Merge bk://212.42.230.204/nf-2.6

into nuts.davemloft.net:/disk1/BK/net-2.6

include/linux/netdevice.h

@@ -925,6 +925,9 @@ extern unsigned long	netdev_fc_xoff;
 extern atomic_t netdev_dropping;
 extern int		netdev_set_master(struct net_device *dev, struct net_device *master);
 extern int skb_checksum_help(struct sk_buff *skb, int inward);
+/* rx skb timestamps */
+extern void		net_enable_timestamp(void);
+extern void		net_disable_timestamp(void);
 
 #ifdef CONFIG_SYSCTL
 extern char *net_sysctl_strdup(const char *s);

include/linux/netfilter.h

@@ -173,6 +173,7 @@ extern void nf_reinject(struct sk_buff *skb,
 			unsigned int verdict);
 
 extern void (*ip_ct_attach)(struct sk_buff *, struct sk_buff *);
+extern void nf_ct_attach(struct sk_buff *, struct sk_buff *);
 
 #ifdef CONFIG_NETFILTER_DEBUG
 extern void nf_dump_skb(int pf, struct sk_buff *skb);
@@ -183,6 +184,7 @@ extern void nf_invalidate_cache(int pf);
 
 #else /* !CONFIG_NETFILTER */
 #define NF_HOOK(pf, hook, skb, indev, outdev, okfn) (okfn)(skb)
+static inline void nf_ct_attach(struct sk_buff *new, struct sk_buff *skb) {}
 #endif /*CONFIG_NETFILTER*/
 
 #endif /*__KERNEL__*/

include/linux/netfilter_ipv4/ip_nat_protocol.h

@@ -18,7 +18,7 @@ struct ip_nat_protocol
 	/* Do a packet translation according to the ip_nat_proto_manip
 	 * and manip type.  Return true if succeeded. */
 	int (*manip_pkt)(struct sk_buff **pskb,
-			 unsigned int hdroff,
+			 unsigned int iphdroff,
 			 const struct ip_conntrack_manip *manip,
 			 enum ip_nat_manip_type maniptype);
 

include/net/sock.h

@@ -1270,19 +1270,7 @@ static inline void sk_eat_skb(struct sock *sk, struct sk_buff *skb)
 	__kfree_skb(skb);
 }
 
-extern atomic_t netstamp_needed;
 extern void sock_enable_timestamp(struct sock *sk);
-
-static inline void net_timestamp(struct timeval *stamp) 
-{ 
-	if (atomic_read(&netstamp_needed)) 
-		do_gettimeofday(stamp);
-	else {
-		stamp->tv_sec = 0;
-		stamp->tv_usec = 0;
-	}		
-} 
-
 extern int sock_get_timestamp(struct sock *, struct timeval __user *);
 
 /* 

net/core/dev.c

@@ -1001,6 +1001,29 @@ int call_netdevice_notifiers(unsigned long val, void *v)
 	return notifier_call_chain(&netdev_chain, val, v);
 }
 
+/* When > 0 there are consumers of rx skb time stamps */
+static atomic_t netstamp_needed = ATOMIC_INIT(0);
+
+void net_enable_timestamp(void)
+{
+	atomic_inc(&netstamp_needed);
+}
+
+void net_disable_timestamp(void)
+{
+	atomic_dec(&netstamp_needed);
+}
+
+static inline void net_timestamp(struct timeval *stamp)
+{
+	if (atomic_read(&netstamp_needed))
+		do_gettimeofday(stamp);
+	else {
+		stamp->tv_sec = 0;
+		stamp->tv_usec = 0;
+	}
+}
+
 /*
  *	Support routine. Sends outgoing frames to any network
  *	taps currently in use.
@@ -3215,6 +3238,8 @@ EXPORT_SYMBOL(skb_checksum_help);
 EXPORT_SYMBOL(synchronize_net);
 EXPORT_SYMBOL(unregister_netdevice);
 EXPORT_SYMBOL(unregister_netdevice_notifier);
+EXPORT_SYMBOL(net_enable_timestamp);
+EXPORT_SYMBOL(net_disable_timestamp);
 
 #if defined(CONFIG_BRIDGE) || defined(CONFIG_BRIDGE_MODULE)
 EXPORT_SYMBOL(br_handle_frame_hook);

net/core/netfilter.c

@@ -802,12 +802,21 @@ EXPORT_SYMBOL(nf_log_register);
 EXPORT_SYMBOL(nf_log_unregister);
 EXPORT_SYMBOL(nf_log_packet);
 
-/* This does not belong here, but ipt_REJECT needs it if connection
-   tracking in use: without this, connection may not be in hash table,
-   and hence manufactured ICMP or RST packets will not be associated
-   with it. */
+/* This does not belong here, but locally generated errors need it if connection
+   tracking in use: without this, connection may not be in hash table, and hence
+   manufactured ICMP or RST packets will not be associated with it. */
 void (*ip_ct_attach)(struct sk_buff *, struct sk_buff *);
 
+void nf_ct_attach(struct sk_buff *new, struct sk_buff *skb)
+{
+	void (*attach)(struct sk_buff *, struct sk_buff *);
+
+	if (skb->nfct && (attach = ip_ct_attach) != NULL) {
+		mb(); /* Just to be sure: must be read before executing this */
+		attach(new, skb);
+	}
+}
+
 void __init netfilter_init(void)
 {
 	int i, h;
@@ -819,6 +828,7 @@ void __init netfilter_init(void)
 }
 
 EXPORT_SYMBOL(ip_ct_attach);
+EXPORT_SYMBOL(nf_ct_attach);
 EXPORT_SYMBOL(nf_getsockopt);
 EXPORT_SYMBOL(nf_hook_slow);
 EXPORT_SYMBOL(nf_hooks);

net/core/sock.c

@@ -179,7 +179,7 @@ static void sock_disable_timestamp(struct sock *sk)
 {	
 	if (sock_flag(sk, SOCK_TIMESTAMP)) { 
 		sock_reset_flag(sk, SOCK_TIMESTAMP);
-		atomic_dec(&netstamp_needed);
+		net_disable_timestamp();
 	}
 }
 
@@ -1226,9 +1226,6 @@ void fastcall release_sock(struct sock *sk)
 }
 EXPORT_SYMBOL(release_sock);
 
-/* When > 0 there are consumers of rx skb time stamps */
-atomic_t netstamp_needed = ATOMIC_INIT(0); 
-
 int sock_get_timestamp(struct sock *sk, struct timeval __user *userstamp)
 { 
 	if (!sock_flag(sk, SOCK_TIMESTAMP))
@@ -1246,7 +1243,7 @@ void sock_enable_timestamp(struct sock *sk)
 {	
 	if (!sock_flag(sk, SOCK_TIMESTAMP)) { 
 		sock_set_flag(sk, SOCK_TIMESTAMP);
-		atomic_inc(&netstamp_needed);
+		net_enable_timestamp();
 	}
 }
 EXPORT_SYMBOL(sock_enable_timestamp); 

net/ipv4/icmp.c

@@ -338,6 +338,8 @@ int icmp_glue_bits(void *from, char *to, int offset, int len, int odd,
 				      to, len, 0);
 
 	skb->csum = csum_block_add(skb->csum, csum, odd);
+	if (icmp_pointers[icmp_param->data.icmph.type].error)
+		nf_ct_attach(skb, icmp_param->skb);
 	return 0;
 }
 

net/ipv4/netfilter/ip_nat_core.c

@@ -684,7 +684,7 @@ manip_pkt(u_int16_t proto,
 	iph = (void *)(*pskb)->data + iphdroff;
 
 	/* Manipulate protcol part. */
-	if (!ip_nat_find_proto(proto)->manip_pkt(pskb, iphdroff + iph->ihl*4,
+	if (!ip_nat_find_proto(proto)->manip_pkt(pskb, iphdroff,
 	                                         manip, maniptype))
 		return 0;
 

net/ipv4/netfilter/ip_nat_proto_icmp.c

@@ -53,11 +53,13 @@ icmp_unique_tuple(struct ip_conntrack_tuple *tuple,
 
 static int
 icmp_manip_pkt(struct sk_buff **pskb,
-	       unsigned int hdroff,
+	       unsigned int iphdroff,
 	       const struct ip_conntrack_manip *manip,
 	       enum ip_nat_manip_type maniptype)
 {
+	struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff);
 	struct icmphdr *hdr;
+	unsigned int hdroff = iphdroff + iph->ihl*4;
 
 	if (!skb_ip_make_writable(pskb, hdroff + sizeof(*hdr)))
 		return 0;

net/ipv4/netfilter/ip_nat_proto_tcp.c

@@ -84,11 +84,13 @@ tcp_unique_tuple(struct ip_conntrack_tuple *tuple,
 
 static int
 tcp_manip_pkt(struct sk_buff **pskb,
-	      unsigned int hdroff,
+	      unsigned int iphdroff,
 	      const struct ip_conntrack_manip *manip,
 	      enum ip_nat_manip_type maniptype)
 {
+	struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff);
 	struct tcphdr *hdr;
+	unsigned int hdroff = iphdroff + iph->ihl*4;
 	u_int32_t oldip;
 	u_int16_t *portptr, oldport;
 	int hdrsize = 8; /* TCP connection tracking guarantees this much */
@@ -106,11 +108,11 @@ tcp_manip_pkt(struct sk_buff **pskb,
 
 	if (maniptype == IP_NAT_MANIP_SRC) {
 		/* Get rid of src ip and src pt */
-		oldip = (*pskb)->nh.iph->saddr;
+		oldip = iph->saddr;
 		portptr = &hdr->source;
 	} else {
 		/* Get rid of dst ip and dst pt */
-		oldip = (*pskb)->nh.iph->daddr;
+		oldip = iph->daddr;
 		portptr = &hdr->dest;
 	}
 

net/ipv4/netfilter/ip_nat_proto_udp.c

@@ -83,11 +83,13 @@ udp_unique_tuple(struct ip_conntrack_tuple *tuple,
 
 static int
 udp_manip_pkt(struct sk_buff **pskb,
-	      unsigned int hdroff,
+	      unsigned int iphdroff,
 	      const struct ip_conntrack_manip *manip,
 	      enum ip_nat_manip_type maniptype)
 {
+	struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff);
 	struct udphdr *hdr;
+	unsigned int hdroff = iphdroff + iph->ihl*4;
 	u_int32_t oldip;
 	u_int16_t *portptr;
 
@@ -97,11 +99,11 @@ udp_manip_pkt(struct sk_buff **pskb,
 	hdr = (void *)(*pskb)->data + hdroff;
 	if (maniptype == IP_NAT_MANIP_SRC) {
 		/* Get rid of src ip and src pt */
-		oldip = (*pskb)->nh.iph->saddr;
+		oldip = iph->saddr;
 		portptr = &hdr->source;
 	} else {
 		/* Get rid of dst ip and dst pt */
-		oldip = (*pskb)->nh.iph->daddr;
+		oldip = iph->daddr;
 		portptr = &hdr->dest;
 	}
 	if (hdr->check) /* 0 is a special case meaning no checksum */

net/ipv4/netfilter/ip_nat_proto_unknown.c

@@ -39,7 +39,7 @@ static int unknown_unique_tuple(struct ip_conntrack_tuple *tuple,
 
 static int
 unknown_manip_pkt(struct sk_buff **pskb,
-		  unsigned int hdroff,
+		  unsigned int iphdroff,
 		  const struct ip_conntrack_manip *manip,
 		  enum ip_nat_manip_type maniptype)
 {

net/ipv4/netfilter/ip_queue.c

@@ -162,6 +162,7 @@ static inline void
 __ipq_reset(void)
 {
 	peer_pid = 0;
+	net_disable_timestamp();
 	__ipq_set_mode(IPQ_COPY_NONE, 0);
 	__ipq_flush(NF_DROP);
 }
@@ -257,7 +258,8 @@ ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp)
 	}
 	
 	if (data_len)
-		memcpy(pmsg->payload, entry->skb->data, data_len);
+		if (skb_copy_bits(entry->skb, 0, pmsg->payload, data_len))
+			BUG();
 		
 	nlh->nlmsg_len = skb->tail - old_tail;
 	return skb;
@@ -362,6 +364,8 @@ ipq_mangle_ipv4(ipq_verdict_msg_t *v, struct ipq_queue_entry *e)
 		}
 		skb_put(e->skb, diff);
 	}
+	if (!skb_ip_make_writable(&e->skb, v->data_len))
+		return -ENOMEM;
 	memcpy(e->skb->data, v->payload, v->data_len);
 	e->skb->nfcache |= NFC_ALTERED;
 
@@ -514,9 +518,10 @@ ipq_rcv_skb(struct sk_buff *skb)
 			write_unlock_bh(&queue_lock);
 			RCV_SKB_FAIL(-EBUSY);
 		}
-	}
-	else
+	} else {
+		net_enable_timestamp();
 		peer_pid = pid;
+	}
 		
 	write_unlock_bh(&queue_lock);
 	

net/ipv4/netfilter/ipt_MASQUERADE.c

@@ -118,49 +118,57 @@ masquerade_target(struct sk_buff **pskb,
 }
 
 static inline int
-device_cmp(const struct ip_conntrack *i, void *_ina)
+device_cmp(const struct ip_conntrack *i, void *ifindex)
 {
-	int ret = 0;
-	struct in_ifaddr *ina = _ina;
+	int ret;
 
 	READ_LOCK(&masq_lock);
-	/* If it's masquerading out this interface with a different address,
-	   or we don't know the new address of this interface. */
-	if (i->nat.masq_index == ina->ifa_dev->dev->ifindex
-	    && i->tuplehash[IP_CT_DIR_REPLY].tuple.dst.ip != ina->ifa_address)
-		ret = 1;
+	ret = (i->nat.masq_index == (int)(long)ifindex);
 	READ_UNLOCK(&masq_lock);
 
 	return ret;
 }
 
-static inline int
-connect_unassure(const struct ip_conntrack *i, void *_ina)
+static int masq_device_event(struct notifier_block *this,
+			     unsigned long event,
+			     void *ptr)
 {
-	struct in_ifaddr *ina = _ina;
+	struct net_device *dev = ptr;
 
-	/* We reset the ASSURED bit on all connections, so they will
-	 * get reaped under memory pressure. */
-	if (i->nat.masq_index == ina->ifa_dev->dev->ifindex)
-		clear_bit(IPS_ASSURED_BIT, (unsigned long *)&i->status);
-	return 0;
+	if (event == NETDEV_DOWN) {
+		/* Device was downed.  Search entire table for
+		   conntracks which were associated with that device,
+		   and forget them. */
+		IP_NF_ASSERT(dev->ifindex != 0);
+
+		ip_ct_selective_cleanup(device_cmp, (void *)(long)dev->ifindex);
+	}
+
+	return NOTIFY_DONE;
 }
 
 static int masq_inet_event(struct notifier_block *this,
 			   unsigned long event,
 			   void *ptr)
 {
-	/* For some configurations, interfaces often come back with
-	 * the same address.  If not, clean up old conntrack
-	 * entries. */
-	if (event == NETDEV_UP)
-		ip_ct_selective_cleanup(device_cmp, ptr);
-	else if (event == NETDEV_DOWN)
-		ip_ct_selective_cleanup(connect_unassure, ptr);
+	struct net_device *dev = ((struct in_ifaddr *)ptr)->ifa_dev->dev;
+
+	if (event == NETDEV_DOWN) {
+		/* IP address was deleted.  Search entire table for
+		   conntracks which were associated with that device,
+		   and forget them. */
+		IP_NF_ASSERT(dev->ifindex != 0);
+
+		ip_ct_selective_cleanup(device_cmp, (void *)(long)dev->ifindex);
+	}
 
 	return NOTIFY_DONE;
 }
 
+static struct notifier_block masq_dev_notifier = {
+	.notifier_call	= masq_device_event,
+};
+
 static struct notifier_block masq_inet_notifier = {
 	.notifier_call	= masq_inet_event,
 };
@@ -178,9 +186,12 @@ static int __init init(void)
 
 	ret = ipt_register_target(&masquerade);
 
-	if (ret == 0)
+	if (ret == 0) {
+		/* Register for device down reports */
+		register_netdevice_notifier(&masq_dev_notifier);
 		/* Register IP address change reports */
 		register_inetaddr_notifier(&masq_inet_notifier);
+	}
 
 	return ret;
 }
@@ -188,6 +199,7 @@ static int __init init(void)
 static void __exit fini(void)
 {
 	ipt_unregister_target(&masquerade);
+	unregister_netdevice_notifier(&masq_dev_notifier);
 	unregister_inetaddr_notifier(&masq_inet_notifier);	
 }
 

net/ipv4/netfilter/ipt_REJECT.c

@@ -38,20 +38,6 @@ MODULE_DESCRIPTION("iptables REJECT target module");
 #define DEBUGP(format, args...)
 #endif
 
-/* If the original packet is part of a connection, but the connection
-   is not confirmed, our manufactured reply will not be associated
-   with it, so we need to do this manually. */
-static void connection_attach(struct sk_buff *new_skb, struct sk_buff *skb)
-{
-	void (*attach)(struct sk_buff *, struct sk_buff *);
-
-	/* Avoid module unload race with ip_ct_attach being NULLed out */
-	if (skb->nfct && (attach = ip_ct_attach) != NULL) {
-		mb(); /* Just to be sure: must be read before executing this */
-		attach(new_skb, skb);
-	}
-}
-
 static inline struct rtable *route_reverse(struct sk_buff *skb, int hook)
 {
 	struct iphdr *iph = skb->nh.iph;
@@ -209,7 +195,7 @@ static void send_reset(struct sk_buff *oldskb, int hook)
 	if (nskb->len > dst_pmtu(nskb->dst))
 		goto free_nskb;
 
-	connection_attach(nskb, oldskb);
+	nf_ct_attach(nskb, oldskb);
 
 	NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, nskb, NULL, nskb->dst->dev,
 		ip_finish_output);
@@ -360,7 +346,7 @@ static void send_unreach(struct sk_buff *skb_in, int code)
 	icmph->checksum = ip_compute_csum((unsigned char *)icmph,
 					  length - sizeof(struct iphdr));
 
-	connection_attach(nskb, skb_in);
+	nf_ct_attach(nskb, skb_in);
 
 	NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, nskb, NULL, nskb->dst->dev,
 		ip_finish_output);

net/ipv6/netfilter/ip6_queue.c

@@ -167,6 +167,7 @@ static inline void
 __ipq_reset(void)
 {
 	peer_pid = 0;
+	net_disable_timestamp();
 	__ipq_set_mode(IPQ_COPY_NONE, 0);
 	__ipq_flush(NF_DROP);
 }
@@ -262,7 +263,8 @@ ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp)
 	}
 	
 	if (data_len)
-		memcpy(pmsg->payload, entry->skb->data, data_len);
+		if (skb_copy_bits(entry->skb, 0, pmsg->payload, data_len))
+			BUG();
 		
 	nlh->nlmsg_len = skb->tail - old_tail;
 	return skb;
@@ -366,6 +368,8 @@ ipq_mangle_ipv6(ipq_verdict_msg_t *v, struct ipq_queue_entry *e)
 		}
 		skb_put(e->skb, diff);
 	}
+	if (!skb_ip_make_writable(&e->skb, v->data_len))
+		return -ENOMEM;
 	memcpy(e->skb->data, v->payload, v->data_len);
 	e->skb->nfcache |= NFC_ALTERED;
 
@@ -517,9 +521,10 @@ ipq_rcv_skb(struct sk_buff *skb)
 			write_unlock_bh(&queue_lock);
 			RCV_SKB_FAIL(-EBUSY);
 		}
-	}
-	else
+	} else {
+		net_enable_timestamp();
 		peer_pid = pid;
+	}
 		
 	write_unlock_bh(&queue_lock);
 	

net/ipv6/netfilter/ip6t_eui64.c

@@ -69,7 +69,7 @@ ip6t_eui64_checkentry(const char *tablename,
 {
 	if (hook_mask
 	    & ~((1 << NF_IP6_PRE_ROUTING) | (1 << NF_IP6_LOCAL_IN) |
-		(1 << NF_IP6_PRE_ROUTING) )) {
+		(1 << NF_IP6_FORWARD))) {
 		printk("ip6t_eui64: only valid for PRE_ROUTING, LOCAL_IN or FORWARD.\n");
 		return 0;
 	}

net/ipv6/netfilter/ip6t_ipv6header.c

@@ -51,7 +51,7 @@ ipv6header_match(const struct sk_buff *skb,
 	temp = 0;
 
         while (ip6t_ext_hdr(nexthdr)) {
-        	struct ipv6_opt_hdr *hdr;
+		struct ipv6_opt_hdr _hdr, *hp;
         	int hdrlen;
 
 		/* Is there enough space for the next ext header? */
@@ -68,15 +68,16 @@ ipv6header_match(const struct sk_buff *skb,
 			break;
 		}
 
-		hdr=(struct ipv6_opt_hdr *)skb->data+ptr;
+		hp = skb_header_pointer(skb, ptr, sizeof(_hdr), &_hdr);
+		BUG_ON(hp == NULL);
 
 		/* Calculate the header length */
                 if (nexthdr == NEXTHDR_FRAGMENT) {
                         hdrlen = 8;
                 } else if (nexthdr == NEXTHDR_AUTH)
-                        hdrlen = (hdr->hdrlen+2)<<2;
+                        hdrlen = (hp->hdrlen+2)<<2;
                 else
-                        hdrlen = ipv6_optlen(hdr);
+                        hdrlen = ipv6_optlen(hp);
 
 		/* set the flag */
 		switch (nexthdr){
@@ -100,7 +101,7 @@ ipv6header_match(const struct sk_buff *skb,
 				break;
 		}
 
-                nexthdr = hdr->nexthdr;
+                nexthdr = hp->nexthdr;
                 len -= hdrlen;
                 ptr += hdrlen;
 		if (ptr > skb->len)
@@ -111,10 +112,14 @@ ipv6header_match(const struct sk_buff *skb,
 		temp |= MASK_PROTO;
 
 	if (info->modeflag)
-		return (!( (temp & info->matchflags)
-			^ info->matchflags) ^ info->invflags);
+		return !((temp ^ info->matchflags ^ info->invflags)
+			 & info->matchflags);
+	else {
+		if (info->invflags)
+			return temp != info->matchflags;
 		else
-		return (!( temp ^ info->matchflags) ^ info->invflags);
+			return temp == info->matchflags;
+	}
 }
 
 static int
@@ -124,11 +129,18 @@ ipv6header_checkentry(const char *tablename,
 		      unsigned int matchsize,
 		      unsigned int hook_mask)
 {
+	const struct ip6t_ipv6header_info *info = matchinfo;
+
 	/* Check for obvious errors */
 	/* This match is valid in all hooks! */
 	if (matchsize != IP6T_ALIGN(sizeof(struct ip6t_ipv6header_info)))
 		return 0;
 
+	/* invflags is 0 or 0xff in hard mode */
+	if ((!info->modeflag) && info->invflags != 0x00
+			      && info->invflags != 0xFF)
+		return 0;
+
 	return 1;
 }