[NETFILTER]: add IPv6-capable TCPMSS target

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

include/linux/netfilter/Kbuild

@@ -33,6 +33,7 @@ header-y += xt_tcpmss.h
 header-y += xt_tcpudp.h
 header-y += xt_SECMARK.h
 header-y += xt_CONNSECMARK.h
+header-y += xt_TCPMSS.h
 
 unifdef-y += nf_conntrack_common.h
 unifdef-y += nf_conntrack_ftp.h

include/linux/netfilter/xt_TCPMSS.h

+#ifndef _XT_TCPMSS_H
+#define _XT_TCPMSS_H
+
+struct xt_tcpmss_info {
+	u_int16_t mss;
+};
+
+#define XT_TCPMSS_CLAMP_PMTU 0xffff
+
+#endif /* _XT_TCPMSS_H */

include/linux/netfilter_ipv4/ipt_TCPMSS.h

 #ifndef _IPT_TCPMSS_H
 #define _IPT_TCPMSS_H
 
-struct ipt_tcpmss_info {
-	u_int16_t mss;
-};
+#include <linux/netfilter/xt_TCPMSS.h>
 
-#define IPT_TCPMSS_CLAMP_PMTU 0xffff
+#define ipt_tcpmss_info		xt_tcpmss_info
+#define IPT_TCPMSS_CLAMP_PMTU	XT_TCPMSS_CLAMP_PMTU
 
 #endif /*_IPT_TCPMSS_H*/

net/ipv4/netfilter/Kconfig

@@ -361,32 +361,6 @@ config IP_NF_TARGET_ULOG
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
-config IP_NF_TARGET_TCPMSS
-	tristate "TCPMSS target support"
-	depends on IP_NF_IPTABLES
-	---help---
-	  This option adds a `TCPMSS' target, which allows you to alter the
-	  MSS value of TCP SYN packets, to control the maximum size for that
-	  connection (usually limiting it to your outgoing interface's MTU
-	  minus 40).
-
-	  This is used to overcome criminally braindead ISPs or servers which
-	  block ICMP Fragmentation Needed packets.  The symptoms of this
-	  problem are that everything works fine from your Linux
-	  firewall/router, but machines behind it can never exchange large
-	  packets:
-	  	1) Web browsers connect, then hang with no data received.
-	  	2) Small mail works fine, but large emails hang.
-	  	3) ssh works fine, but scp hangs after initial handshaking.
-
-	  Workaround: activate this option and add a rule to your firewall
-	  configuration like:
-
-	  iptables -A FORWARD -p tcp --tcp-flags SYN,RST SYN \
-	  		 -j TCPMSS --clamp-mss-to-pmtu
-
-	  To compile it as a module, choose M here.  If unsure, say N.
-
 # NAT + specific targets: ip_conntrack
 config IP_NF_NAT
 	tristate "Full NAT"

net/ipv4/netfilter/Makefile

@@ -103,7 +103,6 @@ obj-$(CONFIG_IP_NF_TARGET_SAME) += ipt_SAME.o
 obj-$(CONFIG_IP_NF_NAT_SNMP_BASIC) += ip_nat_snmp_basic.o
 obj-$(CONFIG_IP_NF_TARGET_LOG) += ipt_LOG.o
 obj-$(CONFIG_IP_NF_TARGET_ULOG) += ipt_ULOG.o
-obj-$(CONFIG_IP_NF_TARGET_TCPMSS) += ipt_TCPMSS.o
 obj-$(CONFIG_IP_NF_TARGET_CLUSTERIP) += ipt_CLUSTERIP.o
 obj-$(CONFIG_IP_NF_TARGET_TTL) += ipt_TTL.o
 

net/netfilter/Kconfig

@@ -395,6 +395,32 @@ config NETFILTER_XT_TARGET_CONNSECMARK
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
+config NETFILTER_XT_TARGET_TCPMSS
+	tristate '"TCPMSS" target support'
+	depends on NETFILTER_XTABLES && (IPV6 || IPV6=n)
+	---help---
+	  This option adds a `TCPMSS' target, which allows you to alter the
+	  MSS value of TCP SYN packets, to control the maximum size for that
+	  connection (usually limiting it to your outgoing interface's MTU
+	  minus 40).
+
+	  This is used to overcome criminally braindead ISPs or servers which
+	  block ICMP Fragmentation Needed packets.  The symptoms of this
+	  problem are that everything works fine from your Linux
+	  firewall/router, but machines behind it can never exchange large
+	  packets:
+	        1) Web browsers connect, then hang with no data received.
+	        2) Small mail works fine, but large emails hang.
+	        3) ssh works fine, but scp hangs after initial handshaking.
+
+	  Workaround: activate this option and add a rule to your firewall
+	  configuration like:
+
+	  iptables -A FORWARD -p tcp --tcp-flags SYN,RST SYN \
+	                 -j TCPMSS --clamp-mss-to-pmtu
+
+	  To compile it as a module, choose M here.  If unsure, say N.
+
 config NETFILTER_XT_MATCH_COMMENT
 	tristate  '"comment" match support'
 	depends on NETFILTER_XTABLES

net/netfilter/Makefile

@@ -45,6 +45,7 @@ obj-$(CONFIG_NETFILTER_XT_TARGET_NFQUEUE) += xt_NFQUEUE.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_NFLOG) += xt_NFLOG.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_NOTRACK) += xt_NOTRACK.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_SECMARK) += xt_SECMARK.o
+obj-$(CONFIG_NETFILTER_XT_TARGET_TCPMSS) += xt_TCPMSS.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_CONNSECMARK) += xt_CONNSECMARK.o
 
 # matches

net/ipv4/netfilter/ipt_TCPMSS.c → net/netfilter/xt_TCPMSS.c

@@ -10,16 +10,23 @@
 
 #include <linux/module.h>
 #include <linux/skbuff.h>
-
 #include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/tcp.h>
+#include <net/ipv6.h>
 #include <net/tcp.h>
 
 #include <linux/netfilter_ipv4/ip_tables.h>
-#include <linux/netfilter_ipv4/ipt_TCPMSS.h>
+#include <linux/netfilter_ipv6/ip6_tables.h>
+#include <linux/netfilter/x_tables.h>
+#include <linux/netfilter/xt_tcpudp.h>
+#include <linux/netfilter/xt_TCPMSS.h>
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Marc Boucher <marc@mbsi.ca>");
-MODULE_DESCRIPTION("iptables TCP MSS modification module");
+MODULE_DESCRIPTION("x_tables TCP MSS modification module");
+MODULE_ALIAS("ipt_TCPMSS");
+MODULE_ALIAS("ip6t_TCPMSS");
 
 static inline unsigned int
 optlen(const u_int8_t *opt, unsigned int offset)
@@ -31,28 +38,23 @@ optlen(const u_int8_t *opt, unsigned int offset)
 		return opt[offset+1];
 }
 
-static unsigned int
-ipt_tcpmss_target(struct sk_buff **pskb,
-		  const struct net_device *in,
-		  const struct net_device *out,
-		  unsigned int hooknum,
-		  const struct xt_target *target,
-		  const void *targinfo)
+static int
+tcpmss_mangle_packet(struct sk_buff **pskb,
+		     const struct xt_tcpmss_info *info,
+		     unsigned int tcphoff,
+		     unsigned int minlen)
 {
-	const struct ipt_tcpmss_info *tcpmssinfo = targinfo;
 	struct tcphdr *tcph;
-	struct iphdr *iph;
-	u_int16_t tcplen, newmss;
-	__be16 newtotlen, oldval;
-	unsigned int i;
-	u_int8_t *opt;
+	unsigned int tcplen, i;
+	__be16 oldval;
+	u16 newmss;
+	u8 *opt;
 
 	if (!skb_make_writable(pskb, (*pskb)->len))
-		return NF_DROP;
+		return -1;
 
-	iph = (*pskb)->nh.iph;
-	tcplen = (*pskb)->len - iph->ihl*4;
-	tcph = (void *)iph + iph->ihl*4;
+	tcplen = (*pskb)->len - tcphoff;
+	tcph = (struct tcphdr *)((*pskb)->nh.raw + tcphoff);
 
 	/* Since it passed flags test in tcp match, we know it is is
 	   not a fragment, and has data >= tcp header length.  SYN
@@ -61,28 +63,24 @@ ipt_tcpmss_target(struct sk_buff **pskb,
 	   badly. --RR */
 	if (tcplen != tcph->doff*4) {
 		if (net_ratelimit())
-			printk(KERN_ERR
-			       "ipt_tcpmss_target: bad length (%d bytes)\n",
+			printk(KERN_ERR "xt_TCPMSS: bad length (%u bytes)\n",
 			       (*pskb)->len);
-		return NF_DROP;
+		return -1;
 	}
 
-	if (tcpmssinfo->mss == IPT_TCPMSS_CLAMP_PMTU) {
-		if (dst_mtu((*pskb)->dst) <= sizeof(struct iphdr) +
-					     sizeof(struct tcphdr)) {
+	if (info->mss == XT_TCPMSS_CLAMP_PMTU) {
+		if (dst_mtu((*pskb)->dst) <= minlen) {
 			if (net_ratelimit())
-				printk(KERN_ERR "ipt_tcpmss_target: "
-				       "unknown or invalid path-MTU (%d)\n",
+				printk(KERN_ERR "xt_TCPMSS: "
+				       "unknown or invalid path-MTU (%u)\n",
 				       dst_mtu((*pskb)->dst));
-			return NF_DROP; /* or IPT_CONTINUE ?? */
+			return -1;
 		}
-
-		newmss = dst_mtu((*pskb)->dst) - sizeof(struct iphdr) -
-						 sizeof(struct tcphdr);
+		newmss = dst_mtu((*pskb)->dst) - minlen;
 	} else
-		newmss = tcpmssinfo->mss;
+		newmss = info->mss;
 
- 	opt = (u_int8_t *)tcph;
+	opt = (u_int8_t *)tcph;
 	for (i = sizeof(struct tcphdr); i < tcph->doff*4; i += optlen(opt, i)) {
 		if (opt[i] == TCPOPT_MSS && tcph->doff*4 - i >= TCPOLEN_MSS &&
 		    opt[i+1] == TCPOLEN_MSS) {
@@ -90,16 +88,16 @@ ipt_tcpmss_target(struct sk_buff **pskb,
 
 			oldmss = (opt[i+2] << 8) | opt[i+3];
 
-			if (tcpmssinfo->mss == IPT_TCPMSS_CLAMP_PMTU &&
+			if (info->mss == XT_TCPMSS_CLAMP_PMTU &&
 			    oldmss <= newmss)
-				return IPT_CONTINUE;
+				return 0;
 
 			opt[i+2] = (newmss & 0xff00) >> 8;
 			opt[i+3] = (newmss & 0x00ff);
 
 			nf_proto_csum_replace2(&tcph->check, *pskb,
-						htons(oldmss), htons(newmss), 0);
-			return IPT_CONTINUE;
+					       htons(oldmss), htons(newmss), 0);
+			return 0;
 		}
 	}
 
@@ -112,20 +110,19 @@ ipt_tcpmss_target(struct sk_buff **pskb,
 		newskb = skb_copy_expand(*pskb, skb_headroom(*pskb),
 					 TCPOLEN_MSS, GFP_ATOMIC);
 		if (!newskb)
-			return NF_DROP;
+			return -1;
 		kfree_skb(*pskb);
 		*pskb = newskb;
-		iph = (*pskb)->nh.iph;
-		tcph = (void *)iph + iph->ihl*4;
+		tcph = (struct tcphdr *)((*pskb)->nh.raw + tcphoff);
 	}
 
 	skb_put((*pskb), TCPOLEN_MSS);
 
- 	opt = (u_int8_t *)tcph + sizeof(struct tcphdr);
+	opt = (u_int8_t *)tcph + sizeof(struct tcphdr);
 	memmove(opt + TCPOLEN_MSS, opt, tcplen - sizeof(struct tcphdr));
 
 	nf_proto_csum_replace2(&tcph->check, *pskb,
-				htons(tcplen), htons(tcplen + TCPOLEN_MSS), 1);
+			       htons(tcplen), htons(tcplen + TCPOLEN_MSS), 1);
 	opt[0] = TCPOPT_MSS;
 	opt[1] = TCPOLEN_MSS;
 	opt[2] = (newmss & 0xff00) >> 8;
@@ -137,71 +134,163 @@ ipt_tcpmss_target(struct sk_buff **pskb,
 	tcph->doff += TCPOLEN_MSS/4;
 	nf_proto_csum_replace2(&tcph->check, *pskb,
 				oldval, ((__be16 *)tcph)[6], 0);
+	return TCPOLEN_MSS;
+}
 
-	newtotlen = htons(ntohs(iph->tot_len) + TCPOLEN_MSS);
-	nf_csum_replace2(&iph->check, iph->tot_len, newtotlen);
-	iph->tot_len = newtotlen;
-	return IPT_CONTINUE;
+static unsigned int
+xt_tcpmss_target4(struct sk_buff **pskb,
+		  const struct net_device *in,
+		  const struct net_device *out,
+		  unsigned int hooknum,
+		  const struct xt_target *target,
+		  const void *targinfo)
+{
+	struct iphdr *iph = (*pskb)->nh.iph;
+	__be16 newlen;
+	int ret;
+
+	ret = tcpmss_mangle_packet(pskb, targinfo, iph->ihl * 4,
+				   sizeof(*iph) + sizeof(struct tcphdr));
+	if (ret < 0)
+		return NF_DROP;
+	if (ret > 0) {
+		iph = (*pskb)->nh.iph;
+		newlen = htons(ntohs(iph->tot_len) + ret);
+		nf_csum_replace2(&iph->check, iph->tot_len, newlen);
+		iph->tot_len = newlen;
+	}
+	return XT_CONTINUE;
 }
 
+#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE)
+static unsigned int
+xt_tcpmss_target6(struct sk_buff **pskb,
+		  const struct net_device *in,
+		  const struct net_device *out,
+		  unsigned int hooknum,
+		  const struct xt_target *target,
+		  const void *targinfo)
+{
+	struct ipv6hdr *ipv6h = (*pskb)->nh.ipv6h;
+	u8 nexthdr;
+	int tcphoff;
+	int ret;
+
+	nexthdr = ipv6h->nexthdr;
+	tcphoff = ipv6_skip_exthdr(*pskb, sizeof(*ipv6h), &nexthdr);
+	if (tcphoff < 0) {
+		WARN_ON(1);
+		return NF_DROP;
+	}
+	ret = tcpmss_mangle_packet(pskb, targinfo, tcphoff,
+				   sizeof(*ipv6h) + sizeof(struct tcphdr));
+	if (ret < 0)
+		return NF_DROP;
+	if (ret > 0) {
+		ipv6h = (*pskb)->nh.ipv6h;
+		ipv6h->payload_len = htons(ntohs(ipv6h->payload_len) + ret);
+	}
+	return XT_CONTINUE;
+}
+#endif
+
 #define TH_SYN 0x02
 
-static inline int find_syn_match(const struct ipt_entry_match *m)
+/* Must specify -p tcp --syn */
+static inline int find_syn_match(const struct xt_entry_match *m)
 {
-	const struct ipt_tcp *tcpinfo = (const struct ipt_tcp *)m->data;
+	const struct xt_tcp *tcpinfo = (const struct xt_tcp *)m->data;
 
 	if (strcmp(m->u.kernel.match->name, "tcp") == 0 &&
 	    tcpinfo->flg_cmp & TH_SYN &&
-	    !(tcpinfo->invflags & IPT_TCP_INV_FLAGS))
+	    !(tcpinfo->invflags & XT_TCP_INV_FLAGS))
 		return 1;
 
 	return 0;
 }
 
-/* Must specify -p tcp --syn/--tcp-flags SYN */
 static int
-ipt_tcpmss_checkentry(const char *tablename,
-		      const void *e_void,
+xt_tcpmss_checkentry4(const char *tablename,
+		      const void *entry,
 		      const struct xt_target *target,
 		      void *targinfo,
 		      unsigned int hook_mask)
 {
-	const struct ipt_tcpmss_info *tcpmssinfo = targinfo;
-	const struct ipt_entry *e = e_void;
+	const struct xt_tcpmss_info *info = targinfo;
+	const struct ipt_entry *e = entry;
 
-	if (tcpmssinfo->mss == IPT_TCPMSS_CLAMP_PMTU &&
+	if (info->mss == XT_TCPMSS_CLAMP_PMTU &&
 	    (hook_mask & ~((1 << NF_IP_FORWARD) |
 			   (1 << NF_IP_LOCAL_OUT) |
 			   (1 << NF_IP_POST_ROUTING))) != 0) {
-		printk("TCPMSS: path-MTU clamping only supported in "
+		printk("xt_TCPMSS: path-MTU clamping only supported in "
 		       "FORWARD, OUTPUT and POSTROUTING hooks\n");
 		return 0;
 	}
-
 	if (IPT_MATCH_ITERATE(e, find_syn_match))
 		return 1;
-	printk("TCPMSS: Only works on TCP SYN packets\n");
+	printk("xt_TCPMSS: Only works on TCP SYN packets\n");
 	return 0;
 }
 
-static struct ipt_target ipt_tcpmss_reg = {
-	.name		= "TCPMSS",
-	.target		= ipt_tcpmss_target,
-	.targetsize	= sizeof(struct ipt_tcpmss_info),
-	.proto		= IPPROTO_TCP,
-	.checkentry	= ipt_tcpmss_checkentry,
-	.me		= THIS_MODULE,
+#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE)
+static int
+xt_tcpmss_checkentry6(const char *tablename,
+		      const void *entry,
+		      const struct xt_target *target,
+		      void *targinfo,
+		      unsigned int hook_mask)
+{
+	const struct xt_tcpmss_info *info = targinfo;
+	const struct ip6t_entry *e = entry;
+
+	if (info->mss == XT_TCPMSS_CLAMP_PMTU &&
+	    (hook_mask & ~((1 << NF_IP6_FORWARD) |
+			   (1 << NF_IP6_LOCAL_OUT) |
+			   (1 << NF_IP6_POST_ROUTING))) != 0) {
+		printk("xt_TCPMSS: path-MTU clamping only supported in "
+		       "FORWARD, OUTPUT and POSTROUTING hooks\n");
+		return 0;
+	}
+	if (IP6T_MATCH_ITERATE(e, find_syn_match))
+		return 1;
+	printk("xt_TCPMSS: Only works on TCP SYN packets\n");
+	return 0;
+}
+#endif
+
+static struct xt_target xt_tcpmss_reg[] = {
+	{
+		.family		= AF_INET,
+		.name		= "TCPMSS",
+		.checkentry	= xt_tcpmss_checkentry4,
+		.target		= xt_tcpmss_target4,
+		.targetsize	= sizeof(struct xt_tcpmss_info),
+		.proto		= IPPROTO_TCP,
+		.me		= THIS_MODULE,
+	},
+#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE)
+	{
+		.family		= AF_INET6,
+		.name		= "TCPMSS",
+		.checkentry	= xt_tcpmss_checkentry6,
+		.target		= xt_tcpmss_target6,
+		.targetsize	= sizeof(struct xt_tcpmss_info),
+		.proto		= IPPROTO_TCP,
+		.me		= THIS_MODULE,
+	},
+#endif
 };
 
-static int __init ipt_tcpmss_init(void)
+static int __init xt_tcpmss_init(void)
 {
-	return ipt_register_target(&ipt_tcpmss_reg);
+	return xt_register_targets(xt_tcpmss_reg, ARRAY_SIZE(xt_tcpmss_reg));
 }
 
-static void __exit ipt_tcpmss_fini(void)
+static void __exit xt_tcpmss_fini(void)
 {
-	ipt_unregister_target(&ipt_tcpmss_reg);
+	xt_unregister_targets(xt_tcpmss_reg, ARRAY_SIZE(xt_tcpmss_reg));
 }
 
-module_init(ipt_tcpmss_init);
-module_exit(ipt_tcpmss_fini);
+module_init(xt_tcpmss_init);
+module_exit(xt_tcpmss_fini);