xfrm: make local error reporting more robust

In xfrm4 and xfrm6 we need to take care about sockets of the other
address family. This could happen because a 6in4 or 4in6 tunnel could
get protected by ipsec.

Because we don't want to have a run-time dependency on ipv6 when only
using ipv4 xfrm we have to embed a pointer to the correct local_error
function in xfrm_state_afinet and look it up when returning an error
depending on the socket address family.

Thanks to vi0ss for the great bug report:
<https://bugzilla.kernel.org/show_bug.cgi?id=58691>

v2:
a) fix two more unsafe interpretations of skb->sk as ipv6 socket
   (xfrm6_local_dontfrag and __xfrm6_output)
v3:
a) add an EXPORT_SYMBOL_GPL(xfrm_local_error) to fix a link error when
   building ipv6 as a module (thanks to Steffen Klassert)

Reported-by: <vi0oss@gmail.com>
Cc: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>

include/net/xfrm.h

@@ -341,10 +341,13 @@ struct xfrm_state_afinfo {
 						  struct sk_buff *skb);
 	int			(*transport_finish)(struct sk_buff *skb,
 						    int async);
+	void			(*local_error)(struct sk_buff *skb, u32 mtu);
 };
 
 extern int xfrm_state_register_afinfo(struct xfrm_state_afinfo *afinfo);
 extern int xfrm_state_unregister_afinfo(struct xfrm_state_afinfo *afinfo);
+extern struct xfrm_state_afinfo *xfrm_state_get_afinfo(unsigned int family);
+extern void xfrm_state_put_afinfo(struct xfrm_state_afinfo *afinfo);
 
 extern void xfrm_state_delete_tunnel(struct xfrm_state *x);
 
@@ -1477,6 +1480,7 @@ extern int xfrm_input_resume(struct sk_buff *skb, int nexthdr);
 extern int xfrm_output_resume(struct sk_buff *skb, int err);
 extern int xfrm_output(struct sk_buff *skb);
 extern int xfrm_inner_extract_output(struct xfrm_state *x, struct sk_buff *skb);
+extern void xfrm_local_error(struct sk_buff *skb, int mtu);
 extern int xfrm4_extract_header(struct sk_buff *skb);
 extern int xfrm4_extract_input(struct xfrm_state *x, struct sk_buff *skb);
 extern int xfrm4_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi,
@@ -1497,6 +1501,7 @@ extern int xfrm4_tunnel_register(struct xfrm_tunnel *handler, unsigned short fam
 extern int xfrm4_tunnel_deregister(struct xfrm_tunnel *handler, unsigned short family);
 extern int xfrm4_mode_tunnel_input_register(struct xfrm_tunnel *handler);
 extern int xfrm4_mode_tunnel_input_deregister(struct xfrm_tunnel *handler);
+extern void xfrm4_local_error(struct sk_buff *skb, u32 mtu);
 extern int xfrm6_extract_header(struct sk_buff *skb);
 extern int xfrm6_extract_input(struct xfrm_state *x, struct sk_buff *skb);
 extern int xfrm6_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi);
@@ -1514,6 +1519,7 @@ extern int xfrm6_output(struct sk_buff *skb);
 extern int xfrm6_output_finish(struct sk_buff *skb);
 extern int xfrm6_find_1stfragopt(struct xfrm_state *x, struct sk_buff *skb,
 				 u8 **prevhdr);
+extern void xfrm6_local_error(struct sk_buff *skb, u32 mtu);
 
 #ifdef CONFIG_XFRM
 extern int xfrm4_udp_encap_rcv(struct sock *sk, struct sk_buff *skb);

net/ipv4/xfrm4_output.c

@@ -33,8 +33,7 @@ static int xfrm4_tunnel_check_size(struct sk_buff *skb)
 	mtu = dst_mtu(dst);
 	if (skb->len > mtu) {
 		if (skb->sk)
-			ip_local_error(skb->sk, EMSGSIZE, ip_hdr(skb)->daddr,
+			xfrm_local_error(skb, mtu);
-				       inet_sk(skb->sk)->inet_dport, mtu);
 		else
 			icmp_send(skb, ICMP_DEST_UNREACH,
 				  ICMP_FRAG_NEEDED, htonl(mtu));
@@ -99,3 +98,12 @@ int xfrm4_output(struct sk_buff *skb)
 			    x->outer_mode->afinfo->output_finish,
 			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }
+
+void xfrm4_local_error(struct sk_buff *skb, u32 mtu)
+{
+	struct iphdr *hdr;
+
+	hdr = skb->encapsulation ? inner_ip_hdr(skb) : ip_hdr(skb);
+	ip_local_error(skb->sk, EMSGSIZE, hdr->daddr,
+		       inet_sk(skb->sk)->inet_dport, mtu);
+}

net/ipv4/xfrm4_state.c

@@ -83,6 +83,7 @@ static struct xfrm_state_afinfo xfrm4_state_afinfo = {
 	.extract_input		= xfrm4_extract_input,
 	.extract_output		= xfrm4_extract_output,
 	.transport_finish	= xfrm4_transport_finish,
+	.local_error		= xfrm4_local_error,
 };
 
 void __init xfrm4_state_init(void)

net/ipv6/xfrm6_output.c

@@ -34,8 +34,10 @@ static int xfrm6_local_dontfrag(struct sk_buff *skb)
 	struct sock *sk = skb->sk;
 
 	if (sk) {
-		proto = sk->sk_protocol;
+		if (sk->sk_family != AF_INET6)
+			return 0;
 
+		proto = sk->sk_protocol;
 		if (proto == IPPROTO_UDP || proto == IPPROTO_RAW)
 			return inet6_sk(sk)->dontfrag;
 	}
@@ -54,7 +56,7 @@ static void xfrm6_local_rxpmtu(struct sk_buff *skb, u32 mtu)
 	ipv6_local_rxpmtu(sk, &fl6, mtu);
 }
 
-static void xfrm6_local_error(struct sk_buff *skb, u32 mtu)
+void xfrm6_local_error(struct sk_buff *skb, u32 mtu)
 {
 	struct flowi6 fl6;
 	struct sock *sk = skb->sk;
@@ -80,7 +82,7 @@ static int xfrm6_tunnel_check_size(struct sk_buff *skb)
 		if (xfrm6_local_dontfrag(skb))
 			xfrm6_local_rxpmtu(skb, mtu);
 		else if (skb->sk)
-			xfrm6_local_error(skb, mtu);
+			xfrm_local_error(skb, mtu);
 		else
 			icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu);
 		ret = -EMSGSIZE;
@@ -142,7 +144,7 @@ static int __xfrm6_output(struct sk_buff *skb)
 		xfrm6_local_rxpmtu(skb, mtu);
 		return -EMSGSIZE;
 	} else if (!skb->local_df && skb->len > mtu && skb->sk) {
-		xfrm6_local_error(skb, mtu);
+		xfrm_local_error(skb, mtu);
 		return -EMSGSIZE;
 	}
 

net/ipv6/xfrm6_state.c

@@ -183,6 +183,7 @@ static struct xfrm_state_afinfo xfrm6_state_afinfo = {
 	.extract_input		= xfrm6_extract_input,
 	.extract_output		= xfrm6_extract_output,
 	.transport_finish	= xfrm6_transport_finish,
+	.local_error		= xfrm6_local_error,
 };
 
 int __init xfrm6_state_init(void)

net/xfrm/xfrm_output.c

@@ -214,5 +214,18 @@ int xfrm_inner_extract_output(struct xfrm_state *x, struct sk_buff *skb)
 	return inner_mode->afinfo->extract_output(x, skb);
 }
 
+void xfrm_local_error(struct sk_buff *skb, int mtu)
+{
+	struct xfrm_state_afinfo *afinfo;
+
+	afinfo = xfrm_state_get_afinfo(skb->sk->sk_family);
+	if (!afinfo)
+		return;
+
+	afinfo->local_error(skb, mtu);
+	xfrm_state_put_afinfo(afinfo);
+}
+
 EXPORT_SYMBOL_GPL(xfrm_output);
 EXPORT_SYMBOL_GPL(xfrm_inner_extract_output);
+EXPORT_SYMBOL_GPL(xfrm_local_error);

net/xfrm/xfrm_state.c

@@ -39,9 +39,6 @@ static DEFINE_SPINLOCK(xfrm_state_lock);
 
 static unsigned int xfrm_state_hashmax __read_mostly = 1 * 1024 * 1024;
 
-static struct xfrm_state_afinfo *xfrm_state_get_afinfo(unsigned int family);
-static void xfrm_state_put_afinfo(struct xfrm_state_afinfo *afinfo);
-
 static inline unsigned int xfrm_dst_hash(struct net *net,
 					 const xfrm_address_t *daddr,
 					 const xfrm_address_t *saddr,
@@ -1860,7 +1857,7 @@ int xfrm_state_unregister_afinfo(struct xfrm_state_afinfo *afinfo)
 }
 EXPORT_SYMBOL(xfrm_state_unregister_afinfo);
 
-static struct xfrm_state_afinfo *xfrm_state_get_afinfo(unsigned int family)
+struct xfrm_state_afinfo *xfrm_state_get_afinfo(unsigned int family)
 {
 	struct xfrm_state_afinfo *afinfo;
 	if (unlikely(family >= NPROTO))
@@ -1872,7 +1869,7 @@ static struct xfrm_state_afinfo *xfrm_state_get_afinfo(unsigned int family)
 	return afinfo;
 }
 
-static void xfrm_state_put_afinfo(struct xfrm_state_afinfo *afinfo)
+void xfrm_state_put_afinfo(struct xfrm_state_afinfo *afinfo)
 {
 	rcu_read_unlock();
 }