Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/gi…

…t/pablo/nf-next

Pablo Neira Ayuso says:

====================
The following patchset contains Netfilter updates for your net-next tree,
they are:

* The new SYNPROXY target for iptables, including IPv4 and IPv6 support,
  from Patrick McHardy.

* nf_defrag_ipv6.o should be only linked to nf_defrag_ipv6.ko, from
  Nathan Hintz.

* Fix an old bug in REJECT, which replies with wrong MAC source address
  from the bridge, by Phil Oester.

* Fix uninitialized helper variable in the expectation support over
  nfnetlink_queue, from Florian Westphal.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

include/linux/netfilter.h

@@ -319,6 +319,7 @@ extern void nf_ct_attach(struct sk_buff *, const struct sk_buff *);
 extern void (*nf_ct_destroy)(struct nf_conntrack *) __rcu;
 
 struct nf_conn;
+enum ip_conntrack_info;
 struct nlattr;
 
 struct nfq_ct_hook {
@@ -327,14 +328,10 @@ struct nfq_ct_hook {
 	int (*parse)(const struct nlattr *attr, struct nf_conn *ct);
 	int (*attach_expect)(const struct nlattr *attr, struct nf_conn *ct,
 			     u32 portid, u32 report);
-};
-extern struct nfq_ct_hook __rcu *nfq_ct_hook;
-
-struct nfq_ct_nat_hook {
 	void (*seq_adjust)(struct sk_buff *skb, struct nf_conn *ct,
-			   u32 ctinfo, s32 off);
+			   enum ip_conntrack_info ctinfo, s32 off);
 };
-extern struct nfq_ct_nat_hook __rcu *nfq_ct_nat_hook;
+extern struct nfq_ct_hook __rcu *nfq_ct_hook;
 #else
 static inline void nf_ct_attach(struct sk_buff *new, struct sk_buff *skb) {}
 #endif

include/net/netfilter/nf_conntrack_extend.h

@@ -10,6 +10,7 @@ enum nf_ct_ext_id {
 #if defined(CONFIG_NF_NAT) || defined(CONFIG_NF_NAT_MODULE)
 	NF_CT_EXT_NAT,
 #endif
+	NF_CT_EXT_SEQADJ,
 	NF_CT_EXT_ACCT,
 #ifdef CONFIG_NF_CONNTRACK_EVENTS
 	NF_CT_EXT_ECACHE,
@@ -25,18 +26,23 @@ enum nf_ct_ext_id {
 #endif
 #ifdef CONFIG_NF_CONNTRACK_LABELS
 	NF_CT_EXT_LABELS,
+#endif
+#if IS_ENABLED(CONFIG_NETFILTER_SYNPROXY)
+	NF_CT_EXT_SYNPROXY,
 #endif
 	NF_CT_EXT_NUM,
 };
 
 #define NF_CT_EXT_HELPER_TYPE struct nf_conn_help
 #define NF_CT_EXT_NAT_TYPE struct nf_conn_nat
+#define NF_CT_EXT_SEQADJ_TYPE struct nf_conn_seqadj
 #define NF_CT_EXT_ACCT_TYPE struct nf_conn_counter
 #define NF_CT_EXT_ECACHE_TYPE struct nf_conntrack_ecache
 #define NF_CT_EXT_ZONE_TYPE struct nf_conntrack_zone
 #define NF_CT_EXT_TSTAMP_TYPE struct nf_conn_tstamp
 #define NF_CT_EXT_TIMEOUT_TYPE struct nf_conn_timeout
 #define NF_CT_EXT_LABELS_TYPE struct nf_conn_labels
+#define NF_CT_EXT_SYNPROXY_TYPE struct nf_conn_synproxy
 
 /* Extensions: optional stuff which isn't permanently in struct. */
 struct nf_ct_ext {

include/net/netfilter/nf_conntrack_seqadj.h

@@ -0,0 +1,51 @@
+#ifndef _NF_CONNTRACK_SEQADJ_H
+#define _NF_CONNTRACK_SEQADJ_H
+
+#include <net/netfilter/nf_conntrack_extend.h>
+
+/**
+ * struct nf_ct_seqadj - sequence number adjustment information
+ *
+ * @correction_pos: position of the last TCP sequence number modification
+ * @offset_before: sequence number offset before last modification
+ * @offset_after: sequence number offset after last modification
+ */
+struct nf_ct_seqadj {
+	u32		correction_pos;
+	s32		offset_before;
+	s32		offset_after;
+};
+
+struct nf_conn_seqadj {
+	struct nf_ct_seqadj	seq[IP_CT_DIR_MAX];
+};
+
+static inline struct nf_conn_seqadj *nfct_seqadj(const struct nf_conn *ct)
+{
+	return nf_ct_ext_find(ct, NF_CT_EXT_SEQADJ);
+}
+
+static inline struct nf_conn_seqadj *nfct_seqadj_ext_add(struct nf_conn *ct)
+{
+	return nf_ct_ext_add(ct, NF_CT_EXT_SEQADJ, GFP_ATOMIC);
+}
+
+extern int nf_ct_seqadj_init(struct nf_conn *ct, enum ip_conntrack_info ctinfo,
+			     s32 off);
+extern int nf_ct_seqadj_set(struct nf_conn *ct, enum ip_conntrack_info ctinfo,
+			    __be32 seq, s32 off);
+extern void nf_ct_tcp_seqadj_set(struct sk_buff *skb,
+				 struct nf_conn *ct,
+				 enum ip_conntrack_info ctinfo,
+				 s32 off);
+
+extern int nf_ct_seq_adjust(struct sk_buff *skb,
+			    struct nf_conn *ct, enum ip_conntrack_info ctinfo,
+			    unsigned int protoff);
+extern s32 nf_ct_seq_offset(const struct nf_conn *ct, enum ip_conntrack_dir,
+			    u32 seq);
+
+extern int nf_conntrack_seqadj_init(void);
+extern void nf_conntrack_seqadj_fini(void);
+
+#endif /* _NF_CONNTRACK_SEQADJ_H */

include/net/netfilter/nf_conntrack_synproxy.h

@@ -0,0 +1,77 @@
+#ifndef _NF_CONNTRACK_SYNPROXY_H
+#define _NF_CONNTRACK_SYNPROXY_H
+
+#include <net/netns/generic.h>
+
+struct nf_conn_synproxy {
+	u32	isn;
+	u32	its;
+	u32	tsoff;
+};
+
+static inline struct nf_conn_synproxy *nfct_synproxy(const struct nf_conn *ct)
+{
+#if IS_ENABLED(CONFIG_NETFILTER_SYNPROXY)
+	return nf_ct_ext_find(ct, NF_CT_EXT_SYNPROXY);
+#else
+	return NULL;
+#endif
+}
+
+static inline struct nf_conn_synproxy *nfct_synproxy_ext_add(struct nf_conn *ct)
+{
+#if IS_ENABLED(CONFIG_NETFILTER_SYNPROXY)
+	return nf_ct_ext_add(ct, NF_CT_EXT_SYNPROXY, GFP_ATOMIC);
+#else
+	return NULL;
+#endif
+}
+
+struct synproxy_stats {
+	unsigned int			syn_received;
+	unsigned int			cookie_invalid;
+	unsigned int			cookie_valid;
+	unsigned int			cookie_retrans;
+	unsigned int			conn_reopened;
+};
+
+struct synproxy_net {
+	struct nf_conn			*tmpl;
+	struct synproxy_stats __percpu	*stats;
+};
+
+extern int synproxy_net_id;
+static inline struct synproxy_net *synproxy_pernet(struct net *net)
+{
+	return net_generic(net, synproxy_net_id);
+}
+
+struct synproxy_options {
+	u8				options;
+	u8				wscale;
+	u16				mss;
+	u32				tsval;
+	u32				tsecr;
+};
+
+struct tcphdr;
+struct xt_synproxy_info;
+extern void synproxy_parse_options(const struct sk_buff *skb, unsigned int doff,
+				   const struct tcphdr *th,
+				   struct synproxy_options *opts);
+extern unsigned int synproxy_options_size(const struct synproxy_options *opts);
+extern void synproxy_build_options(struct tcphdr *th,
+				   const struct synproxy_options *opts);
+
+extern void synproxy_init_timestamp_cookie(const struct xt_synproxy_info *info,
+					   struct synproxy_options *opts);
+extern void synproxy_check_timestamp_cookie(struct synproxy_options *opts);
+
+extern unsigned int synproxy_tstamp_adjust(struct sk_buff *skb,
+					   unsigned int protoff,
+					   struct tcphdr *th,
+					   struct nf_conn *ct,
+					   enum ip_conntrack_info ctinfo,
+					   const struct nf_conn_synproxy *synproxy);
+
+#endif /* _NF_CONNTRACK_SYNPROXY_H */

include/net/netfilter/nf_nat.h

@@ -13,15 +13,6 @@ enum nf_nat_manip_type {
 #define HOOK2MANIP(hooknum) ((hooknum) != NF_INET_POST_ROUTING && \
 			     (hooknum) != NF_INET_LOCAL_IN)
 
-/* NAT sequence number modifications */
-struct nf_nat_seq {
-	/* position of the last TCP sequence number modification (if any) */
-	u_int32_t correction_pos;
-
-	/* sequence number offset before and after last modification */
-	int32_t offset_before, offset_after;
-};
-
 #include <linux/list.h>
 #include <linux/netfilter/nf_conntrack_pptp.h>
 #include <net/netfilter/nf_conntrack_extend.h>
@@ -39,7 +30,6 @@ struct nf_conn;
 /* The structure embedded in the conntrack structure. */
 struct nf_conn_nat {
 	struct hlist_node bysource;
-	struct nf_nat_seq seq[IP_CT_DIR_MAX];
 	struct nf_conn *ct;
 	union nf_conntrack_nat_help help;
 #if defined(CONFIG_IP_NF_TARGET_MASQUERADE) || \

include/net/netfilter/nf_nat_helper.h

@@ -39,28 +39,9 @@ extern int nf_nat_mangle_udp_packet(struct sk_buff *skb,
 				    const char *rep_buffer,
 				    unsigned int rep_len);
 
-extern void nf_nat_set_seq_adjust(struct nf_conn *ct,
-				  enum ip_conntrack_info ctinfo,
-				  __be32 seq, s32 off);
-extern int nf_nat_seq_adjust(struct sk_buff *skb,
-			     struct nf_conn *ct,
-			     enum ip_conntrack_info ctinfo,
-			     unsigned int protoff);
-extern int (*nf_nat_seq_adjust_hook)(struct sk_buff *skb,
-				     struct nf_conn *ct,
-				     enum ip_conntrack_info ctinfo,
-				     unsigned int protoff);
-
 /* Setup NAT on this expected conntrack so it follows master, but goes
  * to port ct->master->saved_proto. */
 extern void nf_nat_follow_master(struct nf_conn *ct,
 				 struct nf_conntrack_expect *this);
 
-extern s32 nf_nat_get_offset(const struct nf_conn *ct,
-			     enum ip_conntrack_dir dir,
-			     u32 seq);
-
-extern void nf_nat_tcp_seq_adjust(struct sk_buff *skb, struct nf_conn *ct,
-				  u32 dir, s32 off);
-
 #endif

include/net/tcp.h

@@ -476,9 +476,13 @@ void inet_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb);
 
 /* From syncookies.c */
 extern __u32 syncookie_secret[2][16-4+SHA_DIGEST_WORDS];
+extern int __cookie_v4_check(const struct iphdr *iph, const struct tcphdr *th,
+			     u32 cookie);
 extern struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, 
 				    struct ip_options *opt);
 #ifdef CONFIG_SYN_COOKIES
+extern u32 __cookie_v4_init_sequence(const struct iphdr *iph,
+				     const struct tcphdr *th, u16 *mssp);
 extern __u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, 
 				     __u16 *mss);
 #else
@@ -495,8 +499,12 @@ extern bool cookie_check_timestamp(struct tcp_options_received *opt,
 				struct net *net, bool *ecn_ok);
 
 /* From net/ipv6/syncookies.c */
+extern int __cookie_v6_check(const struct ipv6hdr *iph, const struct tcphdr *th,
+			     u32 cookie);
 extern struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb);
 #ifdef CONFIG_SYN_COOKIES
+extern u32 __cookie_v6_init_sequence(const struct ipv6hdr *iph,
+				     const struct tcphdr *th, u16 *mssp);
 extern __u32 cookie_v6_init_sequence(struct sock *sk, const struct sk_buff *skb,
 				     __u16 *mss);
 #else

include/uapi/linux/netfilter/nf_conntrack_common.h

@@ -99,7 +99,8 @@ enum ip_conntrack_events {
 	IPCT_PROTOINFO,		/* protocol information has changed */
 	IPCT_HELPER,		/* new helper has been set */
 	IPCT_MARK,		/* new mark has been set */
-	IPCT_NATSEQADJ,		/* NAT is doing sequence adjustment */
+	IPCT_SEQADJ,		/* sequence adjustment has changed */
+	IPCT_NATSEQADJ = IPCT_SEQADJ,
 	IPCT_SECMARK,		/* new security mark has been set */
 	IPCT_LABEL,		/* new connlabel has been set */
 };

include/uapi/linux/netfilter/nfnetlink_conntrack.h

@@ -42,8 +42,10 @@ enum ctattr_type {
 	CTA_ID,
 	CTA_NAT_DST,
 	CTA_TUPLE_MASTER,
-	CTA_NAT_SEQ_ADJ_ORIG,
-	CTA_NAT_SEQ_ADJ_REPLY,
+	CTA_SEQ_ADJ_ORIG,
+	CTA_NAT_SEQ_ADJ_ORIG	= CTA_SEQ_ADJ_ORIG,
+	CTA_SEQ_ADJ_REPLY,
+	CTA_NAT_SEQ_ADJ_REPLY	= CTA_SEQ_ADJ_REPLY,
 	CTA_SECMARK,		/* obsolete */
 	CTA_ZONE,
 	CTA_SECCTX,
@@ -165,6 +167,15 @@ enum ctattr_protonat {
 };
 #define CTA_PROTONAT_MAX (__CTA_PROTONAT_MAX - 1)
 
+enum ctattr_seqadj {
+	CTA_SEQADJ_UNSPEC,
+	CTA_SEQADJ_CORRECTION_POS,
+	CTA_SEQADJ_OFFSET_BEFORE,
+	CTA_SEQADJ_OFFSET_AFTER,
+	__CTA_SEQADJ_MAX
+};
+#define CTA_SEQADJ_MAX (__CTA_SEQADJ_MAX - 1)
+
 enum ctattr_natseq {
 	CTA_NAT_SEQ_UNSPEC,
 	CTA_NAT_SEQ_CORRECTION_POS,

include/uapi/linux/netfilter/xt_SYNPROXY.h

@@ -0,0 +1,16 @@
+#ifndef _XT_SYNPROXY_H
+#define _XT_SYNPROXY_H
+
+#define XT_SYNPROXY_OPT_MSS		0x01
+#define XT_SYNPROXY_OPT_WSCALE		0x02
+#define XT_SYNPROXY_OPT_SACK_PERM	0x04
+#define XT_SYNPROXY_OPT_TIMESTAMP	0x08
+#define XT_SYNPROXY_OPT_ECN		0x10
+
+struct xt_synproxy_info {
+	__u8	options;
+	__u8	wscale;
+	__u16	mss;
+};
+
+#endif /* _XT_SYNPROXY_H */

net/ipv4/netfilter/Kconfig

@@ -110,6 +110,19 @@ config IP_NF_TARGET_REJECT
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
+config IP_NF_TARGET_SYNPROXY
+	tristate "SYNPROXY target support"
+	depends on NF_CONNTRACK && NETFILTER_ADVANCED
+	select NETFILTER_SYNPROXY
+	select SYN_COOKIES
+	help
+	  The SYNPROXY target allows you to intercept TCP connections and
+	  establish them using syncookies before they are passed on to the
+	  server. This allows to avoid conntrack and server resource usage
+	  during SYN-flood attacks.
+
+	  To compile it as a module, choose M here. If unsure, say N.
+
 config IP_NF_TARGET_ULOG
 	tristate "ULOG target support (obsolete)"
 	default m if NETFILTER_ADVANCED=n

net/ipv4/netfilter/Makefile

@@ -46,6 +46,7 @@ obj-$(CONFIG_IP_NF_TARGET_CLUSTERIP) += ipt_CLUSTERIP.o
 obj-$(CONFIG_IP_NF_TARGET_ECN) += ipt_ECN.o
 obj-$(CONFIG_IP_NF_TARGET_MASQUERADE) += ipt_MASQUERADE.o
 obj-$(CONFIG_IP_NF_TARGET_REJECT) += ipt_REJECT.o
+obj-$(CONFIG_IP_NF_TARGET_SYNPROXY) += ipt_SYNPROXY.o
 obj-$(CONFIG_IP_NF_TARGET_ULOG) += ipt_ULOG.o
 
 # generic ARP tables

net/ipv4/netfilter/ipt_REJECT.c

@@ -119,7 +119,26 @@ static void send_reset(struct sk_buff *oldskb, int hook)
 
 	nf_ct_attach(nskb, oldskb);
 
-	ip_local_out(nskb);
+#ifdef CONFIG_BRIDGE_NETFILTER
+	/* If we use ip_local_out for bridged traffic, the MAC source on
+	 * the RST will be ours, instead of the destination's.  This confuses
+	 * some routers/firewalls, and they drop the packet.  So we need to
+	 * build the eth header using the original destination's MAC as the
+	 * source, and send the RST packet directly.
+	 */
+	if (oldskb->nf_bridge) {
+		struct ethhdr *oeth = eth_hdr(oldskb);
+		nskb->dev = oldskb->nf_bridge->physindev;
+		niph->tot_len = htons(nskb->len);
+		ip_send_check(niph);
+		if (dev_hard_header(nskb, nskb->dev, ntohs(nskb->protocol),
+				    oeth->h_source, oeth->h_dest, nskb->len) < 0)
+			goto free_nskb;
+		dev_queue_xmit(nskb);
+	} else
+#endif
+		ip_local_out(nskb);
+
 	return;
 
  free_nskb: