {
	272	case NFPROTO_IPV4:
	273	protoff = ip_hdrlen(skb);
	274	break;
	275	case NFPROTO_IPV6: {
	276	u8 nexthdr = ipv6_hdr(skb)->nexthdr;
	277	__be16 frag_off;
cc570605	278	int ofs;
cae3a262	279
cc570605 JS	280	ofs = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), &nexthdr,
	281	&frag_off);
	282	if (ofs < 0 \|\| (frag_off & htons(~0x7)) != 0) {
cae3a262 JS	283	pr_debug("proto header not found\n");
	284	return NF_ACCEPT;
	285	}
cc570605	286	protoff = ofs;
cae3a262 JS	287	break;
	288	}
	289	default:
	290	WARN_ONCE(1, "helper invoked on non-IP family!");
	291	return NF_DROP;
	292	}
	293
	294	return helper->help(skb, protoff, ct, ctinfo);
	295	}
	296
7f8a436e JS	297	static int handle_fragments(struct net net, struct sw_flow_key key,
	298	u16 zone, struct sk_buff *skb)
	299	{
	300	struct ovs_skb_cb ovs_cb = *OVS_CB(skb);
	301
	302	if (key->eth.type == htons(ETH_P_IP)) {
	303	enum ip_defrag_users user = IP_DEFRAG_CONNTRACK_IN + zone;
	304	int err;
	305
	306	memset(IPCB(skb), 0, sizeof(struct inet_skb_parm));
	307	err = ip_defrag(skb, user);
	308	if (err)
	309	return err;
	310
	311	ovs_cb.mru = IPCB(skb)->frag_max_size;
	312	} else if (key->eth.type == htons(ETH_P_IPV6)) {
	313	#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6)
	314	enum ip6_defrag_users user = IP6_DEFRAG_CONNTRACK_IN + zone;
	315	struct sk_buff *reasm;
	316
	317	memset(IP6CB(skb), 0, sizeof(struct inet6_skb_parm));
	318	reasm = nf_ct_frag6_gather(skb, user);
	319	if (!reasm)
	320	return -EINPROGRESS;
	321
	322	if (skb == reasm)
	323	return -EINVAL;
	324
	325	key->ip.proto = ipv6_hdr(reasm)->nexthdr;
	326	skb_morph(skb, reasm);
	327	consume_skb(reasm);
	328	ovs_cb.mru = IP6CB(skb)->frag_max_size;
	329	#else
	330	return -EPFNOSUPPORT;
	331	#endif
	332	} else {
	333	return -EPFNOSUPPORT;
	334	}
	335
	336	key->ip.frag = OVS_FRAG_TYPE_NONE;
	337	skb_clear_hash(skb);
	338	skb->ignore_df = 1;
	339	*OVS_CB(skb) = ovs_cb;
	340
	341	return 0;
	342	}
	343
	344	static struct nf_conntrack_expect *
	345	ovs_ct_expect_find(struct net net, const struct nf_conntrack_zone zone,
	346	u16 proto, const struct sk_buff *skb)
	347	{
	348	struct nf_conntrack_tuple tuple;
	349
	350	if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), proto, &tuple))
	351	return NULL;
	352	return __nf_ct_expect_find(net, zone, &tuple);
	353	}
	354
	355	/* Determine whether skb->nfct is equal to the result of conntrack lookup. */
	356	static bool skb_nfct_cached(const struct net net, const struct sk_buff skb,
	357	const struct ovs_conntrack_info *info)
	358	{
	359	enum ip_conntrack_info ctinfo;
	360	struct nf_conn *ct;
361
362	ct = nf_ct_get(skb, &ctinfo);
363	if (!ct)
364	return false;
365	if (!net_eq(net, read_pnet(&ct->ct_net)))
366	return false;
367	if (!nf_ct_zone_equal_any(info->ct, nf_ct_zone(ct)))
368	return false;
cae3a262 JS	369	if (info->helper) {
	370	struct nf_conn_help *help;
	371
	372	help = nf_ct_ext_find(ct, NF_CT_EXT_HELPER);
	373	if (help && rcu_access_pointer(help->helper) != info->helper)
	374	return false;
	375	}
7f8a436e JS	376
	377	return true;
	378	}
	379
	380	static int __ovs_ct_lookup(struct net net, const struct sw_flow_key key,
	381	const struct ovs_conntrack_info *info,
	382	struct sk_buff *skb)
	383	{
	384	/* If we are recirculating packets to match on conntrack fields and
	385	* committing with a separate conntrack action, then we don't need to
	386	* actually run the packet through conntrack twice unless it's for a
	387	* different zone.
	388	*/
	389	if (!skb_nfct_cached(net, skb, info)) {
	390	struct nf_conn *tmpl = info->ct;
	391
	392	/* Associate skb with specified zone. */
	393	if (tmpl) {
	394	if (skb->nfct)
	395	nf_conntrack_put(skb->nfct);
	396	nf_conntrack_get(&tmpl->ct_general);
	397	skb->nfct = &tmpl->ct_general;
	398	skb->nfctinfo = IP_CT_NEW;
	399	}
	400
	401	if (nf_conntrack_in(net, info->family, NF_INET_PRE_ROUTING,
	402	skb) != NF_ACCEPT)
	403	return -ENOENT;
cae3a262 JS	404
	405	if (ovs_ct_helper(skb, info->family) != NF_ACCEPT) {
	406	WARN_ONCE(1, "helper rejected packet");
	407	return -EINVAL;
	408	}
7f8a436e JS	409	}
	410
	411	return 0;
	412	}
	413
	414	/* Lookup connection and read fields into key. */
	415	static int ovs_ct_lookup(struct net net, struct sw_flow_key key,
	416	const struct ovs_conntrack_info *info,
	417	struct sk_buff *skb)
	418	{
	419	struct nf_conntrack_expect *exp;
	420
	421	exp = ovs_ct_expect_find(net, &info->zone, info->family, skb);
	422	if (exp) {
	423	u8 state;
	424
	425	state = OVS_CS_F_TRACKED \| OVS_CS_F_NEW \| OVS_CS_F_RELATED;
182e3042	426	__ovs_ct_update_key(key, state, &info->zone, exp->master);
7f8a436e JS	427	} else {
	428	int err;
	429
	430	err = __ovs_ct_lookup(net, key, info, skb);
	431	if (err)
	432	return err;
	433
	434	ovs_ct_update_key(skb, key, true);
	435	}
	436
	437	return 0;
	438	}
	439
	440	/* Lookup connection and confirm if unconfirmed. */
	441	static int ovs_ct_commit(struct net net, struct sw_flow_key key,
	442	const struct ovs_conntrack_info *info,
	443	struct sk_buff *skb)
	444	{
	445	u8 state;
	446	int err;
	447
	448	state = key->ct.state;
	449	if (key->ct.zone == info->zone.id &&
	450	((state & OVS_CS_F_TRACKED) && !(state & OVS_CS_F_NEW))) {
	451	/* Previous lookup has shown that this connection is already
	452	* tracked and committed. Skip committing.
	453	*/
	454	return 0;
	455	}
	456
	457	err = __ovs_ct_lookup(net, key, info, skb);
	458	if (err)
	459	return err;
	460	if (nf_conntrack_confirm(skb) != NF_ACCEPT)
	461	return -EINVAL;
	462
	463	ovs_ct_update_key(skb, key, true);
	464
	465	return 0;
	466	}
	467
33db4125	468	static bool labels_nonzero(const struct ovs_key_ct_labels *labels)
c2ac6673 JS	469	{
	470	size_t i;
	471
33db4125 JS	472	for (i = 0; i < sizeof(*labels); i++)
33db4125 JS	473	if (labels->ct_labels[i])
c2ac6673 JS	474	return true;
	475
	476	return false;
	477	}
	478
7f8a436e JS	479	int ovs_ct_execute(struct net net, struct sk_buff skb,
	480	struct sw_flow_key *key,
	481	const struct ovs_conntrack_info *info)
	482	{
	483	int nh_ofs;
	484	int err;
	485
	486	/* The conntrack module expects to be working at L3. */
	487	nh_ofs = skb_network_offset(skb);
	488	skb_pull(skb, nh_ofs);
	489
	490	if (key->ip.frag != OVS_FRAG_TYPE_NONE) {
	491	err = handle_fragments(net, key, info->zone.id, skb);
	492	if (err)
	493	return err;
	494	}
	495
	496	if (info->flags & OVS_CT_F_COMMIT)
	497	err = ovs_ct_commit(net, key, info, skb);
	498	else
	499	err = ovs_ct_lookup(net, key, info, skb);
182e3042 JS	500	if (err)
182e3042 JS	501	goto err;
7f8a436e	502
c2ac6673	503	if (info->mark.mask) {
182e3042 JS	504	err = ovs_ct_set_mark(skb, key, info->mark.value,
182e3042 JS	505	info->mark.mask);
c2ac6673 JS	506	if (err)
	507	goto err;
	508	}
33db4125 JS	509	if (labels_nonzero(&info->labels.mask))
	510	err = ovs_ct_set_labels(skb, key, &info->labels.value,
	511	&info->labels.mask);
182e3042	512	err:
7f8a436e JS	513	skb_push(skb, nh_ofs);
	514	return err;
	515	}
	516
cae3a262 JS	517	static int ovs_ct_add_helper(struct ovs_conntrack_info info, const char name,
	518	const struct sw_flow_key *key, bool log)
	519	{
	520	struct nf_conntrack_helper *helper;
	521	struct nf_conn_help *help;
	522
	523	helper = nf_conntrack_helper_try_module_get(name, info->family,
	524	key->ip.proto);
	525	if (!helper) {
	526	OVS_NLERR(log, "Unknown helper \"%s\"", name);
	527	return -EINVAL;
	528	}
	529
	530	help = nf_ct_helper_ext_add(info->ct, helper, GFP_KERNEL);
	531	if (!help) {
	532	module_put(helper->me);
	533	return -ENOMEM;
	534	}
	535
	536	rcu_assign_pointer(help->helper, helper);
	537	info->helper = helper;
	538	return 0;
	539	}
	540
7f8a436e JS	541	static const struct ovs_ct_len_tbl ovs_ct_attr_lens[OVS_CT_ATTR_MAX + 1] = {
	542	[OVS_CT_ATTR_FLAGS] = { .minlen = sizeof(u32),
	543	.maxlen = sizeof(u32) },
	544	[OVS_CT_ATTR_ZONE] = { .minlen = sizeof(u16),
	545	.maxlen = sizeof(u16) },
182e3042 JS	546	[OVS_CT_ATTR_MARK] = { .minlen = sizeof(struct md_mark),
182e3042 JS	547	.maxlen = sizeof(struct md_mark) },
33db4125 JS	548	[OVS_CT_ATTR_LABELS] = { .minlen = sizeof(struct md_labels),
33db4125 JS	549	.maxlen = sizeof(struct md_labels) },
cae3a262 JS	550	[OVS_CT_ATTR_HELPER] = { .minlen = 1,
cae3a262 JS	551	.maxlen = NF_CT_HELPER_NAME_LEN }
7f8a436e JS	552	};
	553
	554	static int parse_ct(const struct nlattr attr, struct ovs_conntrack_info info,
cae3a262	555	const char **helper, bool log)
7f8a436e JS	556	{
	557	struct nlattr *a;
	558	int rem;
	559
	560	nla_for_each_nested(a, attr, rem) {
	561	int type = nla_type(a);
	562	int maxlen = ovs_ct_attr_lens[type].maxlen;
	563	int minlen = ovs_ct_attr_lens[type].minlen;
	564
	565	if (type > OVS_CT_ATTR_MAX) {
	566	OVS_NLERR(log,
	567	"Unknown conntrack attr (type=%d, max=%d)",
	568	type, OVS_CT_ATTR_MAX);
	569	return -EINVAL;
	570	}
	571	if (nla_len(a) < minlen \|\| nla_len(a) > maxlen) {
	572	OVS_NLERR(log,
	573	"Conntrack attr type has unexpected length (type=%d, length=%d, expected=%d)",
	574	type, nla_len(a), maxlen);
	575	return -EINVAL;
	576	}
	577
	578	switch (type) {
	579	case OVS_CT_ATTR_FLAGS:
	580	info->flags = nla_get_u32(a);
	581	break;
	582	#ifdef CONFIG_NF_CONNTRACK_ZONES
	583	case OVS_CT_ATTR_ZONE:
	584	info->zone.id = nla_get_u16(a);
	585	break;
182e3042 JS	586	#endif
	587	#ifdef CONFIG_NF_CONNTRACK_MARK
	588	case OVS_CT_ATTR_MARK: {
	589	struct md_mark *mark = nla_data(a);
	590
	591	info->mark = *mark;
	592	break;
	593	}
c2ac6673 JS	594	#endif
c2ac6673 JS	595	#ifdef CONFIG_NF_CONNTRACK_LABELS
33db4125 JS	596	case OVS_CT_ATTR_LABELS: {
33db4125 JS	597	struct md_labels *labels = nla_data(a);
c2ac6673	598
33db4125	599	info->labels = *labels;
c2ac6673 JS	600	break;
c2ac6673 JS	601	}
7f8a436e	602	#endif
cae3a262 JS	603	case OVS_CT_ATTR_HELPER:
	604	*helper = nla_data(a);
	605	if (!memchr(*helper, '\0', nla_len(a))) {
	606	OVS_NLERR(log, "Invalid conntrack helper");
	607	return -EINVAL;
	608	}
	609	break;
7f8a436e JS	610	default:
	611	OVS_NLERR(log, "Unknown conntrack attr (%d)",
	612	type);
	613	return -EINVAL;
	614	}
	615	}
	616
	617	if (rem > 0) {
	618	OVS_NLERR(log, "Conntrack attr has %d unknown bytes", rem);
	619	return -EINVAL;
	620	}
	621
	622	return 0;
	623	}
	624
c2ac6673	625	bool ovs_ct_verify(struct net *net, enum ovs_key_attr attr)
7f8a436e JS	626	{
	627	if (attr == OVS_KEY_ATTR_CT_STATE)
	628	return true;
	629	if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) &&
	630	attr == OVS_KEY_ATTR_CT_ZONE)
	631	return true;
182e3042 JS	632	if (IS_ENABLED(CONFIG_NF_CONNTRACK_MARK) &&
	633	attr == OVS_KEY_ATTR_CT_MARK)
	634	return true;
c2ac6673	635	if (IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS) &&
33db4125	636	attr == OVS_KEY_ATTR_CT_LABELS) {
c2ac6673 JS	637	struct ovs_net *ovs_net = net_generic(net, ovs_net_id);
	638
	639	return ovs_net->xt_label;
	640	}
7f8a436e JS	641
	642	return false;
	643	}
	644
	645	int ovs_ct_copy_action(struct net net, const struct nlattr attr,
	646	const struct sw_flow_key *key,
	647	struct sw_flow_actions **sfa, bool log)
	648	{
	649	struct ovs_conntrack_info ct_info;
cae3a262	650	const char *helper = NULL;
7f8a436e JS	651	u16 family;
	652	int err;
	653
	654	family = key_to_nfproto(key);
	655	if (family == NFPROTO_UNSPEC) {
	656	OVS_NLERR(log, "ct family unspecified");
	657	return -EINVAL;
	658	}
	659
	660	memset(&ct_info, 0, sizeof(ct_info));
	661	ct_info.family = family;
	662
	663	nf_ct_zone_init(&ct_info.zone, NF_CT_DEFAULT_ZONE_ID,
	664	NF_CT_DEFAULT_ZONE_DIR, 0);
	665
cae3a262	666	err = parse_ct(attr, &ct_info, &helper, log);
7f8a436e JS	667	if (err)
	668	return err;
	669
	670	/* Set up template for tracking connections in specific zones. */
	671	ct_info.ct = nf_ct_tmpl_alloc(net, &ct_info.zone, GFP_KERNEL);
	672	if (!ct_info.ct) {
	673	OVS_NLERR(log, "Failed to allocate conntrack template");
	674	return -ENOMEM;
	675	}
cae3a262 JS	676	if (helper) {
	677	err = ovs_ct_add_helper(&ct_info, helper, key, log);
	678	if (err)
	679	goto err_free_ct;
	680	}
7f8a436e JS	681
	682	err = ovs_nla_add_action(sfa, OVS_ACTION_ATTR_CT, &ct_info,
	683	sizeof(ct_info), log);
	684	if (err)
	685	goto err_free_ct;
	686
	687	__set_bit(IPS_CONFIRMED_BIT, &ct_info.ct->status);
	688	nf_conntrack_get(&ct_info.ct->ct_general);
	689	return 0;
	690	err_free_ct:
	691	nf_conntrack_free(ct_info.ct);
	692	return err;
	693	}
	694
	695	int ovs_ct_action_to_attr(const struct ovs_conntrack_info *ct_info,
	696	struct sk_buff *skb)
	697	{
	698	struct nlattr *start;
	699
	700	start = nla_nest_start(skb, OVS_ACTION_ATTR_CT);
	701	if (!start)
	702	return -EMSGSIZE;
	703
	704	if (nla_put_u32(skb, OVS_CT_ATTR_FLAGS, ct_info->flags))
	705	return -EMSGSIZE;
	706	if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) &&
	707	nla_put_u16(skb, OVS_CT_ATTR_ZONE, ct_info->zone.id))
	708	return -EMSGSIZE;
182e3042 JS	709	if (IS_ENABLED(CONFIG_NF_CONNTRACK_MARK) &&
	710	nla_put(skb, OVS_CT_ATTR_MARK, sizeof(ct_info->mark),
	711	&ct_info->mark))
	712	return -EMSGSIZE;
c2ac6673	713	if (IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS) &&
33db4125 JS	714	nla_put(skb, OVS_CT_ATTR_LABELS, sizeof(ct_info->labels),
33db4125 JS	715	&ct_info->labels))
c2ac6673	716	return -EMSGSIZE;
cae3a262 JS	717	if (ct_info->helper) {
	718	if (nla_put_string(skb, OVS_CT_ATTR_HELPER,
	719	ct_info->helper->name))
	720	return -EMSGSIZE;
	721	}
7f8a436e JS	722
	723	nla_nest_end(skb, start);
	724
	725	return 0;
	726	}
	727
	728	void ovs_ct_free_action(const struct nlattr *a)
	729	{
	730	struct ovs_conntrack_info *ct_info = nla_data(a);
	731
cae3a262 JS	732	if (ct_info->helper)
cae3a262 JS	733	module_put(ct_info->helper->me);
7f8a436e JS	734	if (ct_info->ct)
	735	nf_ct_put(ct_info->ct);
	736	}
c2ac6673 JS	737
	738	void ovs_ct_init(struct net *net)
	739	{
33db4125	740	unsigned int n_bits = sizeof(struct ovs_key_ct_labels) * BITS_PER_BYTE;
c2ac6673 JS	741	struct ovs_net *ovs_net = net_generic(net, ovs_net_id);
	742
	743	if (nf_connlabels_get(net, n_bits)) {
	744	ovs_net->xt_label = false;
	745	OVS_NLERR(true, "Failed to set connlabel length");
	746	} else {
	747	ovs_net->xt_label = true;
	748	}
	749	}
	750
	751	void ovs_ct_exit(struct net *net)
	752	{
	753	struct ovs_net *ovs_net = net_generic(net, ovs_net_id);
	754
	755	if (ovs_net->xt_label)
	756	nf_connlabels_put(net);
	757	}