Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf-next

Pablo Neira Ayuso says:

====================
The following batch contains Netfilter/IPVS updates for net-next,
they are:

* Enforce policy to several nfnetlink subsystem, from Daniel
  Borkmann.

* Use xt_socket to match the third packet (to perform simplistic
  socket-based stateful filtering), from Eric Dumazet.

* Avoid large timeout for picked up from the middle TCP flows,
  from Florian Westphal.

* Exclude IPVS from struct net if IPVS is disabled and removal
  of unnecessary included header file, from JunweiZhang.

* Release SCTP connection immediately under load, to mimic current
  TCP behaviour, from Julian Anastasov.

* Replace and enhance SCTP state machine, from Julian Anastasov.

* Add tweak to reduce sync traffic in the presence of persistence,
  also from Julian Anastasov.

* Add tweak for the IPVS SH scheduler not to reject connections
  directed to a server, choose a new one instead, from Alexander
  Frolkin.

* Add support for sloppy TCP and SCTP modes, that creates state
  information on any packet, not only initial handshake packets,
  from Alexander Frolkin.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Documentation/networking/ipvs-sysctl.txt

@@ -181,6 +181,19 @@ snat_reroute - BOOLEAN
 	always be the same as the original route so it is an optimisation
 	to disable snat_reroute and avoid the recalculation.
 
+sync_persist_mode - INTEGER
+	default 0
+
+	Controls the synchronisation of connections when using persistence
+
+	0: All types of connections are synchronised
+	1: Attempt to reduce the synchronisation traffic depending on
+	the connection type. For persistent services avoid synchronisation
+	for normal connections, do it only for persistence templates.
+	In such case, for TCP and SCTP it may need enabling sloppy_tcp and
+	sloppy_sctp flags on backup servers. For non-persistent services
+	such optimization is not applied, mode 0 is assumed.
+
 sync_version - INTEGER
 	default 1
 

include/net/ip_vs.h

@@ -197,31 +197,6 @@ ip_vs_fill_iph_skb(int af, const struct sk_buff *skb, struct ip_vs_iphdr *iphdr)
 	}
 }
 
-/* This function is a faster version of ip_vs_fill_iph_skb().
- * Where we only populate {s,d}addr (and avoid calling ipv6_find_hdr()).
- * This is used by the some of the ip_vs_*_schedule() functions.
- * (Mostly done to avoid ABI breakage of external schedulers)
- */
-static inline void
-ip_vs_fill_iph_addr_only(int af, const struct sk_buff *skb,
-			 struct ip_vs_iphdr *iphdr)
-{
-#ifdef CONFIG_IP_VS_IPV6
-	if (af == AF_INET6) {
-		const struct ipv6hdr *iph =
-			(struct ipv6hdr *)skb_network_header(skb);
-		iphdr->saddr.in6 = iph->saddr;
-		iphdr->daddr.in6 = iph->daddr;
-	} else
-#endif
-	{
-		const struct iphdr *iph =
-			(struct iphdr *)skb_network_header(skb);
-		iphdr->saddr.ip = iph->saddr;
-		iphdr->daddr.ip = iph->daddr;
-	}
-}
-
 static inline void ip_vs_addr_copy(int af, union nf_inet_addr *dst,
 				   const union nf_inet_addr *src)
 {
@@ -405,17 +380,18 @@ enum {
  */
 enum ip_vs_sctp_states {
 	IP_VS_SCTP_S_NONE,
-	IP_VS_SCTP_S_INIT_CLI,
-	IP_VS_SCTP_S_INIT_SER,
-	IP_VS_SCTP_S_INIT_ACK_CLI,
-	IP_VS_SCTP_S_INIT_ACK_SER,
-	IP_VS_SCTP_S_ECHO_CLI,
-	IP_VS_SCTP_S_ECHO_SER,
+	IP_VS_SCTP_S_INIT1,
+	IP_VS_SCTP_S_INIT,
+	IP_VS_SCTP_S_COOKIE_SENT,
+	IP_VS_SCTP_S_COOKIE_REPLIED,
+	IP_VS_SCTP_S_COOKIE_WAIT,
+	IP_VS_SCTP_S_COOKIE,
+	IP_VS_SCTP_S_COOKIE_ECHOED,
 	IP_VS_SCTP_S_ESTABLISHED,
-	IP_VS_SCTP_S_SHUT_CLI,
-	IP_VS_SCTP_S_SHUT_SER,
-	IP_VS_SCTP_S_SHUT_ACK_CLI,
-	IP_VS_SCTP_S_SHUT_ACK_SER,
+	IP_VS_SCTP_S_SHUTDOWN_SENT,
+	IP_VS_SCTP_S_SHUTDOWN_RECEIVED,
+	IP_VS_SCTP_S_SHUTDOWN_ACK_SENT,
+	IP_VS_SCTP_S_REJECTED,
 	IP_VS_SCTP_S_CLOSED,
 	IP_VS_SCTP_S_LAST
 };
@@ -814,7 +790,8 @@ struct ip_vs_scheduler {
 
 	/* selecting a server from the given service */
 	struct ip_vs_dest* (*schedule)(struct ip_vs_service *svc,
-				       const struct sk_buff *skb);
+				       const struct sk_buff *skb,
+				       struct ip_vs_iphdr *iph);
 };
 
 /* The persistence engine object */
@@ -998,10 +975,13 @@ struct netns_ipvs {
 	int			sysctl_snat_reroute;
 	int			sysctl_sync_ver;
 	int			sysctl_sync_ports;
+	int			sysctl_sync_persist_mode;
 	unsigned long		sysctl_sync_qlen_max;
 	int			sysctl_sync_sock_size;
 	int			sysctl_cache_bypass;
 	int			sysctl_expire_nodest_conn;
+	int			sysctl_sloppy_tcp;
+	int			sysctl_sloppy_sctp;
 	int			sysctl_expire_quiescent_template;
 	int			sysctl_sync_threshold[2];
 	unsigned int		sysctl_sync_refresh_period;
@@ -1044,6 +1024,8 @@ struct netns_ipvs {
 #define DEFAULT_SYNC_THRESHOLD	3
 #define DEFAULT_SYNC_PERIOD	50
 #define DEFAULT_SYNC_VER	1
+#define DEFAULT_SLOPPY_TCP	0
+#define DEFAULT_SLOPPY_SCTP	0
 #define DEFAULT_SYNC_REFRESH_PERIOD	(0U * HZ)
 #define DEFAULT_SYNC_RETRIES		0
 #define IPVS_SYNC_WAKEUP_RATE	8
@@ -1080,11 +1062,26 @@ static inline int sysctl_sync_ver(struct netns_ipvs *ipvs)
 	return ipvs->sysctl_sync_ver;
 }
 
+static inline int sysctl_sloppy_tcp(struct netns_ipvs *ipvs)
+{
+	return ipvs->sysctl_sloppy_tcp;
+}
+
+static inline int sysctl_sloppy_sctp(struct netns_ipvs *ipvs)
+{
+	return ipvs->sysctl_sloppy_sctp;
+}
+
 static inline int sysctl_sync_ports(struct netns_ipvs *ipvs)
 {
 	return ACCESS_ONCE(ipvs->sysctl_sync_ports);
 }
 
+static inline int sysctl_sync_persist_mode(struct netns_ipvs *ipvs)
+{
+	return ipvs->sysctl_sync_persist_mode;
+}
+
 static inline unsigned long sysctl_sync_qlen_max(struct netns_ipvs *ipvs)
 {
 	return ipvs->sysctl_sync_qlen_max;
@@ -1133,11 +1130,26 @@ static inline int sysctl_sync_ver(struct netns_ipvs *ipvs)
 	return DEFAULT_SYNC_VER;
 }
 
+static inline int sysctl_sloppy_tcp(struct netns_ipvs *ipvs)
+{
+	return DEFAULT_SLOPPY_TCP;
+}
+
+static inline int sysctl_sloppy_sctp(struct netns_ipvs *ipvs)
+{
+	return DEFAULT_SLOPPY_SCTP;
+}
+
 static inline int sysctl_sync_ports(struct netns_ipvs *ipvs)
 {
 	return 1;
 }
 
+static inline int sysctl_sync_persist_mode(struct netns_ipvs *ipvs)
+{
+	return 0;
+}
+
 static inline unsigned long sysctl_sync_qlen_max(struct netns_ipvs *ipvs)
 {
 	return IPVS_SYNC_QLEN_MAX;

include/net/net_namespace.h

@@ -115,7 +115,9 @@ struct net {
 #ifdef CONFIG_XFRM
 	struct netns_xfrm	xfrm;
 #endif
+#if IS_ENABLED(CONFIG_IP_VS)
 	struct netns_ipvs	*ipvs;
+#endif
 	struct sock		*diag_nlsk;
 	atomic_t		rt_genid;
 	atomic_t		fnhe_genid;

include/uapi/linux/ip_vs.h

@@ -20,6 +20,12 @@
 #define IP_VS_SVC_F_PERSISTENT	0x0001		/* persistent port */
 #define IP_VS_SVC_F_HASHED	0x0002		/* hashed entry */
 #define IP_VS_SVC_F_ONEPACKET	0x0004		/* one-packet scheduling */
+#define IP_VS_SVC_F_SCHED1	0x0008		/* scheduler flag 1 */
+#define IP_VS_SVC_F_SCHED2	0x0010		/* scheduler flag 2 */
+#define IP_VS_SVC_F_SCHED3	0x0020		/* scheduler flag 3 */
+
+#define IP_VS_SVC_F_SCHED_SH_FALLBACK	IP_VS_SVC_F_SCHED1 /* SH fallback */
+#define IP_VS_SVC_F_SCHED_SH_PORT	IP_VS_SVC_F_SCHED2 /* SH use port */
 
 /*
  *      Destination Server Flags

include/uapi/linux/netfilter/nfnetlink_queue.h

@@ -105,5 +105,7 @@ enum nfqnl_attr_config {
 #define NFQA_SKB_CSUMNOTREADY (1 << 0)
 /* packet is GSO (i.e., exceeds device mtu) */
 #define NFQA_SKB_GSO (1 << 1)
+/* csum not validated (incoming device doesn't support hw checksum, etc.) */
+#define NFQA_SKB_CSUM_NOTVERIFIED (1 << 2)
 
 #endif /* _NFNETLINK_QUEUE_H */

include/uapi/linux/netfilter/xt_socket.h

@@ -5,10 +5,17 @@
 
 enum {
 	XT_SOCKET_TRANSPARENT = 1 << 0,
+	XT_SOCKET_NOWILDCARD = 1 << 1,
 };
 
 struct xt_socket_mtinfo1 {
 	__u8 flags;
 };
+#define XT_SOCKET_FLAGS_V1 XT_SOCKET_TRANSPARENT
+
+struct xt_socket_mtinfo2 {
+	__u8 flags;
+};
+#define XT_SOCKET_FLAGS_V2 (XT_SOCKET_TRANSPARENT | XT_SOCKET_NOWILDCARD)
 
 #endif /* _XT_SOCKET_H */

kernel/sysctl_binary.c

@@ -3,7 +3,6 @@
 #include "../fs/xfs/xfs_sysctl.h"
 #include <linux/sunrpc/debug.h>
 #include <linux/string.h>
-#include <net/ip_vs.h>
 #include <linux/syscalls.h>
 #include <linux/namei.h>
 #include <linux/mount.h>

net/netfilter/ipvs/ip_vs_conn.c

@@ -1231,6 +1231,18 @@ void ip_vs_random_dropentry(struct net *net)
 				default:
 					continue;
 				}
+			} else if (cp->protocol == IPPROTO_SCTP) {
+				switch (cp->state) {
+				case IP_VS_SCTP_S_INIT1:
+				case IP_VS_SCTP_S_INIT:
+					break;
+				case IP_VS_SCTP_S_ESTABLISHED:
+					if (todrop_entry(cp))
+						break;
+					continue;
+				default:
+					continue;
+				}
 			} else {
 				if (!todrop_entry(cp))
 					continue;

net/netfilter/ipvs/ip_vs_core.c

@@ -305,7 +305,7 @@ ip_vs_sched_persist(struct ip_vs_service *svc,
 		 * return *ignored=0 i.e. ICMP and NF_DROP
 		 */
 		sched = rcu_dereference(svc->scheduler);
-		dest = sched->schedule(svc, skb);
+		dest = sched->schedule(svc, skb, iph);
 		if (!dest) {
 			IP_VS_DBG(1, "p-schedule: no dest found.\n");
 			kfree(param.pe_data);
@@ -452,7 +452,7 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb,
 	}
 
 	sched = rcu_dereference(svc->scheduler);
-	dest = sched->schedule(svc, skb);
+	dest = sched->schedule(svc, skb, iph);
 	if (dest == NULL) {
 		IP_VS_DBG(1, "Schedule: no dest found.\n");
 		return NULL;

net/netfilter/ipvs/ip_vs_ctl.c

@@ -1714,6 +1714,12 @@ static struct ctl_table vs_vars[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_do_sync_ports,
 	},
+	{
+		.procname	= "sync_persist_mode",
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
 	{
 		.procname	= "sync_qlen_max",
 		.maxlen		= sizeof(unsigned long),
@@ -1738,6 +1744,18 @@ static struct ctl_table vs_vars[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec,
 	},
+	{
+		.procname	= "sloppy_tcp",
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+	{
+		.procname	= "sloppy_sctp",
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
 	{
 		.procname	= "expire_quiescent_template",
 		.maxlen		= sizeof(int),
@@ -3717,12 +3735,15 @@ static int __net_init ip_vs_control_net_init_sysctl(struct net *net)
 	tbl[idx++].data = &ipvs->sysctl_sync_ver;
 	ipvs->sysctl_sync_ports = 1;
 	tbl[idx++].data = &ipvs->sysctl_sync_ports;
+	tbl[idx++].data = &ipvs->sysctl_sync_persist_mode;
 	ipvs->sysctl_sync_qlen_max = nr_free_buffer_pages() / 32;
 	tbl[idx++].data = &ipvs->sysctl_sync_qlen_max;
 	ipvs->sysctl_sync_sock_size = 0;
 	tbl[idx++].data = &ipvs->sysctl_sync_sock_size;
 	tbl[idx++].data = &ipvs->sysctl_cache_bypass;
 	tbl[idx++].data = &ipvs->sysctl_expire_nodest_conn;
+	tbl[idx++].data = &ipvs->sysctl_sloppy_tcp;
+	tbl[idx++].data = &ipvs->sysctl_sloppy_sctp;
 	tbl[idx++].data = &ipvs->sysctl_expire_quiescent_template;
 	ipvs->sysctl_sync_threshold[0] = DEFAULT_SYNC_THRESHOLD;
 	ipvs->sysctl_sync_threshold[1] = DEFAULT_SYNC_PERIOD;

net/netfilter/ipvs/ip_vs_dh.c

@@ -214,18 +214,16 @@ static inline int is_overloaded(struct ip_vs_dest *dest)
  *      Destination hashing scheduling
  */
 static struct ip_vs_dest *
-ip_vs_dh_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
+ip_vs_dh_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
+		  struct ip_vs_iphdr *iph)
 {
 	struct ip_vs_dest *dest;
 	struct ip_vs_dh_state *s;
-	struct ip_vs_iphdr iph;
-
-	ip_vs_fill_iph_addr_only(svc->af, skb, &iph);
 
 	IP_VS_DBG(6, "%s(): Scheduling...\n", __func__);
 
 	s = (struct ip_vs_dh_state *) svc->sched_data;
-	dest = ip_vs_dh_get(svc->af, s, &iph.daddr);
+	dest = ip_vs_dh_get(svc->af, s, &iph->daddr);
 	if (!dest
 	    || !(dest->flags & IP_VS_DEST_F_AVAILABLE)
 	    || atomic_read(&dest->weight) <= 0
@@ -235,7 +233,7 @@ ip_vs_dh_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
 	}
 
 	IP_VS_DBG_BUF(6, "DH: destination IP address %s --> server %s:%d\n",
-		      IP_VS_DBG_ADDR(svc->af, &iph.daddr),
+		      IP_VS_DBG_ADDR(svc->af, &iph->daddr),
 		      IP_VS_DBG_ADDR(svc->af, &dest->addr),
 		      ntohs(dest->port));
 

net/netfilter/ipvs/ip_vs_lblc.c

@@ -487,19 +487,17 @@ is_overloaded(struct ip_vs_dest *dest, struct ip_vs_service *svc)
  *    Locality-Based (weighted) Least-Connection scheduling
  */
 static struct ip_vs_dest *
-ip_vs_lblc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
+ip_vs_lblc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
+		    struct ip_vs_iphdr *iph)
 {
 	struct ip_vs_lblc_table *tbl = svc->sched_data;
-	struct ip_vs_iphdr iph;
 	struct ip_vs_dest *dest = NULL;
 	struct ip_vs_lblc_entry *en;
 
-	ip_vs_fill_iph_addr_only(svc->af, skb, &iph);
-
 	IP_VS_DBG(6, "%s(): Scheduling...\n", __func__);
 
 	/* First look in our cache */
-	en = ip_vs_lblc_get(svc->af, tbl, &iph.daddr);
+	en = ip_vs_lblc_get(svc->af, tbl, &iph->daddr);
 	if (en) {
 		/* We only hold a read lock, but this is atomic */
 		en->lastuse = jiffies;
@@ -529,12 +527,12 @@ ip_vs_lblc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
 	/* If we fail to create a cache entry, we'll just use the valid dest */
 	spin_lock_bh(&svc->sched_lock);
 	if (!tbl->dead)
-		ip_vs_lblc_new(tbl, &iph.daddr, dest);
+		ip_vs_lblc_new(tbl, &iph->daddr, dest);
 	spin_unlock_bh(&svc->sched_lock);
 
 out:
 	IP_VS_DBG_BUF(6, "LBLC: destination IP address %s --> server %s:%d\n",
-		      IP_VS_DBG_ADDR(svc->af, &iph.daddr),
+		      IP_VS_DBG_ADDR(svc->af, &iph->daddr),
 		      IP_VS_DBG_ADDR(svc->af, &dest->addr), ntohs(dest->port));
 
 	return dest;

net/netfilter/ipvs/ip_vs_lblcr.c

@@ -655,19 +655,17 @@ is_overloaded(struct ip_vs_dest *dest, struct ip_vs_service *svc)
  *    Locality-Based (weighted) Least-Connection scheduling
  */
 static struct ip_vs_dest *
-ip_vs_lblcr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
+ip_vs_lblcr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
+		     struct ip_vs_iphdr *iph)
 {
 	struct ip_vs_lblcr_table *tbl = svc->sched_data;
-	struct ip_vs_iphdr iph;
 	struct ip_vs_dest *dest;
 	struct ip_vs_lblcr_entry *en;
 
-	ip_vs_fill_iph_addr_only(svc->af, skb, &iph);
-
 	IP_VS_DBG(6, "%s(): Scheduling...\n", __func__);
 
 	/* First look in our cache */
-	en = ip_vs_lblcr_get(svc->af, tbl, &iph.daddr);
+	en = ip_vs_lblcr_get(svc->af, tbl, &iph->daddr);
 	if (en) {
 		en->lastuse = jiffies;
 
@@ -718,12 +716,12 @@ ip_vs_lblcr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
 	/* If we fail to create a cache entry, we'll just use the valid dest */
 	spin_lock_bh(&svc->sched_lock);
 	if (!tbl->dead)
-		ip_vs_lblcr_new(tbl, &iph.daddr, dest);
+		ip_vs_lblcr_new(tbl, &iph->daddr, dest);
 	spin_unlock_bh(&svc->sched_lock);
 
 out:
 	IP_VS_DBG_BUF(6, "LBLCR: destination IP address %s --> server %s:%d\n",
-		      IP_VS_DBG_ADDR(svc->af, &iph.daddr),
+		      IP_VS_DBG_ADDR(svc->af, &iph->daddr),
 		      IP_VS_DBG_ADDR(svc->af, &dest->addr), ntohs(dest->port));
 
 	return dest;

net/netfilter/ipvs/ip_vs_lc.c

@@ -26,7 +26,8 @@
  *	Least Connection scheduling
  */
 static struct ip_vs_dest *
-ip_vs_lc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
+ip_vs_lc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
+		  struct ip_vs_iphdr *iph)
 {
 	struct ip_vs_dest *dest, *least = NULL;
 	unsigned int loh = 0, doh;

net/netfilter/ipvs/ip_vs_nq.c

@@ -55,7 +55,8 @@ ip_vs_nq_dest_overhead(struct ip_vs_dest *dest)
  *	Weighted Least Connection scheduling
  */
 static struct ip_vs_dest *
-ip_vs_nq_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
+ip_vs_nq_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
+		  struct ip_vs_iphdr *iph)
 {
 	struct ip_vs_dest *dest, *least = NULL;
 	unsigned int loh = 0, doh;

net/netfilter/ipvs/ip_vs_proto_tcp.c

@@ -39,6 +39,7 @@ tcp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd,
 	struct net *net;
 	struct ip_vs_service *svc;
 	struct tcphdr _tcph, *th;
+	struct netns_ipvs *ipvs;
 
 	th = skb_header_pointer(skb, iph->len, sizeof(_tcph), &_tcph);
 	if (th == NULL) {
@@ -46,14 +47,15 @@ tcp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd,
 		return 0;
 	}
 	net = skb_net(skb);
+	ipvs = net_ipvs(net);
 	/* No !th->ack check to allow scheduling on SYN+ACK for Active FTP */
 	rcu_read_lock();
-	if (th->syn &&
+	if ((th->syn || sysctl_sloppy_tcp(ipvs)) && !th->rst &&
 	    (svc = ip_vs_service_find(net, af, skb->mark, iph->protocol,
 				      &iph->daddr, th->dest))) {
 		int ignored;
 
-		if (ip_vs_todrop(net_ipvs(net))) {
+		if (ip_vs_todrop(ipvs)) {
 			/*
 			 * It seems that we are very loaded.
 			 * We have to drop this packet :(
@@ -401,7 +403,7 @@ static struct tcp_states_t tcp_states [] = {
 /*        sNO, sES, sSS, sSR, sFW, sTW, sCL, sCW, sLA, sLI, sSA	*/
 /*syn*/ {{sSR, sES, sES, sSR, sSR, sSR, sSR, sSR, sSR, sSR, sSR }},
 /*fin*/ {{sCL, sCW, sSS, sTW, sTW, sTW, sCL, sCW, sLA, sLI, sTW }},
-/*ack*/ {{sCL, sES, sSS, sES, sFW, sTW, sCL, sCW, sCL, sLI, sES }},
+/*ack*/ {{sES, sES, sSS, sES, sFW, sTW, sCL, sCW, sCL, sLI, sES }},
 /*rst*/ {{sCL, sCL, sCL, sSR, sCL, sCL, sCL, sCL, sLA, sLI, sSR }},
 
 /*	OUTPUT */
@@ -415,7 +417,7 @@ static struct tcp_states_t tcp_states [] = {
 /*        sNO, sES, sSS, sSR, sFW, sTW, sCL, sCW, sLA, sLI, sSA	*/
 /*syn*/ {{sSR, sES, sES, sSR, sSR, sSR, sSR, sSR, sSR, sSR, sSR }},
 /*fin*/ {{sCL, sFW, sSS, sTW, sFW, sTW, sCL, sCW, sLA, sLI, sTW }},
-/*ack*/ {{sCL, sES, sSS, sES, sFW, sTW, sCL, sCW, sCL, sLI, sES }},
+/*ack*/ {{sES, sES, sSS, sES, sFW, sTW, sCL, sCW, sCL, sLI, sES }},
 /*rst*/ {{sCL, sCL, sCL, sSR, sCL, sCL, sCL, sCL, sLA, sLI, sCL }},
 };
 
@@ -424,7 +426,7 @@ static struct tcp_states_t tcp_states_dos [] = {
 /*        sNO, sES, sSS, sSR, sFW, sTW, sCL, sCW, sLA, sLI, sSA	*/
 /*syn*/ {{sSR, sES, sES, sSR, sSR, sSR, sSR, sSR, sSR, sSR, sSA }},
 /*fin*/ {{sCL, sCW, sSS, sTW, sTW, sTW, sCL, sCW, sLA, sLI, sSA }},
-/*ack*/ {{sCL, sES, sSS, sSR, sFW, sTW, sCL, sCW, sCL, sLI, sSA }},
+/*ack*/ {{sES, sES, sSS, sSR, sFW, sTW, sCL, sCW, sCL, sLI, sSA }},
 /*rst*/ {{sCL, sCL, sCL, sSR, sCL, sCL, sCL, sCL, sLA, sLI, sCL }},
 
 /*	OUTPUT */
@@ -438,7 +440,7 @@ static struct tcp_states_t tcp_states_dos [] = {
 /*        sNO, sES, sSS, sSR, sFW, sTW, sCL, sCW, sLA, sLI, sSA	*/
 /*syn*/ {{sSA, sES, sES, sSR, sSA, sSA, sSA, sSA, sSA, sSA, sSA }},
 /*fin*/ {{sCL, sFW, sSS, sTW, sFW, sTW, sCL, sCW, sLA, sLI, sTW }},
-/*ack*/ {{sCL, sES, sSS, sES, sFW, sTW, sCL, sCW, sCL, sLI, sES }},
+/*ack*/ {{sES, sES, sSS, sES, sFW, sTW, sCL, sCW, sCL, sLI, sES }},
 /*rst*/ {{sCL, sCL, sCL, sSR, sCL, sCL, sCL, sCL, sLA, sLI, sCL }},
 };
 

net/netfilter/ipvs/ip_vs_rr.c

@@ -55,7 +55,8 @@ static int ip_vs_rr_del_dest(struct ip_vs_service *svc, struct ip_vs_dest *dest)
  * Round-Robin Scheduling
  */
 static struct ip_vs_dest *
-ip_vs_rr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
+ip_vs_rr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
+		  struct ip_vs_iphdr *iph)
 {
 	struct list_head *p;
 	struct ip_vs_dest *dest, *last;

net/netfilter/ipvs/ip_vs_sed.c

@@ -59,7 +59,8 @@ ip_vs_sed_dest_overhead(struct ip_vs_dest *dest)
  *	Weighted Least Connection scheduling
  */
 static struct ip_vs_dest *
-ip_vs_sed_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
+ip_vs_sed_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
+		   struct ip_vs_iphdr *iph)
 {
 	struct ip_vs_dest *dest, *least;
 	unsigned int loh, doh;

net/netfilter/ipvs/ip_vs_sh.c

@@ -48,6 +48,10 @@
 
 #include <net/ip_vs.h>
 
+#include <net/tcp.h>
+#include <linux/udp.h>
+#include <linux/sctp.h>
+
 
 /*
  *      IPVS SH bucket
@@ -71,10 +75,19 @@ struct ip_vs_sh_state {
 	struct ip_vs_sh_bucket		buckets[IP_VS_SH_TAB_SIZE];
 };
 
+/* Helper function to determine if server is unavailable */
+static inline bool is_unavailable(struct ip_vs_dest *dest)
+{
+	return atomic_read(&dest->weight) <= 0 ||
+	       dest->flags & IP_VS_DEST_F_OVERLOAD;
+}
+
 /*
  *	Returns hash value for IPVS SH entry
  */
-static inline unsigned int ip_vs_sh_hashkey(int af, const union nf_inet_addr *addr)
+static inline unsigned int
+ip_vs_sh_hashkey(int af, const union nf_inet_addr *addr,
+		 __be16 port, unsigned int offset)
 {
 	__be32 addr_fold = addr->ip;
 
@@ -83,7 +96,8 @@ static inline unsigned int ip_vs_sh_hashkey(int af, const union nf_inet_addr *ad
 		addr_fold = addr->ip6[0]^addr->ip6[1]^
 			    addr->ip6[2]^addr->ip6[3];
 #endif
-	return (ntohl(addr_fold)*2654435761UL) & IP_VS_SH_TAB_MASK;
+	return (offset + (ntohs(port) + ntohl(addr_fold))*2654435761UL) &
+		IP_VS_SH_TAB_MASK;
 }
 
 
@@ -91,12 +105,42 @@ static inline unsigned int ip_vs_sh_hashkey(int af, const union nf_inet_addr *ad
  *      Get ip_vs_dest associated with supplied parameters.
  */
 static inline struct ip_vs_dest *
-ip_vs_sh_get(int af, struct ip_vs_sh_state *s, const union nf_inet_addr *addr)
+ip_vs_sh_get(struct ip_vs_service *svc, struct ip_vs_sh_state *s,
+	     const union nf_inet_addr *addr, __be16 port)
 {
-	return rcu_dereference(s->buckets[ip_vs_sh_hashkey(af, addr)].dest);
+	unsigned int hash = ip_vs_sh_hashkey(svc->af, addr, port, 0);
+	struct ip_vs_dest *dest = rcu_dereference(s->buckets[hash].dest);
+
+	return (!dest || is_unavailable(dest)) ? NULL : dest;
 }
 
 
+/* As ip_vs_sh_get, but with fallback if selected server is unavailable */
+static inline struct ip_vs_dest *
+ip_vs_sh_get_fallback(struct ip_vs_service *svc, struct ip_vs_sh_state *s,
+		      const union nf_inet_addr *addr, __be16 port)
+{
+	unsigned int offset;
+	unsigned int hash;
+	struct ip_vs_dest *dest;
+
+	for (offset = 0; offset < IP_VS_SH_TAB_SIZE; offset++) {
+		hash = ip_vs_sh_hashkey(svc->af, addr, port, offset);
+		dest = rcu_dereference(s->buckets[hash].dest);
+		if (!dest)
+			break;
+		if (is_unavailable(dest))
+			IP_VS_DBG_BUF(6, "SH: selected unavailable server "
+				      "%s:%d (offset %d)",
+				      IP_VS_DBG_ADDR(svc->af, &dest->addr),
+				      ntohs(dest->port), offset);
+		else
+			return dest;
+	}
+
+	return NULL;
+}
+
 /*
  *      Assign all the hash buckets of the specified table with the service.
  */
@@ -213,13 +257,33 @@ static int ip_vs_sh_dest_changed(struct ip_vs_service *svc,
 }
 
 
-/*
- *      If the dest flags is set with IP_VS_DEST_F_OVERLOAD,
- *      consider that the server is overloaded here.
- */
-static inline int is_overloaded(struct ip_vs_dest *dest)
+/* Helper function to get port number */
+static inline __be16
+ip_vs_sh_get_port(const struct sk_buff *skb, struct ip_vs_iphdr *iph)
 {
-	return dest->flags & IP_VS_DEST_F_OVERLOAD;
+	__be16 port;
+	struct tcphdr _tcph, *th;
+	struct udphdr _udph, *uh;
+	sctp_sctphdr_t _sctph, *sh;
+
+	switch (iph->protocol) {
+	case IPPROTO_TCP:
+		th = skb_header_pointer(skb, iph->len, sizeof(_tcph), &_tcph);
+		port = th->source;
+		break;
+	case IPPROTO_UDP:
+		uh = skb_header_pointer(skb, iph->len, sizeof(_udph), &_udph);
+		port = uh->source;
+		break;
+	case IPPROTO_SCTP:
+		sh = skb_header_pointer(skb, iph->len, sizeof(_sctph), &_sctph);
+		port = sh->source;
+		break;
+	default:
+		port = 0;
+	}
+
+	return port;
 }
 
 
@@ -227,28 +291,32 @@ static inline int is_overloaded(struct ip_vs_dest *dest)
  *      Source Hashing scheduling
  */
 static struct ip_vs_dest *
-ip_vs_sh_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
+ip_vs_sh_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
+		  struct ip_vs_iphdr *iph)
 {
 	struct ip_vs_dest *dest;
 	struct ip_vs_sh_state *s;
-	struct ip_vs_iphdr iph;
-
-	ip_vs_fill_iph_addr_only(svc->af, skb, &iph);
+	__be16 port = 0;
 
 	IP_VS_DBG(6, "ip_vs_sh_schedule(): Scheduling...\n");
 
+	if (svc->flags & IP_VS_SVC_F_SCHED_SH_PORT)
+		port = ip_vs_sh_get_port(skb, iph);
+
 	s = (struct ip_vs_sh_state *) svc->sched_data;
-	dest = ip_vs_sh_get(svc->af, s, &iph.saddr);
-	if (!dest
-	    || !(dest->flags & IP_VS_DEST_F_AVAILABLE)
-	    || atomic_read(&dest->weight) <= 0
-	    || is_overloaded(dest)) {
+
+	if (svc->flags & IP_VS_SVC_F_SCHED_SH_FALLBACK)
+		dest = ip_vs_sh_get_fallback(svc, s, &iph->saddr, port);
+	else
+		dest = ip_vs_sh_get(svc, s, &iph->saddr, port);
+
+	if (!dest) {
 		ip_vs_scheduler_err(svc, "no destination available");
 		return NULL;
 	}
 
 	IP_VS_DBG_BUF(6, "SH: source IP address %s --> server %s:%d\n",
-		      IP_VS_DBG_ADDR(svc->af, &iph.saddr),
+		      IP_VS_DBG_ADDR(svc->af, &iph->saddr),
 		      IP_VS_DBG_ADDR(svc->af, &dest->addr),
 		      ntohs(dest->port));
 

net/netfilter/ipvs/ip_vs_sync.c

@@ -425,6 +425,16 @@ ip_vs_sync_buff_create_v0(struct netns_ipvs *ipvs)
 	return sb;
 }
 
+/* Check if connection is controlled by persistence */
+static inline bool in_persistence(struct ip_vs_conn *cp)
+{
+	for (cp = cp->control; cp; cp = cp->control) {
+		if (cp->flags & IP_VS_CONN_F_TEMPLATE)
+			return true;
+	}
+	return false;
+}
+
 /* Check if conn should be synced.
  * pkts: conn packets, use sysctl_sync_threshold to avoid packet check
  * - (1) sync_refresh_period: reduce sync rate. Additionally, retry
@@ -447,6 +457,8 @@ static int ip_vs_sync_conn_needed(struct netns_ipvs *ipvs,
 	/* Check if we sync in current state */
 	if (unlikely(cp->flags & IP_VS_CONN_F_TEMPLATE))
 		force = 0;
+	else if (unlikely(sysctl_sync_persist_mode(ipvs) && in_persistence(cp)))
+		return 0;
 	else if (likely(cp->protocol == IPPROTO_TCP)) {
 		if (!((1 << cp->state) &
 		      ((1 << IP_VS_TCP_S_ESTABLISHED) |
@@ -461,9 +473,10 @@ static int ip_vs_sync_conn_needed(struct netns_ipvs *ipvs,
 	} else if (unlikely(cp->protocol == IPPROTO_SCTP)) {
 		if (!((1 << cp->state) &
 		      ((1 << IP_VS_SCTP_S_ESTABLISHED) |
-		       (1 << IP_VS_SCTP_S_CLOSED) |
-		       (1 << IP_VS_SCTP_S_SHUT_ACK_CLI) |
-		       (1 << IP_VS_SCTP_S_SHUT_ACK_SER))))
+		       (1 << IP_VS_SCTP_S_SHUTDOWN_SENT) |
+		       (1 << IP_VS_SCTP_S_SHUTDOWN_RECEIVED) |
+		       (1 << IP_VS_SCTP_S_SHUTDOWN_ACK_SENT) |
+		       (1 << IP_VS_SCTP_S_CLOSED))))
 			return 0;
 		force = cp->state != cp->old_state;
 		if (force && cp->state != IP_VS_SCTP_S_ESTABLISHED)

net/netfilter/ipvs/ip_vs_wlc.c

@@ -31,7 +31,8 @@
  *	Weighted Least Connection scheduling
  */
 static struct ip_vs_dest *
-ip_vs_wlc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
+ip_vs_wlc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
+		   struct ip_vs_iphdr *iph)
 {
 	struct ip_vs_dest *dest, *least;
 	unsigned int loh, doh;

net/netfilter/ipvs/ip_vs_wrr.c

@@ -162,7 +162,8 @@ static int ip_vs_wrr_dest_changed(struct ip_vs_service *svc,
  *    Weighted Round-Robin Scheduling
  */
 static struct ip_vs_dest *
-ip_vs_wrr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
+ip_vs_wrr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
+		   struct ip_vs_iphdr *iph)
 {
 	struct ip_vs_dest *dest, *last, *stop = NULL;
 	struct ip_vs_wrr_mark *mark = svc->sched_data;

net/netfilter/nf_conntrack_netlink.c

@@ -828,7 +828,9 @@ ctnetlink_parse_tuple_ip(struct nlattr *attr, struct nf_conntrack_tuple *tuple)
 	struct nf_conntrack_l3proto *l3proto;
 	int ret = 0;
 
-	nla_parse_nested(tb, CTA_IP_MAX, attr, NULL);
+	ret = nla_parse_nested(tb, CTA_IP_MAX, attr, NULL);
+	if (ret < 0)
+		return ret;
 
 	rcu_read_lock();
 	l3proto = __nf_ct_l3proto_find(tuple->src.l3num);
@@ -895,7 +897,9 @@ ctnetlink_parse_tuple(const struct nlattr * const cda[],
 
 	memset(tuple, 0, sizeof(*tuple));
 
-	nla_parse_nested(tb, CTA_TUPLE_MAX, cda[type], tuple_nla_policy);
+	err = nla_parse_nested(tb, CTA_TUPLE_MAX, cda[type], tuple_nla_policy);
+	if (err < 0)
+		return err;
 
 	if (!tb[CTA_TUPLE_IP])
 		return -EINVAL;
@@ -946,9 +950,12 @@ static inline int
 ctnetlink_parse_help(const struct nlattr *attr, char **helper_name,
 		     struct nlattr **helpinfo)
 {
+	int err;
 	struct nlattr *tb[CTA_HELP_MAX+1];
 
-	nla_parse_nested(tb, CTA_HELP_MAX, attr, help_nla_policy);
+	err = nla_parse_nested(tb, CTA_HELP_MAX, attr, help_nla_policy);
+	if (err < 0)
+		return err;
 
 	if (!tb[CTA_HELP_NAME])
 		return -EINVAL;
@@ -1431,7 +1438,9 @@ ctnetlink_change_protoinfo(struct nf_conn *ct, const struct nlattr * const cda[]
 	struct nf_conntrack_l4proto *l4proto;
 	int err = 0;
 
-	nla_parse_nested(tb, CTA_PROTOINFO_MAX, attr, protoinfo_policy);
+	err = nla_parse_nested(tb, CTA_PROTOINFO_MAX, attr, protoinfo_policy);
+	if (err < 0)
+		return err;
 
 	rcu_read_lock();
 	l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct));
@@ -1452,9 +1461,12 @@ static const struct nla_policy nat_seq_policy[CTA_NAT_SEQ_MAX+1] = {
 static inline int
 change_nat_seq_adj(struct nf_nat_seq *natseq, const struct nlattr * const attr)
 {
+	int err;
 	struct nlattr *cda[CTA_NAT_SEQ_MAX+1];
 
-	nla_parse_nested(cda, CTA_NAT_SEQ_MAX, attr, nat_seq_policy);
+	err = nla_parse_nested(cda, CTA_NAT_SEQ_MAX, attr, nat_seq_policy);
+	if (err < 0)
+		return err;
 
 	if (!cda[CTA_NAT_SEQ_CORRECTION_POS])
 		return -EINVAL;
@@ -2115,7 +2127,9 @@ ctnetlink_nfqueue_parse(const struct nlattr *attr, struct nf_conn *ct)
 	struct nlattr *cda[CTA_MAX+1];
 	int ret;
 
-	nla_parse_nested(cda, CTA_MAX, attr, ct_nla_policy);
+	ret = nla_parse_nested(cda, CTA_MAX, attr, ct_nla_policy);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_bh(&nf_conntrack_lock);
 	ret = ctnetlink_nfqueue_parse_ct((const struct nlattr **)cda, ct);
@@ -2710,7 +2724,9 @@ ctnetlink_parse_expect_nat(const struct nlattr *attr,
 	struct nf_conntrack_tuple nat_tuple = {};
 	int err;
 
-	nla_parse_nested(tb, CTA_EXPECT_NAT_MAX, attr, exp_nat_nla_policy);
+	err = nla_parse_nested(tb, CTA_EXPECT_NAT_MAX, attr, exp_nat_nla_policy);
+	if (err < 0)
+		return err;
 
 	if (!tb[CTA_EXPECT_NAT_DIR] || !tb[CTA_EXPECT_NAT_TUPLE])
 		return -EINVAL;

net/netfilter/nf_conntrack_proto_tcp.c

@@ -1043,6 +1043,12 @@ static int tcp_packet(struct nf_conn *ct,
 			nf_ct_kill_acct(ct, ctinfo, skb);
 			return NF_ACCEPT;
 		}
+		/* ESTABLISHED without SEEN_REPLY, i.e. mid-connection
+		 * pickup with loose=1. Avoid large ESTABLISHED timeout.
+		 */
+		if (new_state == TCP_CONNTRACK_ESTABLISHED &&
+		    timeout > timeouts[TCP_CONNTRACK_UNACK])
+			timeout = timeouts[TCP_CONNTRACK_UNACK];
 	} else if (!test_bit(IPS_ASSURED_BIT, &ct->status)
 		   && (old_state == TCP_CONNTRACK_SYN_RECV
 		       || old_state == TCP_CONNTRACK_ESTABLISHED)

net/netfilter/nfnetlink_cthelper.c

@@ -67,9 +67,12 @@ static int
 nfnl_cthelper_parse_tuple(struct nf_conntrack_tuple *tuple,
 			  const struct nlattr *attr)
 {
+	int err;
 	struct nlattr *tb[NFCTH_TUPLE_MAX+1];
 
-	nla_parse_nested(tb, NFCTH_TUPLE_MAX, attr, nfnl_cthelper_tuple_pol);
+	err = nla_parse_nested(tb, NFCTH_TUPLE_MAX, attr, nfnl_cthelper_tuple_pol);
+	if (err < 0)
+		return err;
 
 	if (!tb[NFCTH_TUPLE_L3PROTONUM] || !tb[NFCTH_TUPLE_L4PROTONUM])
 		return -EINVAL;
@@ -121,9 +124,12 @@ static int
 nfnl_cthelper_expect_policy(struct nf_conntrack_expect_policy *expect_policy,
 			    const struct nlattr *attr)
 {
+	int err;
 	struct nlattr *tb[NFCTH_POLICY_MAX+1];
 
-	nla_parse_nested(tb, NFCTH_POLICY_MAX, attr, nfnl_cthelper_expect_pol);
+	err = nla_parse_nested(tb, NFCTH_POLICY_MAX, attr, nfnl_cthelper_expect_pol);
+	if (err < 0)
+		return err;
 
 	if (!tb[NFCTH_POLICY_NAME] ||
 	    !tb[NFCTH_POLICY_EXPECT_MAX] ||
@@ -153,8 +159,10 @@ nfnl_cthelper_parse_expect_policy(struct nf_conntrack_helper *helper,
 	struct nf_conntrack_expect_policy *expect_policy;
 	struct nlattr *tb[NFCTH_POLICY_SET_MAX+1];
 
-	nla_parse_nested(tb, NFCTH_POLICY_SET_MAX, attr,
-					nfnl_cthelper_expect_policy_set);
+	ret = nla_parse_nested(tb, NFCTH_POLICY_SET_MAX, attr,
+			       nfnl_cthelper_expect_policy_set);
+	if (ret < 0)
+		return ret;
 
 	if (!tb[NFCTH_POLICY_SET_NUM])
 		return -EINVAL;

net/netfilter/nfnetlink_cttimeout.c

@@ -59,8 +59,10 @@ ctnl_timeout_parse_policy(struct ctnl_timeout *timeout,
 	if (likely(l4proto->ctnl_timeout.nlattr_to_obj)) {
 		struct nlattr *tb[l4proto->ctnl_timeout.nlattr_max+1];
 
-		nla_parse_nested(tb, l4proto->ctnl_timeout.nlattr_max,
-				 attr, l4proto->ctnl_timeout.nla_policy);
+		ret = nla_parse_nested(tb, l4proto->ctnl_timeout.nlattr_max,
+				       attr, l4proto->ctnl_timeout.nla_policy);
+		if (ret < 0)
+			return ret;
 
 		ret = l4proto->ctnl_timeout.nlattr_to_obj(tb, net,
 							  &timeout->data);

net/netfilter/nfnetlink_queue_core.c

@@ -280,12 +280,17 @@ nfqnl_zcopy(struct sk_buff *to, const struct sk_buff *from, int len, int hlen)
 	skb_shinfo(to)->nr_frags = j;
 }
 
-static int nfqnl_put_packet_info(struct sk_buff *nlskb, struct sk_buff *packet)
+static int
+nfqnl_put_packet_info(struct sk_buff *nlskb, struct sk_buff *packet,
+		      bool csum_verify)
 {
 	__u32 flags = 0;
 
 	if (packet->ip_summed == CHECKSUM_PARTIAL)
 		flags = NFQA_SKB_CSUMNOTREADY;
+	else if (csum_verify)
+		flags = NFQA_SKB_CSUM_NOTVERIFIED;
+
 	if (skb_is_gso(packet))
 		flags |= NFQA_SKB_GSO;
 
@@ -310,6 +315,7 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue,
 	struct net_device *outdev;
 	struct nf_conn *ct = NULL;
 	enum ip_conntrack_info uninitialized_var(ctinfo);
+	bool csum_verify;
 
 	size =    nlmsg_total_size(sizeof(struct nfgenmsg))
 		+ nla_total_size(sizeof(struct nfqnl_msg_packet_hdr))
@@ -327,6 +333,12 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue,
 	if (entskb->tstamp.tv64)
 		size += nla_total_size(sizeof(struct nfqnl_msg_packet_timestamp));
 
+	if (entry->hook <= NF_INET_FORWARD ||
+	   (entry->hook == NF_INET_POST_ROUTING && entskb->sk == NULL))
+		csum_verify = !skb_csum_unnecessary(entskb);
+	else
+		csum_verify = false;
+
 	outdev = entry->outdev;
 
 	switch ((enum nfqnl_config_mode)ACCESS_ONCE(queue->copy_mode)) {
@@ -476,7 +488,7 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue,
 	    nla_put_be32(skb, NFQA_CAP_LEN, htonl(cap_len)))
 		goto nla_put_failure;
 
-	if (nfqnl_put_packet_info(skb, entskb))
+	if (nfqnl_put_packet_info(skb, entskb, csum_verify))
 		goto nla_put_failure;
 
 	if (data_len) {

net/netfilter/xt_socket.c

@@ -163,8 +163,11 @@ socket_match(const struct sk_buff *skb, struct xt_action_param *par,
 		bool wildcard;
 		bool transparent = true;
 
-		/* Ignore sockets listening on INADDR_ANY */
-		wildcard = (sk->sk_state != TCP_TIME_WAIT &&
+		/* Ignore sockets listening on INADDR_ANY,
+		 * unless XT_SOCKET_NOWILDCARD is set
+		 */
+		wildcard = (!(info->flags & XT_SOCKET_NOWILDCARD) &&
+			    sk->sk_state != TCP_TIME_WAIT &&
 			    inet_sk(sk)->inet_rcv_saddr == 0);
 
 		/* Ignore non-transparent sockets,
@@ -197,7 +200,7 @@ socket_mt4_v0(const struct sk_buff *skb, struct xt_action_param *par)
 }
 
 static bool
-socket_mt4_v1(const struct sk_buff *skb, struct xt_action_param *par)
+socket_mt4_v1_v2(const struct sk_buff *skb, struct xt_action_param *par)
 {
 	return socket_match(skb, par, par->matchinfo);
 }
@@ -259,7 +262,7 @@ extract_icmp6_fields(const struct sk_buff *skb,
 }
 
 static bool
-socket_mt6_v1(const struct sk_buff *skb, struct xt_action_param *par)
+socket_mt6_v1_v2(const struct sk_buff *skb, struct xt_action_param *par)
 {
 	struct ipv6hdr *iph = ipv6_hdr(skb);
 	struct udphdr _hdr, *hp = NULL;
@@ -302,8 +305,11 @@ socket_mt6_v1(const struct sk_buff *skb, struct xt_action_param *par)
 		bool wildcard;
 		bool transparent = true;
 
-		/* Ignore sockets listening on INADDR_ANY */
-		wildcard = (sk->sk_state != TCP_TIME_WAIT &&
+		/* Ignore sockets listening on INADDR_ANY
+		 * unless XT_SOCKET_NOWILDCARD is set
+		 */
+		wildcard = (!(info->flags & XT_SOCKET_NOWILDCARD) &&
+			    sk->sk_state != TCP_TIME_WAIT &&
 			    ipv6_addr_any(&inet6_sk(sk)->rcv_saddr));
 
 		/* Ignore non-transparent sockets,
@@ -331,6 +337,28 @@ socket_mt6_v1(const struct sk_buff *skb, struct xt_action_param *par)
 }
 #endif
 
+static int socket_mt_v1_check(const struct xt_mtchk_param *par)
+{
+	const struct xt_socket_mtinfo1 *info = (struct xt_socket_mtinfo1 *) par->matchinfo;
+
+	if (info->flags & ~XT_SOCKET_FLAGS_V1) {
+		pr_info("unknown flags 0x%x\n", info->flags & ~XT_SOCKET_FLAGS_V1);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int socket_mt_v2_check(const struct xt_mtchk_param *par)
+{
+	const struct xt_socket_mtinfo2 *info = (struct xt_socket_mtinfo2 *) par->matchinfo;
+
+	if (info->flags & ~XT_SOCKET_FLAGS_V2) {
+		pr_info("unknown flags 0x%x\n", info->flags & ~XT_SOCKET_FLAGS_V2);
+		return -EINVAL;
+	}
+	return 0;
+}
+
 static struct xt_match socket_mt_reg[] __read_mostly = {
 	{
 		.name		= "socket",
@@ -345,7 +373,8 @@ static struct xt_match socket_mt_reg[] __read_mostly = {
 		.name		= "socket",
 		.revision	= 1,
 		.family		= NFPROTO_IPV4,
-		.match		= socket_mt4_v1,
+		.match		= socket_mt4_v1_v2,
+		.checkentry	= socket_mt_v1_check,
 		.matchsize	= sizeof(struct xt_socket_mtinfo1),
 		.hooks		= (1 << NF_INET_PRE_ROUTING) |
 				  (1 << NF_INET_LOCAL_IN),
@@ -356,7 +385,32 @@ static struct xt_match socket_mt_reg[] __read_mostly = {
 		.name		= "socket",
 		.revision	= 1,
 		.family		= NFPROTO_IPV6,
-		.match		= socket_mt6_v1,
+		.match		= socket_mt6_v1_v2,
+		.checkentry	= socket_mt_v1_check,
+		.matchsize	= sizeof(struct xt_socket_mtinfo1),
+		.hooks		= (1 << NF_INET_PRE_ROUTING) |
+				  (1 << NF_INET_LOCAL_IN),
+		.me		= THIS_MODULE,
+	},
+#endif
+	{
+		.name		= "socket",
+		.revision	= 2,
+		.family		= NFPROTO_IPV4,
+		.match		= socket_mt4_v1_v2,
+		.checkentry	= socket_mt_v2_check,
+		.matchsize	= sizeof(struct xt_socket_mtinfo1),
+		.hooks		= (1 << NF_INET_PRE_ROUTING) |
+				  (1 << NF_INET_LOCAL_IN),
+		.me		= THIS_MODULE,
+	},
+#ifdef XT_SOCKET_HAVE_IPV6
+	{
+		.name		= "socket",
+		.revision	= 2,
+		.family		= NFPROTO_IPV6,
+		.match		= socket_mt6_v1_v2,
+		.checkentry	= socket_mt_v2_check,
 		.matchsize	= sizeof(struct xt_socket_mtinfo1),
 		.hooks		= (1 << NF_INET_PRE_ROUTING) |
 				  (1 << NF_INET_LOCAL_IN),