net: Compute protocol sequence numbers and fragment IDs using MD5.

Computers have become a lot faster since we compromised on the
partial MD4 hash which we use currently for performance reasons.

MD5 is a much safer choice, and is inline with both RFC1948 and
other ISS generators (OpenBSD, Solaris, etc.)

Furthermore, only having 24-bits of the sequence number be truly
unpredictable is a very serious limitation.  So the periodic
regeneration and 8-bit counter have been removed.  We compute and
use a full 32-bit sequence number.

For ipv6, DCCP was found to use a 32-bit truncated initial sequence
number (it needs 43-bits) and that is fixed here as well.

Reported-by: Dan Kaminsky <dan@doxpara.com>
Tested-by: Willy Tarreau <w@1wt.eu>
Signed-off-by: David S. Miller <davem@davemloft.net>

11 files changed, 195 insertions, 8 deletions

diff --git a/net/core/Makefile b/net/core/Makefile
index 8a04dd22cf7..0d357b1c4e5 100644
--- a/net/core/Makefile
+++ b/net/core/Makefile
@@ -3,7 +3,7 @@
 #
 
 obj-y := sock.o request_sock.o skbuff.o iovec.o datagram.o stream.o scm.o \
-	 gen_stats.o gen_estimator.o net_namespace.o
+	 gen_stats.o gen_estimator.o net_namespace.o secure_seq.o
 
 obj-$(CONFIG_SYSCTL) += sysctl_net_core.o
 
diff --git a/net/core/secure_seq.c b/net/core/secure_seq.c
new file mode 100644
index 00000000000..45329d7c9dd
--- /dev/null
+++ b/net/core/secure_seq.c
@@ -0,0 +1,184 @@
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/cryptohash.h>
+#include <linux/module.h>
+#include <linux/cache.h>
+#include <linux/random.h>
+#include <linux/hrtimer.h>
+#include <linux/ktime.h>
+#include <linux/string.h>
+
+#include <net/secure_seq.h>
+
+static u32 net_secret[MD5_MESSAGE_BYTES / 4] ____cacheline_aligned;
+
+static int __init net_secret_init(void)
+{
+	get_random_bytes(net_secret, sizeof(net_secret));
+	return 0;
+}
+late_initcall(net_secret_init);
+
+static u32 seq_scale(u32 seq)
+{
+	/*
+	 *	As close as possible to RFC 793, which
+	 *	suggests using a 250 kHz clock.
+	 *	Further reading shows this assumes 2 Mb/s networks.
+	 *	For 10 Mb/s Ethernet, a 1 MHz clock is appropriate.
+	 *	For 10 Gb/s Ethernet, a 1 GHz clock should be ok, but
+	 *	we also need to limit the resolution so that the u32 seq
+	 *	overlaps less than one time per MSL (2 minutes).
+	 *	Choosing a clock of 64 ns period is OK. (period of 274 s)
+	 */
+	return seq + (ktime_to_ns(ktime_get_real()) >> 6);
+}
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+__u32 secure_tcpv6_sequence_number(__be32 *saddr, __be32 *daddr,
+				   __be16 sport, __be16 dport)
+{
+	u32 secret[MD5_MESSAGE_BYTES / 4];
+	u32 hash[MD5_DIGEST_WORDS];
+	u32 i;
+
+	memcpy(hash, saddr, 16);
+	for (i = 0; i < 4; i++)
+		secret[i] = net_secret[i] + daddr[i];
+	secret[4] = net_secret[4] +
+		(((__force u16)sport << 16) + (__force u16)dport);
+	for (i = 5; i < MD5_MESSAGE_BYTES / 4; i++)
+		secret[i] = net_secret[i];
+
+	md5_transform(hash, secret);
+
+	return seq_scale(hash[0]);
+}
+EXPORT_SYMBOL(secure_tcpv6_sequence_number);
+
+u32 secure_ipv6_port_ephemeral(const __be32 *saddr, const __be32 *daddr,
+			       __be16 dport)
+{
+	u32 secret[MD5_MESSAGE_BYTES / 4];
+	u32 hash[MD5_DIGEST_WORDS];
+	u32 i;
+
+	memcpy(hash, saddr, 16);
+	for (i = 0; i < 4; i++)
+		secret[i] = net_secret[i] + (__force u32) daddr[i];
+	secret[4] = net_secret[4] + (__force u32)dport;
+	for (i = 5; i < MD5_MESSAGE_BYTES / 4; i++)
+		secret[i] = net_secret[i];
+
+	md5_transform(hash, secret);
+
+	return hash[0];
+}
+#endif
+
+#ifdef CONFIG_INET
+__u32 secure_ip_id(__be32 daddr)
+{
+	u32 hash[MD5_DIGEST_WORDS];
+
+	hash[0] = (__force __u32) daddr;
+	hash[1] = net_secret[13];
+	hash[2] = net_secret[14];
+	hash[3] = net_secret[15];
+
+	md5_transform(hash, net_secret);
+
+	return hash[0];
+}
+
+__u32 secure_ipv6_id(const __be32 daddr[4])
+{
+	__u32 hash[4];
+
+	memcpy(hash, daddr, 16);
+	md5_transform(hash, net_secret);
+
+	return hash[0];
+}
+
+__u32 secure_tcp_sequence_number(__be32 saddr, __be32 daddr,
+				 __be16 sport, __be16 dport)
+{
+	u32 hash[MD5_DIGEST_WORDS];
+
+	hash[0] = (__force u32)saddr;
+	hash[1] = (__force u32)daddr;
+	hash[2] = ((__force u16)sport << 16) + (__force u16)dport;
+	hash[3] = net_secret[15];
+
+	md5_transform(hash, net_secret);
+
+	return seq_scale(hash[0]);
+}
+
+u32 secure_ipv4_port_ephemeral(__be32 saddr, __be32 daddr, __be16 dport)
+{
+	u32 hash[MD5_DIGEST_WORDS];
+
+	hash[0] = (__force u32)saddr;
+	hash[1] = (__force u32)daddr;
+	hash[2] = (__force u32)dport ^ net_secret[14];
+	hash[3] = net_secret[15];
+
+	md5_transform(hash, net_secret);
+
+	return hash[0];
+}
+EXPORT_SYMBOL_GPL(secure_ipv4_port_ephemeral);
+#endif
+
+#if defined(CONFIG_IP_DCCP) || defined(CONFIG_IP_DCCP_MODULE)
+u64 secure_dccp_sequence_number(__be32 saddr, __be32 daddr,
+				__be16 sport, __be16 dport)
+{
+	u32 hash[MD5_DIGEST_WORDS];
+	u64 seq;
+
+	hash[0] = (__force u32)saddr;
+	hash[1] = (__force u32)daddr;
+	hash[2] = ((__force u16)sport << 16) + (__force u16)dport;
+	hash[3] = net_secret[15];
+
+	md5_transform(hash, net_secret);
+
+	seq = hash[0] | (((u64)hash[1]) << 32);
+	seq += ktime_to_ns(ktime_get_real());
+	seq &= (1ull << 48) - 1;
+
+	return seq;
+}
+EXPORT_SYMBOL(secure_dccp_sequence_number);
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+u64 secure_dccpv6_sequence_number(__be32 *saddr, __be32 *daddr,
+				  __be16 sport, __be16 dport)
+{
+	u32 secret[MD5_MESSAGE_BYTES / 4];
+	u32 hash[MD5_DIGEST_WORDS];
+	u64 seq;
+	u32 i;
+
+	memcpy(hash, saddr, 16);
+	for (i = 0; i < 4; i++)
+		secret[i] = net_secret[i] + daddr[i];
+	secret[4] = net_secret[4] +
+		(((__force u16)sport << 16) + (__force u16)dport);
+	for (i = 5; i < MD5_MESSAGE_BYTES / 4; i++)
+		secret[i] = net_secret[i];
+
+	md5_transform(hash, secret);
+
+	seq = hash[0] | (((u64)hash[1]) << 32);
+	seq += ktime_to_ns(ktime_get_real());
+	seq &= (1ull << 48) - 1;
+
+	return seq;
+}
+EXPORT_SYMBOL(secure_dccpv6_sequence_number);
+#endif
+#endif
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 8c36adfd191..332639b56f4 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -26,6 +26,7 @@
 #include <net/timewait_sock.h>
 #include <net/tcp_states.h>
 #include <net/xfrm.h>
+#include <net/secure_seq.h>
 
 #include "ackvec.h"
 #include "ccid.h"
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 8dc4348774a..b74f76117dc 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -29,6 +29,7 @@
 #include <net/transp_v6.h>
 #include <net/ip6_checksum.h>
 #include <net/xfrm.h>
+#include <net/secure_seq.h>
 
 #include "dccp.h"
 #include "ipv6.h"
@@ -69,13 +70,7 @@ static inline void dccp_v6_send_check(struct sock *sk, struct sk_buff *skb)
 	dh->dccph_checksum = dccp_v6_csum_finish(skb, &np->saddr, &np->daddr);
 }
 
-static inline __u32 secure_dccpv6_sequence_number(__be32 *saddr, __be32 *daddr,
-						  __be16 sport, __be16 dport   )
-{
-	return secure_tcpv6_sequence_number(saddr, daddr, sport, dport);
-}
-
-static inline __u32 dccp_v6_init_sequence(struct sk_buff *skb)
+static inline __u64 dccp_v6_init_sequence(struct sk_buff *skb)
 {
 	return secure_dccpv6_sequence_number(ipv6_hdr(skb)->daddr.s6_addr32,
 					     ipv6_hdr(skb)->saddr.s6_addr32,
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 3c0369a3a66..984ec656b03 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -21,6 +21,7 @@
 
 #include <net/inet_connection_sock.h>
 #include <net/inet_hashtables.h>
+#include <net/secure_seq.h>
 #include <net/ip.h>
 
 /*
diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c
index e38213817d0..86f13c67ea8 100644
--- a/net/ipv4/inetpeer.c
+++ b/net/ipv4/inetpeer.c
@@ -19,6 +19,7 @@
 #include <linux/net.h>
 #include <net/ip.h>
 #include <net/inetpeer.h>
+#include <net/secure_seq.h>
 
 /*
  *  Theory of operations.
diff --git a/net/ipv4/netfilter/nf_nat_proto_common.c b/net/ipv4/netfilter/nf_nat_proto_common.c
index 3e61faf23a9..f52d41ea069 100644
--- a/net/ipv4/netfilter/nf_nat_proto_common.c
+++ b/net/ipv4/netfilter/nf_nat_proto_common.c
@@ -12,6 +12,7 @@
 #include <linux/ip.h>
 
 #include <linux/netfilter.h>
+#include <net/secure_seq.h>
 #include <net/netfilter/nf_nat.h>
 #include <net/netfilter/nf_nat_core.h>
 #include <net/netfilter/nf_nat_rule.h>
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 6afc4eb5059..e3dec1c9f09 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -109,6 +109,7 @@
 #include <linux/sysctl.h>
 #endif
 #include <net/atmclip.h>
+#include <net/secure_seq.h>
 
 #define RT_FL_TOS(oldflp4) \
     ((u32)(oldflp4->flowi4_tos & (IPTOS_RT_MASK | RTO_ONLINK)))
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 955b8e65b69..1c12b8ec849 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -72,6 +72,7 @@
 #include <net/timewait_sock.h>
 #include <net/xfrm.h>
 #include <net/netdma.h>
+#include <net/secure_seq.h>
 
 #include <linux/inet.h>
 #include <linux/ipv6.h>
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index b5319723370..73f1a00a96a 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -20,6 +20,7 @@
 #include <net/inet_connection_sock.h>
 #include <net/inet_hashtables.h>
 #include <net/inet6_hashtables.h>
+#include <net/secure_seq.h>
 #include <net/ip.h>
 
 int __inet6_hash(struct sock *sk, struct inet_timewait_sock *tw)
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 78aa53492b3..d1fb63f4aeb 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -61,6 +61,7 @@
 #include <net/timewait_sock.h>
 #include <net/netdma.h>
 #include <net/inet_common.h>
+#include <net/secure_seq.h>
 
 #include <asm/uaccess.h>