X-Git-Url: https://gerrit.opnfv.org/gerrit/gitweb?a=blobdiff_plain;f=kernel%2Fnet%2Fnetfilter%2Fipvs%2Fip_vs_core.c;h=4da560005b0e9168760c3e76825f18551d60d238;hb=52f993b8e89487ec9ee15a7fb4979e0f09a45b27;hp=5d2b806a862e6834ff6c61aee5c0e0a899bbe4b8;hpb=9ca8dbcc65cfc63d6f5ef3312a33184e1d726e00;p=kvmfornfv.git

diff --git a/kernel/net/netfilter/ipvs/ip_vs_core.c b/kernel/net/netfilter/ipvs/ip_vs_core.c
index 5d2b806a8..4da560005 100644
--- a/kernel/net/netfilter/ipvs/ip_vs_core.c
+++ b/kernel/net/netfilter/ipvs/ip_vs_core.c
@@ -112,7 +112,7 @@ static inline void
 ip_vs_in_stats(struct ip_vs_conn *cp, struct sk_buff *skb)
 {
 	struct ip_vs_dest *dest = cp->dest;
-	struct netns_ipvs *ipvs = net_ipvs(skb_net(skb));
+	struct netns_ipvs *ipvs = cp->ipvs;
 
 	if (dest && (dest->flags & IP_VS_DEST_F_AVAILABLE)) {
 		struct ip_vs_cpu_stats *s;
@@ -146,7 +146,7 @@ static inline void
 ip_vs_out_stats(struct ip_vs_conn *cp, struct sk_buff *skb)
 {
 	struct ip_vs_dest *dest = cp->dest;
-	struct netns_ipvs *ipvs = net_ipvs(skb_net(skb));
+	struct netns_ipvs *ipvs = cp->ipvs;
 
 	if (dest && (dest->flags & IP_VS_DEST_F_AVAILABLE)) {
 		struct ip_vs_cpu_stats *s;
@@ -179,7 +179,7 @@ ip_vs_out_stats(struct ip_vs_conn *cp, struct sk_buff *skb)
 static inline void
 ip_vs_conn_stats(struct ip_vs_conn *cp, struct ip_vs_service *svc)
 {
-	struct netns_ipvs *ipvs = net_ipvs(svc->net);
+	struct netns_ipvs *ipvs = svc->ipvs;
 	struct ip_vs_cpu_stats *s;
 
 	s = this_cpu_ptr(cp->dest->stats.cpustats);
@@ -215,7 +215,7 @@ ip_vs_conn_fill_param_persist(const struct ip_vs_service *svc,
 			      const union nf_inet_addr *vaddr, __be16 vport,
 			      struct ip_vs_conn_param *p)
 {
-	ip_vs_conn_fill_param(svc->net, svc->af, protocol, caddr, cport, vaddr,
+	ip_vs_conn_fill_param(svc->ipvs, svc->af, protocol, caddr, cport, vaddr,
 			      vport, p);
 	p->pe = rcu_dereference(svc->pe);
 	if (p->pe && p->pe->fill_param)
@@ -245,20 +245,30 @@ ip_vs_sched_persist(struct ip_vs_service *svc,
 	const union nf_inet_addr fwmark = { .ip = htonl(svc->fwmark) };
 	union nf_inet_addr snet;	/* source network of the client,
 					   after masking */
+	const union nf_inet_addr *src_addr, *dst_addr;
+
+	if (likely(!ip_vs_iph_inverse(iph))) {
+		src_addr = &iph->saddr;
+		dst_addr = &iph->daddr;
+	} else {
+		src_addr = &iph->daddr;
+		dst_addr = &iph->saddr;
+	}
+
 
 	/* Mask saddr with the netmask to adjust template granularity */
 #ifdef CONFIG_IP_VS_IPV6
 	if (svc->af == AF_INET6)
-		ipv6_addr_prefix(&snet.in6, &iph->saddr.in6,
+		ipv6_addr_prefix(&snet.in6, &src_addr->in6,
 				 (__force __u32) svc->netmask);
 	else
 #endif
-		snet.ip = iph->saddr.ip & svc->netmask;
+		snet.ip = src_addr->ip & svc->netmask;
 
 	IP_VS_DBG_BUF(6, "p-schedule: src %s:%u dest %s:%u "
 		      "mnet %s\n",
-		      IP_VS_DBG_ADDR(svc->af, &iph->saddr), ntohs(src_port),
-		      IP_VS_DBG_ADDR(svc->af, &iph->daddr), ntohs(dst_port),
+		      IP_VS_DBG_ADDR(svc->af, src_addr), ntohs(src_port),
+		      IP_VS_DBG_ADDR(svc->af, dst_addr), ntohs(dst_port),
 		      IP_VS_DBG_ADDR(svc->af, &snet));
 
 	/*
@@ -276,7 +286,7 @@ ip_vs_sched_persist(struct ip_vs_service *svc,
 	 */
 	{
 		int protocol = iph->protocol;
-		const union nf_inet_addr *vaddr = &iph->daddr;
+		const union nf_inet_addr *vaddr = dst_addr;
 		__be16 vport = 0;
 
 		if (dst_port == svc->port) {
@@ -319,7 +329,13 @@ 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, iph);
+		if (sched) {
+			/* read svc->sched_data after svc->scheduler */
+			smp_rmb();
+			dest = sched->schedule(svc, skb, iph);
+		} else {
+			dest = NULL;
+		}
 		if (!dest) {
 			IP_VS_DBG(1, "p-schedule: no dest found.\n");
 			kfree(param.pe_data);
@@ -360,8 +376,8 @@ ip_vs_sched_persist(struct ip_vs_service *svc,
 	/*
 	 *    Create a new connection according to the template
 	 */
-	ip_vs_conn_fill_param(svc->net, svc->af, iph->protocol, &iph->saddr,
-			      src_port, &iph->daddr, dst_port, &param);
+	ip_vs_conn_fill_param(svc->ipvs, svc->af, iph->protocol, src_addr,
+			      src_port, dst_addr, dst_port, &param);
 
 	cp = ip_vs_conn_new(&param, dest->af, &dest->addr, dport, flags, dest,
 			    skb->mark);
@@ -412,7 +428,8 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb,
 	struct ip_vs_conn *cp = NULL;
 	struct ip_vs_scheduler *sched;
 	struct ip_vs_dest *dest;
-	__be16 _ports[2], *pptr;
+	__be16 _ports[2], *pptr, cport, vport;
+	const void *caddr, *vaddr;
 	unsigned int flags;
 
 	*ignored = 1;
@@ -423,14 +440,26 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb,
 	if (pptr == NULL)
 		return NULL;
 
+	if (likely(!ip_vs_iph_inverse(iph))) {
+		cport = pptr[0];
+		caddr = &iph->saddr;
+		vport = pptr[1];
+		vaddr = &iph->daddr;
+	} else {
+		cport = pptr[1];
+		caddr = &iph->daddr;
+		vport = pptr[0];
+		vaddr = &iph->saddr;
+	}
+
 	/*
 	 * FTPDATA needs this check when using local real server.
 	 * Never schedule Active FTPDATA connections from real server.
 	 * For LVS-NAT they must be already created. For other methods
 	 * with persistence the connection is created on SYN+ACK.
 	 */
-	if (pptr[0] == FTPDATA) {
-		IP_VS_DBG_PKT(12, svc->af, pp, skb, 0,
+	if (cport == FTPDATA) {
+		IP_VS_DBG_PKT(12, svc->af, pp, skb, iph->off,
 			      "Not scheduling FTPDATA");
 		return NULL;
 	}
@@ -438,19 +467,25 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb,
 	/*
 	 *    Do not schedule replies from local real server.
 	 */
-	if ((!skb->dev || skb->dev->flags & IFF_LOOPBACK) &&
-	    (cp = pp->conn_in_get(svc->af, skb, iph, 1))) {
-		IP_VS_DBG_PKT(12, svc->af, pp, skb, 0,
-			      "Not scheduling reply for existing connection");
-		__ip_vs_conn_put(cp);
-		return NULL;
+	if ((!skb->dev || skb->dev->flags & IFF_LOOPBACK)) {
+		iph->hdr_flags ^= IP_VS_HDR_INVERSE;
+		cp = pp->conn_in_get(svc->ipvs, svc->af, skb, iph);
+		iph->hdr_flags ^= IP_VS_HDR_INVERSE;
+
+		if (cp) {
+			IP_VS_DBG_PKT(12, svc->af, pp, skb, iph->off,
+				      "Not scheduling reply for existing"
+				      " connection");
+			__ip_vs_conn_put(cp);
+			return NULL;
+		}
 	}
 
 	/*
 	 *    Persistent service
 	 */
 	if (svc->flags & IP_VS_SVC_F_PERSISTENT)
-		return ip_vs_sched_persist(svc, skb, pptr[0], pptr[1], ignored,
+		return ip_vs_sched_persist(svc, skb, cport, vport, ignored,
 					   iph);
 
 	*ignored = 0;
@@ -458,7 +493,7 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb,
 	/*
 	 *    Non-persistent service
 	 */
-	if (!svc->fwmark && pptr[1] != svc->port) {
+	if (!svc->fwmark && vport != svc->port) {
 		if (!svc->port)
 			pr_err("Schedule: port zero only supported "
 			       "in persistent services, "
@@ -467,7 +502,13 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb,
 	}
 
 	sched = rcu_dereference(svc->scheduler);
-	dest = sched->schedule(svc, skb, iph);
+	if (sched) {
+		/* read svc->sched_data after svc->scheduler */
+		smp_rmb();
+		dest = sched->schedule(svc, skb, iph);
+	} else {
+		dest = NULL;
+	}
 	if (dest == NULL) {
 		IP_VS_DBG(1, "Schedule: no dest found.\n");
 		return NULL;
@@ -483,11 +524,10 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb,
 	{
 		struct ip_vs_conn_param p;
 
-		ip_vs_conn_fill_param(svc->net, svc->af, iph->protocol,
-				      &iph->saddr, pptr[0], &iph->daddr,
-				      pptr[1], &p);
+		ip_vs_conn_fill_param(svc->ipvs, svc->af, iph->protocol,
+				      caddr, cport, vaddr, vport, &p);
 		cp = ip_vs_conn_new(&p, dest->af, &dest->addr,
-				    dest->port ? dest->port : pptr[1],
+				    dest->port ? dest->port : vport,
 				    flags, dest, skb->mark);
 		if (!cp) {
 			*ignored = -1;
@@ -507,6 +547,15 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb,
 	return cp;
 }
 
+static inline int ip_vs_addr_is_unicast(struct net *net, int af,
+					union nf_inet_addr *addr)
+{
+#ifdef CONFIG_IP_VS_IPV6
+	if (af == AF_INET6)
+		return ipv6_addr_type(&addr->in6) & IPV6_ADDR_UNICAST;
+#endif
+	return (inet_addr_type(net, addr->ip) == RTN_UNICAST);
+}
 
 /*
  *  Pass or drop the packet.
@@ -516,33 +565,21 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb,
 int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb,
 		struct ip_vs_proto_data *pd, struct ip_vs_iphdr *iph)
 {
-	__be16 _ports[2], *pptr;
-#ifdef CONFIG_SYSCTL
-	struct net *net;
-	struct netns_ipvs *ipvs;
-	int unicast;
-#endif
+	__be16 _ports[2], *pptr, dport;
+	struct netns_ipvs *ipvs = svc->ipvs;
+	struct net *net = ipvs->net;
 
 	pptr = frag_safe_skb_hp(skb, iph->len, sizeof(_ports), _ports, iph);
-	if (pptr == NULL) {
+	if (!pptr)
 		return NF_DROP;
-	}
-
-#ifdef CONFIG_SYSCTL
-	net = skb_net(skb);
-
-#ifdef CONFIG_IP_VS_IPV6
-	if (svc->af == AF_INET6)
-		unicast = ipv6_addr_type(&iph->daddr.in6) & IPV6_ADDR_UNICAST;
-	else
-#endif
-		unicast = (inet_addr_type(net, iph->daddr.ip) == RTN_UNICAST);
+	dport = likely(!ip_vs_iph_inverse(iph)) ? pptr[1] : pptr[0];
 
 	/* if it is fwmark-based service, the cache_bypass sysctl is up
 	   and the destination is a non-local unicast, then create
 	   a cache_bypass connection entry */
-	ipvs = net_ipvs(net);
-	if (ipvs->sysctl_cache_bypass && svc->fwmark && unicast) {
+	if (sysctl_cache_bypass(ipvs) && svc->fwmark &&
+	    !(iph->hdr_flags & (IP_VS_HDR_INVERSE | IP_VS_HDR_ICMP)) &&
+	    ip_vs_addr_is_unicast(net, svc->af, &iph->daddr)) {
 		int ret;
 		struct ip_vs_conn *cp;
 		unsigned int flags = (svc->flags & IP_VS_SVC_F_ONEPACKET &&
@@ -554,7 +591,7 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb,
 		IP_VS_DBG(6, "%s(): create a cache_bypass entry\n", __func__);
 		{
 			struct ip_vs_conn_param p;
-			ip_vs_conn_fill_param(svc->net, svc->af, iph->protocol,
+			ip_vs_conn_fill_param(svc->ipvs, svc->af, iph->protocol,
 					      &iph->saddr, pptr[0],
 					      &iph->daddr, pptr[1], &p);
 			cp = ip_vs_conn_new(&p, svc->af, &daddr, 0,
@@ -578,7 +615,6 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb,
 		ip_vs_conn_put(cp);
 		return ret;
 	}
-#endif
 
 	/*
 	 * When the virtual ftp service is presented, packets destined
@@ -586,9 +622,12 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb,
 	 * listed in the ipvs table), pass the packets, because it is
 	 * not ipvs job to decide to drop the packets.
 	 */
-	if ((svc->port == FTPPORT) && (pptr[1] != FTPPORT))
+	if (svc->port == FTPPORT && dport != FTPPORT)
 		return NF_ACCEPT;
 
+	if (unlikely(ip_vs_iph_icmp(iph)))
+		return NF_DROP;
+
 	/*
 	 * Notify the client that the destination is unreachable, and
 	 * release the socket buffer.
@@ -598,11 +637,8 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb,
 	 */
 #ifdef CONFIG_IP_VS_IPV6
 	if (svc->af == AF_INET6) {
-		if (!skb->dev) {
-			struct net *net_ = dev_net(skb_dst(skb)->dev);
-
-			skb->dev = net_->loopback_dev;
-		}
+		if (!skb->dev)
+			skb->dev = net->loopback_dev;
 		icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0);
 	} else
 #endif
@@ -613,15 +649,13 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb,
 
 #ifdef CONFIG_SYSCTL
 
-static int sysctl_snat_reroute(struct sk_buff *skb)
+static int sysctl_snat_reroute(struct netns_ipvs *ipvs)
 {
-	struct netns_ipvs *ipvs = net_ipvs(skb_net(skb));
 	return ipvs->sysctl_snat_reroute;
 }
 
-static int sysctl_nat_icmp_send(struct net *net)
+static int sysctl_nat_icmp_send(struct netns_ipvs *ipvs)
 {
-	struct netns_ipvs *ipvs = net_ipvs(net);
 	return ipvs->sysctl_nat_icmp_send;
 }
 
@@ -632,8 +666,8 @@ static int sysctl_expire_nodest_conn(struct netns_ipvs *ipvs)
 
 #else
 
-static int sysctl_snat_reroute(struct sk_buff *skb) { return 0; }
-static int sysctl_nat_icmp_send(struct net *net) { return 0; }
+static int sysctl_snat_reroute(struct netns_ipvs *ipvs) { return 0; }
+static int sysctl_nat_icmp_send(struct netns_ipvs *ipvs) { return 0; }
 static int sysctl_expire_nodest_conn(struct netns_ipvs *ipvs) { return 0; }
 
 #endif
@@ -652,12 +686,13 @@ static inline enum ip_defrag_users ip_vs_defrag_user(unsigned int hooknum)
 	return IP_DEFRAG_VS_OUT;
 }
 
-static inline int ip_vs_gather_frags(struct sk_buff *skb, u_int32_t user)
+static inline int ip_vs_gather_frags(struct netns_ipvs *ipvs,
+				     struct sk_buff *skb, u_int32_t user)
 {
 	int err;
 
 	local_bh_disable();
-	err = ip_defrag(skb, user);
+	err = ip_defrag(ipvs->net, skb, user);
 	local_bh_enable();
 	if (!err)
 		ip_send_check(ip_hdr(skb));
@@ -665,10 +700,10 @@ static inline int ip_vs_gather_frags(struct sk_buff *skb, u_int32_t user)
 	return err;
 }
 
-static int ip_vs_route_me_harder(int af, struct sk_buff *skb,
-				 unsigned int hooknum)
+static int ip_vs_route_me_harder(struct netns_ipvs *ipvs, int af,
+				 struct sk_buff *skb, unsigned int hooknum)
 {
-	if (!sysctl_snat_reroute(skb))
+	if (!sysctl_snat_reroute(ipvs))
 		return 0;
 	/* Reroute replies only to remote clients (FORWARD and LOCAL_OUT) */
 	if (NF_INET_LOCAL_IN == hooknum)
@@ -678,12 +713,12 @@ static int ip_vs_route_me_harder(int af, struct sk_buff *skb,
 		struct dst_entry *dst = skb_dst(skb);
 
 		if (dst->dev && !(dst->dev->flags & IFF_LOOPBACK) &&
-		    ip6_route_me_harder(skb) != 0)
+		    ip6_route_me_harder(ipvs->net, skb) != 0)
 			return 1;
 	} else
 #endif
 		if (!(skb_rtable(skb)->rt_flags & RTCF_LOCAL) &&
-		    ip_route_me_harder(skb, RTN_LOCAL) != 0)
+		    ip_route_me_harder(ipvs->net, skb, RTN_LOCAL) != 0)
 			return 1;
 
 	return 0;
@@ -836,7 +871,7 @@ static int handle_response_icmp(int af, struct sk_buff *skb,
 #endif
 		ip_vs_nat_icmp(skb, pp, cp, 1);
 
-	if (ip_vs_route_me_harder(af, skb, hooknum))
+	if (ip_vs_route_me_harder(cp->ipvs, af, skb, hooknum))
 		goto out;
 
 	/* do the statistics and put it back */
@@ -860,8 +895,8 @@ out:
  *	Find any that might be relevant, check against existing connections.
  *	Currently handles error types - unreachable, quench, ttl exceeded.
  */
-static int ip_vs_out_icmp(struct sk_buff *skb, int *related,
-			  unsigned int hooknum)
+static int ip_vs_out_icmp(struct netns_ipvs *ipvs, struct sk_buff *skb,
+			  int *related, unsigned int hooknum)
 {
 	struct iphdr *iph;
 	struct icmphdr	_icmph, *ic;
@@ -876,7 +911,7 @@ static int ip_vs_out_icmp(struct sk_buff *skb, int *related,
 
 	/* reassemble IP fragments */
 	if (ip_is_fragment(ip_hdr(skb))) {
-		if (ip_vs_gather_frags(skb, ip_vs_defrag_user(hooknum)))
+		if (ip_vs_gather_frags(ipvs, skb, ip_vs_defrag_user(hooknum)))
 			return NF_STOLEN;
 	}
 
@@ -922,10 +957,10 @@ static int ip_vs_out_icmp(struct sk_buff *skb, int *related,
 	IP_VS_DBG_PKT(11, AF_INET, pp, skb, offset,
 		      "Checking outgoing ICMP for");
 
-	ip_vs_fill_ip4hdr(cih, &ciph);
-	ciph.len += offset;
+	ip_vs_fill_iph_skb_icmp(AF_INET, skb, offset, true, &ciph);
+
 	/* The embedded headers contain source and dest in reverse order */
-	cp = pp->conn_out_get(AF_INET, skb, &ciph, 1);
+	cp = pp->conn_out_get(ipvs, AF_INET, skb, &ciph);
 	if (!cp)
 		return NF_ACCEPT;
 
@@ -935,16 +970,16 @@ static int ip_vs_out_icmp(struct sk_buff *skb, int *related,
 }
 
 #ifdef CONFIG_IP_VS_IPV6
-static int ip_vs_out_icmp_v6(struct sk_buff *skb, int *related,
-			     unsigned int hooknum, struct ip_vs_iphdr *ipvsh)
+static int ip_vs_out_icmp_v6(struct netns_ipvs *ipvs, struct sk_buff *skb,
+			     int *related,  unsigned int hooknum,
+			     struct ip_vs_iphdr *ipvsh)
 {
 	struct icmp6hdr	_icmph, *ic;
-	struct ipv6hdr _ip6h, *ip6h; /* The ip header contained within ICMP */
 	struct ip_vs_iphdr ciph = {.flags = 0, .fragoffs = 0};/*Contained IP */
 	struct ip_vs_conn *cp;
 	struct ip_vs_protocol *pp;
 	union nf_inet_addr snet;
-	unsigned int writable;
+	unsigned int offset;
 
 	*related = 1;
 	ic = frag_safe_skb_hp(skb, ipvsh->len, sizeof(_icmph), &_icmph, ipvsh);
@@ -972,31 +1007,23 @@ static int ip_vs_out_icmp_v6(struct sk_buff *skb, int *related,
 		  ic->icmp6_type, ntohs(icmpv6_id(ic)),
 		  &ipvsh->saddr, &ipvsh->daddr);
 
-	/* Now find the contained IP header */
-	ciph.len = ipvsh->len + sizeof(_icmph);
-	ip6h = skb_header_pointer(skb, ciph.len, sizeof(_ip6h), &_ip6h);
-	if (ip6h == NULL)
+	if (!ip_vs_fill_iph_skb_icmp(AF_INET6, skb, ipvsh->len + sizeof(_icmph),
+				     true, &ciph))
 		return NF_ACCEPT; /* The packet looks wrong, ignore */
-	ciph.saddr.in6 = ip6h->saddr; /* conn_out_get() handles reverse order */
-	ciph.daddr.in6 = ip6h->daddr;
-	/* skip possible IPv6 exthdrs of contained IPv6 packet */
-	ciph.protocol = ipv6_find_hdr(skb, &ciph.len, -1, &ciph.fragoffs, NULL);
-	if (ciph.protocol < 0)
-		return NF_ACCEPT; /* Contained IPv6 hdr looks wrong, ignore */
 
 	pp = ip_vs_proto_get(ciph.protocol);
 	if (!pp)
 		return NF_ACCEPT;
 
 	/* The embedded headers contain source and dest in reverse order */
-	cp = pp->conn_out_get(AF_INET6, skb, &ciph, 1);
+	cp = pp->conn_out_get(ipvs, AF_INET6, skb, &ciph);
 	if (!cp)
 		return NF_ACCEPT;
 
 	snet.in6 = ciph.saddr.in6;
-	writable = ciph.len;
+	offset = ciph.len;
 	return handle_response_icmp(AF_INET6, skb, &snet, ciph.protocol, cp,
-				    pp, writable, sizeof(struct ipv6hdr),
+				    pp, offset, sizeof(struct ipv6hdr),
 				    hooknum);
 }
 #endif
@@ -1081,7 +1108,7 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd,
 {
 	struct ip_vs_protocol *pp = pd->pp;
 
-	IP_VS_DBG_PKT(11, af, pp, skb, 0, "Outgoing packet");
+	IP_VS_DBG_PKT(11, af, pp, skb, iph->off, "Outgoing packet");
 
 	if (!skb_make_writable(skb, iph->len))
 		goto drop;
@@ -1115,10 +1142,10 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd,
 	 * if it came from this machine itself.  So re-compute
 	 * the routing information.
 	 */
-	if (ip_vs_route_me_harder(af, skb, hooknum))
+	if (ip_vs_route_me_harder(cp->ipvs, af, skb, hooknum))
 		goto drop;
 
-	IP_VS_DBG_PKT(10, af, pp, skb, 0, "After SNAT");
+	IP_VS_DBG_PKT(10, af, pp, skb, iph->off, "After SNAT");
 
 	ip_vs_out_stats(cp, skb);
 	ip_vs_set_state(cp, IP_VS_DIR_OUTPUT, skb, pd);
@@ -1143,13 +1170,13 @@ drop:
  *	Check if outgoing packet belongs to the established ip_vs_conn.
  */
 static unsigned int
-ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af)
+ip_vs_out(struct netns_ipvs *ipvs, unsigned int hooknum, struct sk_buff *skb, int af)
 {
-	struct net *net = NULL;
 	struct ip_vs_iphdr iph;
 	struct ip_vs_protocol *pp;
 	struct ip_vs_proto_data *pd;
 	struct ip_vs_conn *cp;
+	struct sock *sk;
 
 	EnterFunction(11);
 
@@ -1157,29 +1184,27 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af)
 	if (skb->ipvs_property)
 		return NF_ACCEPT;
 
+	sk = skb_to_full_sk(skb);
 	/* Bad... Do not break raw sockets */
-	if (unlikely(skb->sk != NULL && hooknum == NF_INET_LOCAL_OUT &&
+	if (unlikely(sk && hooknum == NF_INET_LOCAL_OUT &&
 		     af == AF_INET)) {
-		struct sock *sk = skb->sk;
-		struct inet_sock *inet = inet_sk(skb->sk);
 
-		if (inet && sk->sk_family == PF_INET && inet->nodefrag)
+		if (sk->sk_family == PF_INET && inet_sk(sk)->nodefrag)
 			return NF_ACCEPT;
 	}
 
 	if (unlikely(!skb_dst(skb)))
 		return NF_ACCEPT;
 
-	net = skb_net(skb);
-	if (!net_ipvs(net)->enable)
+	if (!ipvs->enable)
 		return NF_ACCEPT;
 
-	ip_vs_fill_iph_skb(af, skb, &iph);
+	ip_vs_fill_iph_skb(af, skb, false, &iph);
 #ifdef CONFIG_IP_VS_IPV6
 	if (af == AF_INET6) {
 		if (unlikely(iph.protocol == IPPROTO_ICMPV6)) {
 			int related;
-			int verdict = ip_vs_out_icmp_v6(skb, &related,
+			int verdict = ip_vs_out_icmp_v6(ipvs, skb, &related,
 							hooknum, &iph);
 
 			if (related)
@@ -1189,13 +1214,13 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af)
 #endif
 		if (unlikely(iph.protocol == IPPROTO_ICMP)) {
 			int related;
-			int verdict = ip_vs_out_icmp(skb, &related, hooknum);
+			int verdict = ip_vs_out_icmp(ipvs, skb, &related, hooknum);
 
 			if (related)
 				return verdict;
 		}
 
-	pd = ip_vs_proto_data_get(net, iph.protocol);
+	pd = ip_vs_proto_data_get(ipvs, iph.protocol);
 	if (unlikely(!pd))
 		return NF_ACCEPT;
 	pp = pd->pp;
@@ -1205,21 +1230,21 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af)
 	if (af == AF_INET)
 #endif
 		if (unlikely(ip_is_fragment(ip_hdr(skb)) && !pp->dont_defrag)) {
-			if (ip_vs_gather_frags(skb,
+			if (ip_vs_gather_frags(ipvs, skb,
 					       ip_vs_defrag_user(hooknum)))
 				return NF_STOLEN;
 
-			ip_vs_fill_ip4hdr(skb_network_header(skb), &iph);
+			ip_vs_fill_iph_skb(AF_INET, skb, false, &iph);
 		}
 
 	/*
 	 * Check if the packet belongs to an existing entry
 	 */
-	cp = pp->conn_out_get(af, skb, &iph, 0);
+	cp = pp->conn_out_get(ipvs, af, skb, &iph);
 
 	if (likely(cp))
 		return handle_response(af, skb, pd, cp, &iph, hooknum);
-	if (sysctl_nat_icmp_send(net) &&
+	if (sysctl_nat_icmp_send(ipvs) &&
 	    (pp->protocol == IPPROTO_TCP ||
 	     pp->protocol == IPPROTO_UDP ||
 	     pp->protocol == IPPROTO_SCTP)) {
@@ -1229,7 +1254,7 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af)
 					 sizeof(_ports), _ports, &iph);
 		if (pptr == NULL)
 			return NF_ACCEPT;	/* Not for me */
-		if (ip_vs_has_real_service(net, af, iph.protocol, &iph.saddr,
+		if (ip_vs_has_real_service(ipvs, af, iph.protocol, &iph.saddr,
 					   pptr[0])) {
 			/*
 			 * Notify the real server: there is no
@@ -1246,7 +1271,7 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af)
 #ifdef CONFIG_IP_VS_IPV6
 				if (af == AF_INET6) {
 					if (!skb->dev)
-						skb->dev = net->loopback_dev;
+						skb->dev = ipvs->net->loopback_dev;
 					icmpv6_send(skb,
 						    ICMPV6_DEST_UNREACH,
 						    ICMPV6_PORT_UNREACH,
@@ -1260,7 +1285,7 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af)
 			}
 		}
 	}
-	IP_VS_DBG_PKT(12, af, pp, skb, 0,
+	IP_VS_DBG_PKT(12, af, pp, skb, iph.off,
 		      "ip_vs_out: packet continues traversal as normal");
 	return NF_ACCEPT;
 }
@@ -1271,10 +1296,10 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af)
  *	Check if packet is reply for established ip_vs_conn.
  */
 static unsigned int
-ip_vs_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_reply4(void *priv, struct sk_buff *skb,
 	     const struct nf_hook_state *state)
 {
-	return ip_vs_out(ops->hooknum, skb, AF_INET);
+	return ip_vs_out(net_ipvs(state->net), state->hook, skb, AF_INET);
 }
 
 /*
@@ -1282,10 +1307,10 @@ ip_vs_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *	Check if packet is reply for established ip_vs_conn.
  */
 static unsigned int
-ip_vs_local_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_local_reply4(void *priv, struct sk_buff *skb,
 		   const struct nf_hook_state *state)
 {
-	return ip_vs_out(ops->hooknum, skb, AF_INET);
+	return ip_vs_out(net_ipvs(state->net), state->hook, skb, AF_INET);
 }
 
 #ifdef CONFIG_IP_VS_IPV6
@@ -1296,10 +1321,10 @@ ip_vs_local_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *	Check if packet is reply for established ip_vs_conn.
  */
 static unsigned int
-ip_vs_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_reply6(void *priv, struct sk_buff *skb,
 	     const struct nf_hook_state *state)
 {
-	return ip_vs_out(ops->hooknum, skb, AF_INET6);
+	return ip_vs_out(net_ipvs(state->net), state->hook, skb, AF_INET6);
 }
 
 /*
@@ -1307,14 +1332,51 @@ ip_vs_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *	Check if packet is reply for established ip_vs_conn.
  */
 static unsigned int
-ip_vs_local_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_local_reply6(void *priv, struct sk_buff *skb,
 		   const struct nf_hook_state *state)
 {
-	return ip_vs_out(ops->hooknum, skb, AF_INET6);
+	return ip_vs_out(net_ipvs(state->net), state->hook, skb, AF_INET6);
 }
 
 #endif
 
+static unsigned int
+ip_vs_try_to_schedule(struct netns_ipvs *ipvs, int af, struct sk_buff *skb,
+		      struct ip_vs_proto_data *pd,
+		      int *verdict, struct ip_vs_conn **cpp,
+		      struct ip_vs_iphdr *iph)
+{
+	struct ip_vs_protocol *pp = pd->pp;
+
+	if (!iph->fragoffs) {
+		/* No (second) fragments need to enter here, as nf_defrag_ipv6
+		 * replayed fragment zero will already have created the cp
+		 */
+
+		/* Schedule and create new connection entry into cpp */
+		if (!pp->conn_schedule(ipvs, af, skb, pd, verdict, cpp, iph))
+			return 0;
+	}
+
+	if (unlikely(!*cpp)) {
+		/* sorry, all this trouble for a no-hit :) */
+		IP_VS_DBG_PKT(12, af, pp, skb, iph->off,
+			      "ip_vs_in: packet continues traversal as normal");
+		if (iph->fragoffs) {
+			/* Fragment that couldn't be mapped to a conn entry
+			 * is missing module nf_defrag_ipv6
+			 */
+			IP_VS_DBG_RL("Unhandled frag, load nf_defrag_ipv6\n");
+			IP_VS_DBG_PKT(7, af, pp, skb, iph->off,
+				      "unhandled fragment");
+		}
+		*verdict = NF_ACCEPT;
+		return 0;
+	}
+
+	return 1;
+}
+
 /*
  *	Handle ICMP messages in the outside-to-inside direction (incoming).
  *	Find any that might be relevant, check against existing connections,
@@ -1322,9 +1384,9 @@ ip_vs_local_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *	Currently handles error types - unreachable, quench, ttl exceeded.
  */
 static int
-ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum)
+ip_vs_in_icmp(struct netns_ipvs *ipvs, struct sk_buff *skb, int *related,
+	      unsigned int hooknum)
 {
-	struct net *net = NULL;
 	struct iphdr *iph;
 	struct icmphdr	_icmph, *ic;
 	struct iphdr	_ciph, *cih;	/* The ip header contained within the ICMP */
@@ -1333,13 +1395,13 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum)
 	struct ip_vs_protocol *pp;
 	struct ip_vs_proto_data *pd;
 	unsigned int offset, offset2, ihl, verdict;
-	bool ipip;
+	bool ipip, new_cp = false;
 
 	*related = 1;
 
 	/* reassemble IP fragments */
 	if (ip_is_fragment(ip_hdr(skb))) {
-		if (ip_vs_gather_frags(skb, ip_vs_defrag_user(hooknum)))
+		if (ip_vs_gather_frags(ipvs, skb, ip_vs_defrag_user(hooknum)))
 			return NF_STOLEN;
 	}
 
@@ -1373,8 +1435,6 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum)
 	if (cih == NULL)
 		return NF_ACCEPT; /* The packet looks wrong, ignore */
 
-	net = skb_net(skb);
-
 	/* Special case for errors for IPIP packets */
 	ipip = false;
 	if (cih->protocol == IPPROTO_IPIP) {
@@ -1390,7 +1450,7 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum)
 		ipip = true;
 	}
 
-	pd = ip_vs_proto_data_get(net, cih->protocol);
+	pd = ip_vs_proto_data_get(ipvs, cih->protocol);
 	if (!pd)
 		return NF_ACCEPT;
 	pp = pd->pp;
@@ -1404,15 +1464,24 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum)
 		      "Checking incoming ICMP for");
 
 	offset2 = offset;
-	ip_vs_fill_ip4hdr(cih, &ciph);
-	ciph.len += offset;
+	ip_vs_fill_iph_skb_icmp(AF_INET, skb, offset, !ipip, &ciph);
 	offset = ciph.len;
+
 	/* The embedded headers contain source and dest in reverse order.
 	 * For IPIP this is error for request, not for reply.
 	 */
-	cp = pp->conn_in_get(AF_INET, skb, &ciph, ipip ? 0 : 1);
-	if (!cp)
-		return NF_ACCEPT;
+	cp = pp->conn_in_get(ipvs, AF_INET, skb, &ciph);
+
+	if (!cp) {
+		int v;
+
+		if (!sysctl_schedule_icmp(ipvs))
+			return NF_ACCEPT;
+
+		if (!ip_vs_try_to_schedule(ipvs, AF_INET, skb, pd, &v, &cp, &ciph))
+			return v;
+		new_cp = true;
+	}
 
 	verdict = NF_DROP;
 
@@ -1443,7 +1512,7 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum)
 			skb_reset_network_header(skb);
 			IP_VS_DBG(12, "ICMP for IPIP %pI4->%pI4: mtu=%u\n",
 				&ip_hdr(skb)->saddr, &ip_hdr(skb)->daddr, mtu);
-			ipv4_update_pmtu(skb, dev_net(skb->dev),
+			ipv4_update_pmtu(skb, ipvs->net,
 					 mtu, 0, 0, 0, 0);
 			/* Client uses PMTUD? */
 			if (!(frag_off & htons(IP_DF)))
@@ -1489,23 +1558,26 @@ ignore_ipip:
 	verdict = ip_vs_icmp_xmit(skb, cp, pp, offset, hooknum, &ciph);
 
 out:
-	__ip_vs_conn_put(cp);
+	if (likely(!new_cp))
+		__ip_vs_conn_put(cp);
+	else
+		ip_vs_conn_put(cp);
 
 	return verdict;
 }
 
 #ifdef CONFIG_IP_VS_IPV6
-static int ip_vs_in_icmp_v6(struct sk_buff *skb, int *related,
-			    unsigned int hooknum, struct ip_vs_iphdr *iph)
+static int ip_vs_in_icmp_v6(struct netns_ipvs *ipvs, struct sk_buff *skb,
+			    int *related, unsigned int hooknum,
+			    struct ip_vs_iphdr *iph)
 {
-	struct net *net = NULL;
-	struct ipv6hdr _ip6h, *ip6h;
 	struct icmp6hdr	_icmph, *ic;
 	struct ip_vs_iphdr ciph = {.flags = 0, .fragoffs = 0};/*Contained IP */
 	struct ip_vs_conn *cp;
 	struct ip_vs_protocol *pp;
 	struct ip_vs_proto_data *pd;
-	unsigned int offs_ciph, writable, verdict;
+	unsigned int offset, verdict;
+	bool new_cp = false;
 
 	*related = 1;
 
@@ -1534,21 +1606,11 @@ static int ip_vs_in_icmp_v6(struct sk_buff *skb, int *related,
 		  ic->icmp6_type, ntohs(icmpv6_id(ic)),
 		  &iph->saddr, &iph->daddr);
 
-	/* Now find the contained IP header */
-	ciph.len = iph->len + sizeof(_icmph);
-	offs_ciph = ciph.len; /* Save ip header offset */
-	ip6h = skb_header_pointer(skb, ciph.len, sizeof(_ip6h), &_ip6h);
-	if (ip6h == NULL)
-		return NF_ACCEPT; /* The packet looks wrong, ignore */
-	ciph.saddr.in6 = ip6h->saddr; /* conn_in_get() handles reverse order */
-	ciph.daddr.in6 = ip6h->daddr;
-	/* skip possible IPv6 exthdrs of contained IPv6 packet */
-	ciph.protocol = ipv6_find_hdr(skb, &ciph.len, -1, &ciph.fragoffs, NULL);
-	if (ciph.protocol < 0)
-		return NF_ACCEPT; /* Contained IPv6 hdr looks wrong, ignore */
-
-	net = skb_net(skb);
-	pd = ip_vs_proto_data_get(net, ciph.protocol);
+	offset = iph->len + sizeof(_icmph);
+	if (!ip_vs_fill_iph_skb_icmp(AF_INET6, skb, offset, true, &ciph))
+		return NF_ACCEPT;
+
+	pd = ip_vs_proto_data_get(ipvs, ciph.protocol);
 	if (!pd)
 		return NF_ACCEPT;
 	pp = pd->pp;
@@ -1557,36 +1619,49 @@ static int ip_vs_in_icmp_v6(struct sk_buff *skb, int *related,
 	if (ciph.fragoffs)
 		return NF_ACCEPT;
 
-	IP_VS_DBG_PKT(11, AF_INET6, pp, skb, offs_ciph,
+	IP_VS_DBG_PKT(11, AF_INET6, pp, skb, offset,
 		      "Checking incoming ICMPv6 for");
 
 	/* The embedded headers contain source and dest in reverse order
 	 * if not from localhost
 	 */
-	cp = pp->conn_in_get(AF_INET6, skb, &ciph,
-			     (hooknum == NF_INET_LOCAL_OUT) ? 0 : 1);
+	cp = pp->conn_in_get(ipvs, AF_INET6, skb, &ciph);
+
+	if (!cp) {
+		int v;
+
+		if (!sysctl_schedule_icmp(ipvs))
+			return NF_ACCEPT;
+
+		if (!ip_vs_try_to_schedule(ipvs, AF_INET6, skb, pd, &v, &cp, &ciph))
+			return v;
+
+		new_cp = true;
+	}
 
-	if (!cp)
-		return NF_ACCEPT;
 	/* VS/TUN, VS/DR and LOCALNODE just let it go */
 	if ((hooknum == NF_INET_LOCAL_OUT) &&
 	    (IP_VS_FWD_METHOD(cp) != IP_VS_CONN_F_MASQ)) {
-		__ip_vs_conn_put(cp);
-		return NF_ACCEPT;
+		verdict = NF_ACCEPT;
+		goto out;
 	}
 
 	/* do the statistics and put it back */
 	ip_vs_in_stats(cp, skb);
 
 	/* Need to mangle contained IPv6 header in ICMPv6 packet */
-	writable = ciph.len;
+	offset = ciph.len;
 	if (IPPROTO_TCP == ciph.protocol || IPPROTO_UDP == ciph.protocol ||
 	    IPPROTO_SCTP == ciph.protocol)
-		writable += 2 * sizeof(__u16); /* Also mangle ports */
+		offset += 2 * sizeof(__u16); /* Also mangle ports */
 
-	verdict = ip_vs_icmp_xmit_v6(skb, cp, pp, writable, hooknum, &ciph);
+	verdict = ip_vs_icmp_xmit_v6(skb, cp, pp, offset, hooknum, &ciph);
 
-	__ip_vs_conn_put(cp);
+out:
+	if (likely(!new_cp))
+		__ip_vs_conn_put(cp);
+	else
+		ip_vs_conn_put(cp);
 
 	return verdict;
 }
@@ -1598,16 +1673,15 @@ static int ip_vs_in_icmp_v6(struct sk_buff *skb, int *related,
  *	and send it on its way...
  */
 static unsigned int
-ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af)
+ip_vs_in(struct netns_ipvs *ipvs, unsigned int hooknum, struct sk_buff *skb, int af)
 {
-	struct net *net;
 	struct ip_vs_iphdr iph;
 	struct ip_vs_protocol *pp;
 	struct ip_vs_proto_data *pd;
 	struct ip_vs_conn *cp;
 	int ret, pkts;
-	struct netns_ipvs *ipvs;
 	int conn_reuse_mode;
+	struct sock *sk;
 
 	/* Already marked as IPVS request or reply? */
 	if (skb->ipvs_property)
@@ -1621,7 +1695,7 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af)
 	if (unlikely((skb->pkt_type != PACKET_HOST &&
 		      hooknum != NF_INET_LOCAL_OUT) ||
 		     !skb_dst(skb))) {
-		ip_vs_fill_iph_skb(af, skb, &iph);
+		ip_vs_fill_iph_skb(af, skb, false, &iph);
 		IP_VS_DBG_BUF(12, "packet type=%d proto=%d daddr=%s"
 			      " ignored in hook %u\n",
 			      skb->pkt_type, iph.protocol,
@@ -1629,20 +1703,17 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af)
 		return NF_ACCEPT;
 	}
 	/* ipvs enabled in this netns ? */
-	net = skb_net(skb);
-	ipvs = net_ipvs(net);
 	if (unlikely(sysctl_backup_only(ipvs) || !ipvs->enable))
 		return NF_ACCEPT;
 
-	ip_vs_fill_iph_skb(af, skb, &iph);
+	ip_vs_fill_iph_skb(af, skb, false, &iph);
 
 	/* Bad... Do not break raw sockets */
-	if (unlikely(skb->sk != NULL && hooknum == NF_INET_LOCAL_OUT &&
+	sk = skb_to_full_sk(skb);
+	if (unlikely(sk && hooknum == NF_INET_LOCAL_OUT &&
 		     af == AF_INET)) {
-		struct sock *sk = skb->sk;
-		struct inet_sock *inet = inet_sk(skb->sk);
 
-		if (inet && sk->sk_family == PF_INET && inet->nodefrag)
+		if (sk->sk_family == PF_INET && inet_sk(sk)->nodefrag)
 			return NF_ACCEPT;
 	}
 
@@ -1650,8 +1721,8 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af)
 	if (af == AF_INET6) {
 		if (unlikely(iph.protocol == IPPROTO_ICMPV6)) {
 			int related;
-			int verdict = ip_vs_in_icmp_v6(skb, &related, hooknum,
-						       &iph);
+			int verdict = ip_vs_in_icmp_v6(ipvs, skb, &related,
+						       hooknum, &iph);
 
 			if (related)
 				return verdict;
@@ -1660,60 +1731,71 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af)
 #endif
 		if (unlikely(iph.protocol == IPPROTO_ICMP)) {
 			int related;
-			int verdict = ip_vs_in_icmp(skb, &related, hooknum);
+			int verdict = ip_vs_in_icmp(ipvs, skb, &related,
+						    hooknum);
 
 			if (related)
 				return verdict;
 		}
 
 	/* Protocol supported? */
-	pd = ip_vs_proto_data_get(net, iph.protocol);
-	if (unlikely(!pd))
+	pd = ip_vs_proto_data_get(ipvs, iph.protocol);
+	if (unlikely(!pd)) {
+		/* The only way we'll see this packet again is if it's
+		 * encapsulated, so mark it with ipvs_property=1 so we
+		 * skip it if we're ignoring tunneled packets
+		 */
+		if (sysctl_ignore_tunneled(ipvs))
+			skb->ipvs_property = 1;
+
 		return NF_ACCEPT;
+	}
 	pp = pd->pp;
 	/*
 	 * Check if the packet belongs to an existing connection entry
 	 */
-	cp = pp->conn_in_get(af, skb, &iph, 0);
+	cp = pp->conn_in_get(ipvs, af, skb, &iph);
 
 	conn_reuse_mode = sysctl_conn_reuse_mode(ipvs);
-	if (conn_reuse_mode && !iph.fragoffs &&
-	    is_new_conn(skb, &iph) && cp &&
-	    ((unlikely(sysctl_expire_nodest_conn(ipvs)) && cp->dest &&
-	      unlikely(!atomic_read(&cp->dest->weight))) ||
-	     unlikely(is_new_conn_expected(cp, conn_reuse_mode)))) {
-		if (!atomic_read(&cp->n_control))
-			ip_vs_conn_expire_now(cp);
-		__ip_vs_conn_put(cp);
-		cp = NULL;
+	if (conn_reuse_mode && !iph.fragoffs && is_new_conn(skb, &iph) && cp) {
+		bool uses_ct = false, resched = false;
+
+		if (unlikely(sysctl_expire_nodest_conn(ipvs)) && cp->dest &&
+		    unlikely(!atomic_read(&cp->dest->weight))) {
+			resched = true;
+			uses_ct = ip_vs_conn_uses_conntrack(cp, skb);
+		} else if (is_new_conn_expected(cp, conn_reuse_mode)) {
+			uses_ct = ip_vs_conn_uses_conntrack(cp, skb);
+			if (!atomic_read(&cp->n_control)) {
+				resched = true;
+			} else {
+				/* Do not reschedule controlling connection
+				 * that uses conntrack while it is still
+				 * referenced by controlled connection(s).
+				 */
+				resched = !uses_ct;
+			}
+		}
+
+		if (resched) {
+			if (!atomic_read(&cp->n_control))
+				ip_vs_conn_expire_now(cp);
+			__ip_vs_conn_put(cp);
+			if (uses_ct)
+				return NF_DROP;
+			cp = NULL;
+		}
 	}
 
-	if (unlikely(!cp) && !iph.fragoffs) {
-		/* No (second) fragments need to enter here, as nf_defrag_ipv6
-		 * replayed fragment zero will already have created the cp
-		 */
+	if (unlikely(!cp)) {
 		int v;
 
-		/* Schedule and create new connection entry into &cp */
-		if (!pp->conn_schedule(af, skb, pd, &v, &cp, &iph))
+		if (!ip_vs_try_to_schedule(ipvs, af, skb, pd, &v, &cp, &iph))
 			return v;
 	}
 
-	if (unlikely(!cp)) {
-		/* sorry, all this trouble for a no-hit :) */
-		IP_VS_DBG_PKT(12, af, pp, skb, 0,
-			      "ip_vs_in: packet continues traversal as normal");
-		if (iph.fragoffs) {
-			/* Fragment that couldn't be mapped to a conn entry
-			 * is missing module nf_defrag_ipv6
-			 */
-			IP_VS_DBG_RL("Unhandled frag, load nf_defrag_ipv6\n");
-			IP_VS_DBG_PKT(7, af, pp, skb, 0, "unhandled fragment");
-		}
-		return NF_ACCEPT;
-	}
+	IP_VS_DBG_PKT(11, af, pp, skb, iph.off, "Incoming packet");
 
-	IP_VS_DBG_PKT(11, af, pp, skb, 0, "Incoming packet");
 	/* Check the server status */
 	if (cp->dest && !(cp->dest->flags & IP_VS_DEST_F_AVAILABLE)) {
 		/* the destination server is not available */
@@ -1753,7 +1835,7 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af)
 		pkts = atomic_add_return(1, &cp->in_pkts);
 
 	if (ipvs->sync_state & IP_VS_STATE_MASTER)
-		ip_vs_sync_conn(net, cp, pkts);
+		ip_vs_sync_conn(ipvs, cp, pkts);
 
 	ip_vs_conn_put(cp);
 	return ret;
@@ -1764,10 +1846,10 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af)
  *	Schedule and forward packets from remote clients
  */
 static unsigned int
-ip_vs_remote_request4(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_remote_request4(void *priv, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
-	return ip_vs_in(ops->hooknum, skb, AF_INET);
+	return ip_vs_in(net_ipvs(state->net), state->hook, skb, AF_INET);
 }
 
 /*
@@ -1775,10 +1857,10 @@ ip_vs_remote_request4(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *	Schedule and forward packets from local clients
  */
 static unsigned int
-ip_vs_local_request4(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_local_request4(void *priv, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	return ip_vs_in(ops->hooknum, skb, AF_INET);
+	return ip_vs_in(net_ipvs(state->net), state->hook, skb, AF_INET);
 }
 
 #ifdef CONFIG_IP_VS_IPV6
@@ -1788,10 +1870,10 @@ ip_vs_local_request4(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *	Schedule and forward packets from remote clients
  */
 static unsigned int
-ip_vs_remote_request6(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_remote_request6(void *priv, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
-	return ip_vs_in(ops->hooknum, skb, AF_INET6);
+	return ip_vs_in(net_ipvs(state->net), state->hook, skb, AF_INET6);
 }
 
 /*
@@ -1799,10 +1881,10 @@ ip_vs_remote_request6(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *	Schedule and forward packets from local clients
  */
 static unsigned int
-ip_vs_local_request6(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_local_request6(void *priv, struct sk_buff *skb,
 		     const struct nf_hook_state *state)
 {
-	return ip_vs_in(ops->hooknum, skb, AF_INET6);
+	return ip_vs_in(net_ipvs(state->net), state->hook, skb, AF_INET6);
 }
 
 #endif
@@ -1818,46 +1900,40 @@ ip_vs_local_request6(const struct nf_hook_ops *ops, struct sk_buff *skb,
  *      and send them to ip_vs_in_icmp.
  */
 static unsigned int
-ip_vs_forward_icmp(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_forward_icmp(void *priv, struct sk_buff *skb,
 		   const struct nf_hook_state *state)
 {
 	int r;
-	struct net *net;
-	struct netns_ipvs *ipvs;
+	struct netns_ipvs *ipvs = net_ipvs(state->net);
 
 	if (ip_hdr(skb)->protocol != IPPROTO_ICMP)
 		return NF_ACCEPT;
 
 	/* ipvs enabled in this netns ? */
-	net = skb_net(skb);
-	ipvs = net_ipvs(net);
 	if (unlikely(sysctl_backup_only(ipvs) || !ipvs->enable))
 		return NF_ACCEPT;
 
-	return ip_vs_in_icmp(skb, &r, ops->hooknum);
+	return ip_vs_in_icmp(ipvs, skb, &r, state->hook);
 }
 
 #ifdef CONFIG_IP_VS_IPV6
 static unsigned int
-ip_vs_forward_icmp_v6(const struct nf_hook_ops *ops, struct sk_buff *skb,
+ip_vs_forward_icmp_v6(void *priv, struct sk_buff *skb,
 		      const struct nf_hook_state *state)
 {
 	int r;
-	struct net *net;
-	struct netns_ipvs *ipvs;
+	struct netns_ipvs *ipvs = net_ipvs(state->net);
 	struct ip_vs_iphdr iphdr;
 
-	ip_vs_fill_iph_skb(AF_INET6, skb, &iphdr);
+	ip_vs_fill_iph_skb(AF_INET6, skb, false, &iphdr);
 	if (iphdr.protocol != IPPROTO_ICMPV6)
 		return NF_ACCEPT;
 
 	/* ipvs enabled in this netns ? */
-	net = skb_net(skb);
-	ipvs = net_ipvs(net);
 	if (unlikely(sysctl_backup_only(ipvs) || !ipvs->enable))
 		return NF_ACCEPT;
 
-	return ip_vs_in_icmp_v6(skb, &r, ops->hooknum, &iphdr);
+	return ip_vs_in_icmp_v6(ipvs, skb, &r, state->hook, &iphdr);
 }
 #endif
 
@@ -1866,7 +1942,6 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 	/* After packet filtering, change source only for VS/NAT */
 	{
 		.hook		= ip_vs_reply4,
-		.owner		= THIS_MODULE,
 		.pf		= NFPROTO_IPV4,
 		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP_PRI_NAT_SRC - 2,
@@ -1876,7 +1951,6 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 	 * applied to IPVS. */
 	{
 		.hook		= ip_vs_remote_request4,
-		.owner		= THIS_MODULE,
 		.pf		= NFPROTO_IPV4,
 		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP_PRI_NAT_SRC - 1,
@@ -1884,7 +1958,6 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 	/* Before ip_vs_in, change source only for VS/NAT */
 	{
 		.hook		= ip_vs_local_reply4,
-		.owner		= THIS_MODULE,
 		.pf		= NFPROTO_IPV4,
 		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP_PRI_NAT_DST + 1,
@@ -1892,7 +1965,6 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 	/* After mangle, schedule and forward local requests */
 	{
 		.hook		= ip_vs_local_request4,
-		.owner		= THIS_MODULE,
 		.pf		= NFPROTO_IPV4,
 		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP_PRI_NAT_DST + 2,
@@ -1901,7 +1973,6 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 	 * destined for 0.0.0.0/0, which is for incoming IPVS connections */
 	{
 		.hook		= ip_vs_forward_icmp,
-		.owner		= THIS_MODULE,
 		.pf		= NFPROTO_IPV4,
 		.hooknum	= NF_INET_FORWARD,
 		.priority	= 99,
@@ -1909,7 +1980,6 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 	/* After packet filtering, change source only for VS/NAT */
 	{
 		.hook		= ip_vs_reply4,
-		.owner		= THIS_MODULE,
 		.pf		= NFPROTO_IPV4,
 		.hooknum	= NF_INET_FORWARD,
 		.priority	= 100,
@@ -1918,7 +1988,6 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 	/* After packet filtering, change source only for VS/NAT */
 	{
 		.hook		= ip_vs_reply6,
-		.owner		= THIS_MODULE,
 		.pf		= NFPROTO_IPV6,
 		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP6_PRI_NAT_SRC - 2,
@@ -1928,7 +1997,6 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 	 * applied to IPVS. */
 	{
 		.hook		= ip_vs_remote_request6,
-		.owner		= THIS_MODULE,
 		.pf		= NFPROTO_IPV6,
 		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP6_PRI_NAT_SRC - 1,
@@ -1936,7 +2004,6 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 	/* Before ip_vs_in, change source only for VS/NAT */
 	{
 		.hook		= ip_vs_local_reply6,
-		.owner		= THIS_MODULE,
 		.pf		= NFPROTO_IPV6,
 		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP6_PRI_NAT_DST + 1,
@@ -1944,7 +2011,6 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 	/* After mangle, schedule and forward local requests */
 	{
 		.hook		= ip_vs_local_request6,
-		.owner		= THIS_MODULE,
 		.pf		= NFPROTO_IPV6,
 		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP6_PRI_NAT_DST + 2,
@@ -1953,7 +2019,6 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 	 * destined for 0.0.0.0/0, which is for incoming IPVS connections */
 	{
 		.hook		= ip_vs_forward_icmp_v6,
-		.owner		= THIS_MODULE,
 		.pf		= NFPROTO_IPV6,
 		.hooknum	= NF_INET_FORWARD,
 		.priority	= 99,
@@ -1961,7 +2026,6 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 	/* After packet filtering, change source only for VS/NAT */
 	{
 		.hook		= ip_vs_reply6,
-		.owner		= THIS_MODULE,
 		.pf		= NFPROTO_IPV6,
 		.hooknum	= NF_INET_FORWARD,
 		.priority	= 100,
@@ -1987,22 +2051,22 @@ static int __net_init __ip_vs_init(struct net *net)
 	atomic_inc(&ipvs_netns_cnt);
 	net->ipvs = ipvs;
 
-	if (ip_vs_estimator_net_init(net) < 0)
+	if (ip_vs_estimator_net_init(ipvs) < 0)
 		goto estimator_fail;
 
-	if (ip_vs_control_net_init(net) < 0)
+	if (ip_vs_control_net_init(ipvs) < 0)
 		goto control_fail;
 
-	if (ip_vs_protocol_net_init(net) < 0)
+	if (ip_vs_protocol_net_init(ipvs) < 0)
 		goto protocol_fail;
 
-	if (ip_vs_app_net_init(net) < 0)
+	if (ip_vs_app_net_init(ipvs) < 0)
 		goto app_fail;
 
-	if (ip_vs_conn_net_init(net) < 0)
+	if (ip_vs_conn_net_init(ipvs) < 0)
 		goto conn_fail;
 
-	if (ip_vs_sync_net_init(net) < 0)
+	if (ip_vs_sync_net_init(ipvs) < 0)
 		goto sync_fail;
 
 	printk(KERN_INFO "IPVS: Creating netns size=%zu id=%d\n",
@@ -2013,15 +2077,15 @@ static int __net_init __ip_vs_init(struct net *net)
  */
 
 sync_fail:
-	ip_vs_conn_net_cleanup(net);
+	ip_vs_conn_net_cleanup(ipvs);
 conn_fail:
-	ip_vs_app_net_cleanup(net);
+	ip_vs_app_net_cleanup(ipvs);
 app_fail:
-	ip_vs_protocol_net_cleanup(net);
+	ip_vs_protocol_net_cleanup(ipvs);
 protocol_fail:
-	ip_vs_control_net_cleanup(net);
+	ip_vs_control_net_cleanup(ipvs);
 control_fail:
-	ip_vs_estimator_net_cleanup(net);
+	ip_vs_estimator_net_cleanup(ipvs);
 estimator_fail:
 	net->ipvs = NULL;
 	return -ENOMEM;
@@ -2029,22 +2093,25 @@ estimator_fail:
 
 static void __net_exit __ip_vs_cleanup(struct net *net)
 {
-	ip_vs_service_net_cleanup(net);	/* ip_vs_flush() with locks */
-	ip_vs_conn_net_cleanup(net);
-	ip_vs_app_net_cleanup(net);
-	ip_vs_protocol_net_cleanup(net);
-	ip_vs_control_net_cleanup(net);
-	ip_vs_estimator_net_cleanup(net);
-	IP_VS_DBG(2, "ipvs netns %d released\n", net_ipvs(net)->gen);
+	struct netns_ipvs *ipvs = net_ipvs(net);
+
+	ip_vs_service_net_cleanup(ipvs);	/* ip_vs_flush() with locks */
+	ip_vs_conn_net_cleanup(ipvs);
+	ip_vs_app_net_cleanup(ipvs);
+	ip_vs_protocol_net_cleanup(ipvs);
+	ip_vs_control_net_cleanup(ipvs);
+	ip_vs_estimator_net_cleanup(ipvs);
+	IP_VS_DBG(2, "ipvs netns %d released\n", ipvs->gen);
 	net->ipvs = NULL;
 }
 
 static void __net_exit __ip_vs_dev_cleanup(struct net *net)
 {
+	struct netns_ipvs *ipvs = net_ipvs(net);
 	EnterFunction(2);
-	net_ipvs(net)->enable = 0;	/* Disable packet reception */
+	ipvs->enable = 0;	/* Disable packet reception */
 	smp_wmb();
-	ip_vs_sync_net_cleanup(net);
+	ip_vs_sync_net_cleanup(ipvs);
 	LeaveFunction(2);
 }