Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 2a5538e9 authored by Pablo Neira Ayuso's avatar Pablo Neira Ayuso
Browse files

netfilter: nat: move specific NAT IPv6 to core



Move the specific NAT IPv6 core functions that are called from the
hooks from ip6table_nat.c to nf_nat_l3proto_ipv6.c. This prepares the
ground to allow iptables and nft to use the same NAT engine code that
comes in a follow up patch.

This also renames nf_nat_ipv6_fn to nft_nat_ipv6_fn in
net/ipv6/netfilter/nft_chain_nat_ipv6.c to avoid a compilation breakage.

Signed-off-by: default avatarPablo Neira Ayuso <pablo@netfilter.org>
parent 65cd90ac
Loading
Loading
Loading
Loading
+37 −0
Original line number Diff line number Diff line
@@ -84,4 +84,41 @@ int nf_nat_icmpv6_reply_translation(struct sk_buff *skb, struct nf_conn *ct,
				    enum ip_conntrack_info ctinfo,
				    unsigned int hooknum, unsigned int hdrlen);

unsigned int nf_nat_ipv6_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
			    const struct net_device *in,
			    const struct net_device *out,
			    unsigned int (*do_chain)(const struct nf_hook_ops *ops,
						     struct sk_buff *skb,
						     const struct net_device *in,
						     const struct net_device *out,
						     struct nf_conn *ct));

unsigned int nf_nat_ipv6_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
			     const struct net_device *in,
			     const struct net_device *out,
			     unsigned int (*do_chain)(const struct nf_hook_ops *ops,
						      struct sk_buff *skb,
						      const struct net_device *in,
						      const struct net_device *out,
						      struct nf_conn *ct));

unsigned int nf_nat_ipv6_local_fn(const struct nf_hook_ops *ops,
				  struct sk_buff *skb,
				  const struct net_device *in,
				  const struct net_device *out,
				  unsigned int (*do_chain)(const struct nf_hook_ops *ops,
							   struct sk_buff *skb,
							   const struct net_device *in,
							   const struct net_device *out,
							   struct nf_conn *ct));

unsigned int nf_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
			    const struct net_device *in,
			    const struct net_device *out,
			    unsigned int (*do_chain)(const struct nf_hook_ops *ops,
						     struct sk_buff *skb,
						     const struct net_device *in,
						     const struct net_device *out,
						     struct nf_conn *ct));

#endif /* _NF_NAT_L3PROTO_H */
+34 −199
Original line number Diff line number Diff line
@@ -30,222 +30,57 @@ static const struct xt_table nf_nat_ipv6_table = {
	.af		= NFPROTO_IPV6,
};

static unsigned int alloc_null_binding(struct nf_conn *ct, unsigned int hooknum)
{
	/* Force range to this IP; let proto decide mapping for
	 * per-proto parts (hence not IP_NAT_RANGE_PROTO_SPECIFIED).
	 */
	struct nf_nat_range range;

	range.flags = 0;
	pr_debug("Allocating NULL binding for %p (%pI6)\n", ct,
		 HOOK2MANIP(hooknum) == NF_NAT_MANIP_SRC ?
		 &ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip6 :
		 &ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip6);

	return nf_nat_setup_info(ct, &range, HOOK2MANIP(hooknum));
}

static unsigned int nf_nat_rule_find(struct sk_buff *skb, unsigned int hooknum,
static unsigned int ip6table_nat_do_chain(const struct nf_hook_ops *ops,
					  struct sk_buff *skb,
					  const struct net_device *in,
					  const struct net_device *out,
					  struct nf_conn *ct)
{
	struct net *net = nf_ct_net(ct);
	unsigned int ret;

	ret = ip6t_do_table(skb, hooknum, in, out, net->ipv6.ip6table_nat);
	if (ret == NF_ACCEPT) {
		if (!nf_nat_initialized(ct, HOOK2MANIP(hooknum)))
			ret = alloc_null_binding(ct, hooknum);
	}
	return ret;
	return ip6t_do_table(skb, ops->hooknum, in, out, net->ipv6.ip6table_nat);
}

static unsigned int
nf_nat_ipv6_fn(const struct nf_hook_ops *ops,
static unsigned int ip6table_nat_fn(const struct nf_hook_ops *ops,
				    struct sk_buff *skb,
				    const struct net_device *in,
				    const struct net_device *out,
				    int (*okfn)(struct sk_buff *))
{
	struct nf_conn *ct;
	enum ip_conntrack_info ctinfo;
	struct nf_conn_nat *nat;
	enum nf_nat_manip_type maniptype = HOOK2MANIP(ops->hooknum);
	__be16 frag_off;
	int hdrlen;
	u8 nexthdr;

	ct = nf_ct_get(skb, &ctinfo);
	/* Can't track?  It's not due to stress, or conntrack would
	 * have dropped it.  Hence it's the user's responsibilty to
	 * packet filter it out, or implement conntrack/NAT for that
	 * protocol. 8) --RR
	 */
	if (!ct)
		return NF_ACCEPT;

	/* Don't try to NAT if this packet is not conntracked */
	if (nf_ct_is_untracked(ct))
		return NF_ACCEPT;

	nat = nf_ct_nat_ext_add(ct);
	if (nat == NULL)
		return NF_ACCEPT;

	switch (ctinfo) {
	case IP_CT_RELATED:
	case IP_CT_RELATED_REPLY:
		nexthdr = ipv6_hdr(skb)->nexthdr;
		hdrlen = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr),
					  &nexthdr, &frag_off);

		if (hdrlen >= 0 && nexthdr == IPPROTO_ICMPV6) {
			if (!nf_nat_icmpv6_reply_translation(skb, ct, ctinfo,
							     ops->hooknum,
							     hdrlen))
				return NF_DROP;
			else
				return NF_ACCEPT;
		}
		/* Fall thru... (Only ICMPs can be IP_CT_IS_REPLY) */
	case IP_CT_NEW:
		/* Seen it before?  This can happen for loopback, retrans,
		 * or local packets.
		 */
		if (!nf_nat_initialized(ct, maniptype)) {
			unsigned int ret;

			ret = nf_nat_rule_find(skb, ops->hooknum, in, out, ct);
			if (ret != NF_ACCEPT)
				return ret;
		} else {
			pr_debug("Already setup manip %s for ct %p\n",
				 maniptype == NF_NAT_MANIP_SRC ? "SRC" : "DST",
				 ct);
			if (nf_nat_oif_changed(ops->hooknum, ctinfo, nat, out))
				goto oif_changed;
		}
		break;

	default:
		/* ESTABLISHED */
		NF_CT_ASSERT(ctinfo == IP_CT_ESTABLISHED ||
			     ctinfo == IP_CT_ESTABLISHED_REPLY);
		if (nf_nat_oif_changed(ops->hooknum, ctinfo, nat, out))
			goto oif_changed;
	}

	return nf_nat_packet(ct, ctinfo, ops->hooknum, skb);

oif_changed:
	nf_ct_kill_acct(ct, ctinfo, skb);
	return NF_DROP;
	return nf_nat_ipv6_fn(ops, skb, in, out, ip6table_nat_do_chain);
}

static unsigned int
nf_nat_ipv6_in(const struct nf_hook_ops *ops,
static unsigned int ip6table_nat_in(const struct nf_hook_ops *ops,
				    struct sk_buff *skb,
				    const struct net_device *in,
				    const struct net_device *out,
				    int (*okfn)(struct sk_buff *))
{
	unsigned int ret;
	struct in6_addr daddr = ipv6_hdr(skb)->daddr;

	ret = nf_nat_ipv6_fn(ops, skb, in, out, okfn);
	if (ret != NF_DROP && ret != NF_STOLEN &&
	    ipv6_addr_cmp(&daddr, &ipv6_hdr(skb)->daddr))
		skb_dst_drop(skb);

	return ret;
	return nf_nat_ipv6_in(ops, skb, in, out, ip6table_nat_do_chain);
}

static unsigned int
nf_nat_ipv6_out(const struct nf_hook_ops *ops,
static unsigned int ip6table_nat_out(const struct nf_hook_ops *ops,
				     struct sk_buff *skb,
				     const struct net_device *in,
				     const struct net_device *out,
				     int (*okfn)(struct sk_buff *))
{
#ifdef CONFIG_XFRM
	const struct nf_conn *ct;
	enum ip_conntrack_info ctinfo;
	int err;
#endif
	unsigned int ret;

	/* root is playing with raw sockets. */
	if (skb->len < sizeof(struct ipv6hdr))
		return NF_ACCEPT;

	ret = nf_nat_ipv6_fn(ops, skb, in, out, okfn);
#ifdef CONFIG_XFRM
	if (ret != NF_DROP && ret != NF_STOLEN &&
	    !(IP6CB(skb)->flags & IP6SKB_XFRM_TRANSFORMED) &&
	    (ct = nf_ct_get(skb, &ctinfo)) != NULL) {
		enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);

		if (!nf_inet_addr_cmp(&ct->tuplehash[dir].tuple.src.u3,
				      &ct->tuplehash[!dir].tuple.dst.u3) ||
		    (ct->tuplehash[dir].tuple.dst.protonum != IPPROTO_ICMPV6 &&
		     ct->tuplehash[dir].tuple.src.u.all !=
		     ct->tuplehash[!dir].tuple.dst.u.all)) {
			err = nf_xfrm_me_harder(skb, AF_INET6);
			if (err < 0)
				ret = NF_DROP_ERR(err);
		}
	}
#endif
	return ret;
	return nf_nat_ipv6_out(ops, skb, in, out, ip6table_nat_do_chain);
}

static unsigned int
nf_nat_ipv6_local_fn(const struct nf_hook_ops *ops,
static unsigned int ip6table_nat_local_fn(const struct nf_hook_ops *ops,
					  struct sk_buff *skb,
					  const struct net_device *in,
					  const struct net_device *out,
					  int (*okfn)(struct sk_buff *))
{
	const struct nf_conn *ct;
	enum ip_conntrack_info ctinfo;
	unsigned int ret;
	int err;

	/* root is playing with raw sockets. */
	if (skb->len < sizeof(struct ipv6hdr))
		return NF_ACCEPT;

	ret = nf_nat_ipv6_fn(ops, skb, in, out, okfn);
	if (ret != NF_DROP && ret != NF_STOLEN &&
	    (ct = nf_ct_get(skb, &ctinfo)) != NULL) {
		enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);

		if (!nf_inet_addr_cmp(&ct->tuplehash[dir].tuple.dst.u3,
				      &ct->tuplehash[!dir].tuple.src.u3)) {
			err = ip6_route_me_harder(skb);
			if (err < 0)
				ret = NF_DROP_ERR(err);
		}
#ifdef CONFIG_XFRM
		else if (!(IP6CB(skb)->flags & IP6SKB_XFRM_TRANSFORMED) &&
			 ct->tuplehash[dir].tuple.dst.protonum != IPPROTO_ICMPV6 &&
			 ct->tuplehash[dir].tuple.dst.u.all !=
			 ct->tuplehash[!dir].tuple.src.u.all) {
			err = nf_xfrm_me_harder(skb, AF_INET6);
			if (err < 0)
				ret = NF_DROP_ERR(err);
		}
#endif
	}
	return ret;
	return nf_nat_ipv6_local_fn(ops, skb, in, out, ip6table_nat_do_chain);
}

static struct nf_hook_ops nf_nat_ipv6_ops[] __read_mostly = {
	/* Before packet filtering, change destination */
	{
		.hook		= nf_nat_ipv6_in,
		.hook		= ip6table_nat_in,
		.owner		= THIS_MODULE,
		.pf		= NFPROTO_IPV6,
		.hooknum	= NF_INET_PRE_ROUTING,
@@ -253,7 +88,7 @@ static struct nf_hook_ops nf_nat_ipv6_ops[] __read_mostly = {
	},
	/* After packet filtering, change source */
	{
		.hook		= nf_nat_ipv6_out,
		.hook		= ip6table_nat_out,
		.owner		= THIS_MODULE,
		.pf		= NFPROTO_IPV6,
		.hooknum	= NF_INET_POST_ROUTING,
@@ -261,7 +96,7 @@ static struct nf_hook_ops nf_nat_ipv6_ops[] __read_mostly = {
	},
	/* Before packet filtering, change destination */
	{
		.hook		= nf_nat_ipv6_local_fn,
		.hook		= ip6table_nat_local_fn,
		.owner		= THIS_MODULE,
		.pf		= NFPROTO_IPV6,
		.hooknum	= NF_INET_LOCAL_OUT,
@@ -269,7 +104,7 @@ static struct nf_hook_ops nf_nat_ipv6_ops[] __read_mostly = {
	},
	/* After packet filtering, change source */
	{
		.hook		= nf_nat_ipv6_fn,
		.hook		= ip6table_nat_fn,
		.owner		= THIS_MODULE,
		.pf		= NFPROTO_IPV6,
		.hooknum	= NF_INET_LOCAL_IN,
+199 −0
Original line number Diff line number Diff line
@@ -261,6 +261,205 @@ int nf_nat_icmpv6_reply_translation(struct sk_buff *skb,
}
EXPORT_SYMBOL_GPL(nf_nat_icmpv6_reply_translation);

unsigned int
nf_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
	       const struct net_device *in, const struct net_device *out,
	       unsigned int (*do_chain)(const struct nf_hook_ops *ops,
					struct sk_buff *skb,
					const struct net_device *in,
					const struct net_device *out,
					struct nf_conn *ct))
{
	struct nf_conn *ct;
	enum ip_conntrack_info ctinfo;
	struct nf_conn_nat *nat;
	enum nf_nat_manip_type maniptype = HOOK2MANIP(ops->hooknum);
	__be16 frag_off;
	int hdrlen;
	u8 nexthdr;

	ct = nf_ct_get(skb, &ctinfo);
	/* Can't track?  It's not due to stress, or conntrack would
	 * have dropped it.  Hence it's the user's responsibilty to
	 * packet filter it out, or implement conntrack/NAT for that
	 * protocol. 8) --RR
	 */
	if (!ct)
		return NF_ACCEPT;

	/* Don't try to NAT if this packet is not conntracked */
	if (nf_ct_is_untracked(ct))
		return NF_ACCEPT;

	nat = nf_ct_nat_ext_add(ct);
	if (nat == NULL)
		return NF_ACCEPT;

	switch (ctinfo) {
	case IP_CT_RELATED:
	case IP_CT_RELATED_REPLY:
		nexthdr = ipv6_hdr(skb)->nexthdr;
		hdrlen = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr),
					  &nexthdr, &frag_off);

		if (hdrlen >= 0 && nexthdr == IPPROTO_ICMPV6) {
			if (!nf_nat_icmpv6_reply_translation(skb, ct, ctinfo,
							     ops->hooknum,
							     hdrlen))
				return NF_DROP;
			else
				return NF_ACCEPT;
		}
		/* Fall thru... (Only ICMPs can be IP_CT_IS_REPLY) */
	case IP_CT_NEW:
		/* Seen it before?  This can happen for loopback, retrans,
		 * or local packets.
		 */
		if (!nf_nat_initialized(ct, maniptype)) {
			unsigned int ret;

			ret = do_chain(ops, skb, in, out, ct);
			if (ret != NF_ACCEPT)
				return ret;

			if (nf_nat_initialized(ct, HOOK2MANIP(ops->hooknum)))
				break;

			ret = nf_nat_alloc_null_binding(ct, ops->hooknum);
			if (ret != NF_ACCEPT)
				return ret;
		} else {
			pr_debug("Already setup manip %s for ct %p\n",
				 maniptype == NF_NAT_MANIP_SRC ? "SRC" : "DST",
				 ct);
			if (nf_nat_oif_changed(ops->hooknum, ctinfo, nat, out))
				goto oif_changed;
		}
		break;

	default:
		/* ESTABLISHED */
		NF_CT_ASSERT(ctinfo == IP_CT_ESTABLISHED ||
			     ctinfo == IP_CT_ESTABLISHED_REPLY);
		if (nf_nat_oif_changed(ops->hooknum, ctinfo, nat, out))
			goto oif_changed;
	}

	return nf_nat_packet(ct, ctinfo, ops->hooknum, skb);

oif_changed:
	nf_ct_kill_acct(ct, ctinfo, skb);
	return NF_DROP;
}
EXPORT_SYMBOL_GPL(nf_nat_ipv6_fn);

unsigned int
nf_nat_ipv6_in(const struct nf_hook_ops *ops, struct sk_buff *skb,
	       const struct net_device *in, const struct net_device *out,
	       unsigned int (*do_chain)(const struct nf_hook_ops *ops,
					struct sk_buff *skb,
					const struct net_device *in,
					const struct net_device *out,
					struct nf_conn *ct))
{
	unsigned int ret;
	struct in6_addr daddr = ipv6_hdr(skb)->daddr;

	ret = nf_nat_ipv6_fn(ops, skb, in, out, do_chain);
	if (ret != NF_DROP && ret != NF_STOLEN &&
	    ipv6_addr_cmp(&daddr, &ipv6_hdr(skb)->daddr))
		skb_dst_drop(skb);

	return ret;
}
EXPORT_SYMBOL_GPL(nf_nat_ipv6_in);

unsigned int
nf_nat_ipv6_out(const struct nf_hook_ops *ops, struct sk_buff *skb,
		const struct net_device *in, const struct net_device *out,
		unsigned int (*do_chain)(const struct nf_hook_ops *ops,
					 struct sk_buff *skb,
					 const struct net_device *in,
					 const struct net_device *out,
					 struct nf_conn *ct))
{
#ifdef CONFIG_XFRM
	const struct nf_conn *ct;
	enum ip_conntrack_info ctinfo;
	int err;
#endif
	unsigned int ret;

	/* root is playing with raw sockets. */
	if (skb->len < sizeof(struct ipv6hdr))
		return NF_ACCEPT;

	ret = nf_nat_ipv6_fn(ops, skb, in, out, do_chain);
#ifdef CONFIG_XFRM
	if (ret != NF_DROP && ret != NF_STOLEN &&
	    !(IP6CB(skb)->flags & IP6SKB_XFRM_TRANSFORMED) &&
	    (ct = nf_ct_get(skb, &ctinfo)) != NULL) {
		enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);

		if (!nf_inet_addr_cmp(&ct->tuplehash[dir].tuple.src.u3,
				      &ct->tuplehash[!dir].tuple.dst.u3) ||
		    (ct->tuplehash[dir].tuple.dst.protonum != IPPROTO_ICMPV6 &&
		     ct->tuplehash[dir].tuple.src.u.all !=
		     ct->tuplehash[!dir].tuple.dst.u.all)) {
			err = nf_xfrm_me_harder(skb, AF_INET6);
			if (err < 0)
				ret = NF_DROP_ERR(err);
		}
	}
#endif
	return ret;
}
EXPORT_SYMBOL_GPL(nf_nat_ipv6_out);

unsigned int
nf_nat_ipv6_local_fn(const struct nf_hook_ops *ops, struct sk_buff *skb,
		     const struct net_device *in, const struct net_device *out,
		     unsigned int (*do_chain)(const struct nf_hook_ops *ops,
					      struct sk_buff *skb,
					      const struct net_device *in,
					      const struct net_device *out,
					      struct nf_conn *ct))
{
	const struct nf_conn *ct;
	enum ip_conntrack_info ctinfo;
	unsigned int ret;
	int err;

	/* root is playing with raw sockets. */
	if (skb->len < sizeof(struct ipv6hdr))
		return NF_ACCEPT;

	ret = nf_nat_ipv6_fn(ops, skb, in, out, do_chain);
	if (ret != NF_DROP && ret != NF_STOLEN &&
	    (ct = nf_ct_get(skb, &ctinfo)) != NULL) {
		enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);

		if (!nf_inet_addr_cmp(&ct->tuplehash[dir].tuple.dst.u3,
				      &ct->tuplehash[!dir].tuple.src.u3)) {
			err = ip6_route_me_harder(skb);
			if (err < 0)
				ret = NF_DROP_ERR(err);
		}
#ifdef CONFIG_XFRM
		else if (!(IP6CB(skb)->flags & IP6SKB_XFRM_TRANSFORMED) &&
			 ct->tuplehash[dir].tuple.dst.protonum != IPPROTO_ICMPV6 &&
			 ct->tuplehash[dir].tuple.dst.u.all !=
			 ct->tuplehash[!dir].tuple.src.u.all) {
			err = nf_xfrm_me_harder(skb, AF_INET6);
			if (err < 0)
				ret = NF_DROP_ERR(err);
		}
#endif
	}
	return ret;
}
EXPORT_SYMBOL_GPL(nf_nat_ipv6_local_fn);

static int __init nf_nat_l3proto_ipv6_init(void)
{
	int err;
+5 −5
Original line number Diff line number Diff line
@@ -28,7 +28,7 @@
 * IPv6 NAT chains
 */

static unsigned int nf_nat_ipv6_fn(const struct nf_hook_ops *ops,
static unsigned int nft_nat_ipv6_fn(const struct nf_hook_ops *ops,
			      struct sk_buff *skb,
			      const struct net_device *in,
			      const struct net_device *out,
@@ -97,7 +97,7 @@ static unsigned int nf_nat_ipv6_prerouting(const struct nf_hook_ops *ops,
	struct in6_addr daddr = ipv6_hdr(skb)->daddr;
	unsigned int ret;

	ret = nf_nat_ipv6_fn(ops, skb, in, out, okfn);
	ret = nft_nat_ipv6_fn(ops, skb, in, out, okfn);
	if (ret != NF_DROP && ret != NF_STOLEN &&
	    ipv6_addr_cmp(&daddr, &ipv6_hdr(skb)->daddr))
		skb_dst_drop(skb);
@@ -115,7 +115,7 @@ static unsigned int nf_nat_ipv6_postrouting(const struct nf_hook_ops *ops,
	const struct nf_conn *ct __maybe_unused;
	unsigned int ret;

	ret = nf_nat_ipv6_fn(ops, skb, in, out, okfn);
	ret = nft_nat_ipv6_fn(ops, skb, in, out, okfn);
#ifdef CONFIG_XFRM
	if (ret != NF_DROP && ret != NF_STOLEN &&
	    !(IP6CB(skb)->flags & IP6SKB_XFRM_TRANSFORMED) &&
@@ -143,7 +143,7 @@ static unsigned int nf_nat_ipv6_output(const struct nf_hook_ops *ops,
	const struct nf_conn *ct;
	unsigned int ret;

	ret = nf_nat_ipv6_fn(ops, skb, in, out, okfn);
	ret = nft_nat_ipv6_fn(ops, skb, in, out, okfn);
	if (ret != NF_DROP && ret != NF_STOLEN &&
	    (ct = nf_ct_get(skb, &ctinfo)) != NULL) {
		enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
@@ -177,7 +177,7 @@ static const struct nf_chain_type nft_chain_nat_ipv6 = {
		[NF_INET_PRE_ROUTING]	= nf_nat_ipv6_prerouting,
		[NF_INET_POST_ROUTING]	= nf_nat_ipv6_postrouting,
		[NF_INET_LOCAL_OUT]	= nf_nat_ipv6_output,
		[NF_INET_LOCAL_IN]	= nf_nat_ipv6_fn,
		[NF_INET_LOCAL_IN]	= nft_nat_ipv6_fn,
	},
};