Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf-next

Pablo Neira Ayuso says:

====================
Netfilter updates for net-next

The following patchset contains Netfilter updates for your net-next tree:

1) Get rid of nf_sk_is_transparent(), use inet_sk_transparent() instead.
   From Máté Eckl.

2) Move shared tproxy infrastructure to nf_tproxy_ipv4 and nf_tproxy_ipv6.
   Also from Máté.

3) Add hashtable to speed up chain lookups by name, from Florian Westphal.

4) Patch series to add connlimit support reusing part of the
   nf_conncount infrastructure. This includes preparation changes such
   passing context to the object and expression destroy interface;
   garbage collection for expressions embedded into set elements, and
   the introduction of the clone_destroy interface for expressions.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/netfilter/nf_conntrack_count.h

@@ -13,4 +13,15 @@ unsigned int nf_conncount_count(struct net *net,
 				const u32 *key,
 				const struct nf_conntrack_tuple *tuple,
 				const struct nf_conntrack_zone *zone);
+
+unsigned int nf_conncount_lookup(struct net *net, struct hlist_head *head,
+				 const struct nf_conntrack_tuple *tuple,
+				 const struct nf_conntrack_zone *zone,
+				 bool *addit);
+
+bool nf_conncount_add(struct hlist_head *head,
+		      const struct nf_conntrack_tuple *tuple);
+
+void nf_conncount_cache_free(struct hlist_head *hhead);
+
 #endif

include/net/netfilter/nf_socket.h

@@ -3,19 +3,6 @@
 #define _NF_SOCK_H_
 
 #include <net/sock.h>
-#include <net/inet_timewait_sock.h>
-
-static inline bool nf_sk_is_transparent(struct sock *sk)
-{
-	switch (sk->sk_state) {
-	case TCP_TIME_WAIT:
-		return inet_twsk(sk)->tw_transparent;
-	case TCP_NEW_SYN_RECV:
-		return inet_rsk(inet_reqsk(sk))->no_srccheck;
-	default:
-		return inet_sk(sk)->transparent;
-	}
-}
 
 struct sock *nf_sk_lookup_slow_v4(struct net *net, const struct sk_buff *skb,
 				  const struct net_device *indev);

include/net/netfilter/nf_tables.h

@@ -9,6 +9,7 @@
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter/nf_tables.h>
 #include <linux/u64_stats_sync.h>
+#include <linux/rhashtable.h>
 #include <net/netfilter/nf_flow_table.h>
 #include <net/netlink.h>
 
@@ -342,6 +343,7 @@ struct nft_set_ops {
 						const struct nft_set_desc *desc,
 						const struct nlattr * const nla[]);
 	void				(*destroy)(const struct nft_set *set);
+	void				(*gc_init)(const struct nft_set *set);
 
 	unsigned int			elemsize;
 };
@@ -370,6 +372,8 @@ void nft_unregister_set(struct nft_set_type *type);
  *
  *	@list: table set list node
  *	@bindings: list of set bindings
+ *	@table: table this set belongs to
+ *	@net: netnamespace this set belongs to
  * 	@name: name of the set
  *	@handle: unique handle of the set
  * 	@ktype: key type (numeric type defined by userspace, not used in the kernel)
@@ -393,6 +397,8 @@ void nft_unregister_set(struct nft_set_type *type);
 struct nft_set {
 	struct list_head		list;
 	struct list_head		bindings;
+	struct nft_table		*table;
+	possible_net_t			net;
 	char				*name;
 	u64				handle;
 	u32				ktype;
@@ -708,6 +714,7 @@ struct nft_expr_type {
 };
 
 #define NFT_EXPR_STATEFUL		0x1
+#define NFT_EXPR_GC			0x2
 
 /**
  *	struct nft_expr_ops - nf_tables expression operations
@@ -739,11 +746,15 @@ struct nft_expr_ops {
 						      const struct nft_expr *expr);
 	void				(*destroy)(const struct nft_ctx *ctx,
 						   const struct nft_expr *expr);
+	void				(*destroy_clone)(const struct nft_ctx *ctx,
+							 const struct nft_expr *expr);
 	int				(*dump)(struct sk_buff *skb,
 						const struct nft_expr *expr);
 	int				(*validate)(const struct nft_ctx *ctx,
 						    const struct nft_expr *expr,
 						    const struct nft_data **data);
+	bool				(*gc)(struct net *net,
+					      const struct nft_expr *expr);
 	const struct nft_expr_type	*type;
 	void				*data;
 };
@@ -850,6 +861,7 @@ enum nft_chain_flags {
  *
  *	@rules: list of rules in the chain
  *	@list: used internally
+ *	@rhlhead: used internally
  *	@table: table that this chain belongs to
  *	@handle: chain handle
  *	@use: number of jump references to this chain
@@ -862,6 +874,7 @@ struct nft_chain {
 	struct nft_rule			*__rcu *rules_gen_1;
 	struct list_head		rules;
 	struct list_head		list;
+	struct rhlist_head		rhlhead;
 	struct nft_table		*table;
 	u64				handle;
 	u32				use;
@@ -955,7 +968,8 @@ unsigned int nft_do_chain(struct nft_pktinfo *pkt, void *priv);
  *	struct nft_table - nf_tables table
  *
  *	@list: used internally
- *	@chains: chains in the table
+ *	@chains_ht: chains in the table
+ *	@chains: same, for stable walks
  *	@sets: sets in the table
  *	@objects: stateful objects in the table
  *	@flowtables: flow tables in the table
@@ -969,6 +983,7 @@ unsigned int nft_do_chain(struct nft_pktinfo *pkt, void *priv);
  */
 struct nft_table {
 	struct list_head		list;
+	struct rhltable			chains_ht;
 	struct list_head		chains;
 	struct list_head		sets;
 	struct list_head		objects;
@@ -1070,7 +1085,8 @@ struct nft_object_ops {
 	int				(*init)(const struct nft_ctx *ctx,
 						const struct nlattr *const tb[],
 						struct nft_object *obj);
-	void				(*destroy)(struct nft_object *obj);
+	void				(*destroy)(const struct nft_ctx *ctx,
+						   struct nft_object *obj);
 	int				(*dump)(struct sk_buff *skb,
 						struct nft_object *obj,
 						bool reset);

include/net/netfilter/nf_tproxy.h

+#ifndef _NF_TPROXY_H_
+#define _NF_TPROXY_H_
+
+#include <net/tcp.h>
+
+enum nf_tproxy_lookup_t {
+	 NF_TPROXY_LOOKUP_LISTENER,
+	 NF_TPROXY_LOOKUP_ESTABLISHED,
+};
+
+static inline bool nf_tproxy_sk_is_transparent(struct sock *sk)
+{
+	if (inet_sk_transparent(sk))
+		return true;
+
+	sock_gen_put(sk);
+	return false;
+}
+
+__be32 nf_tproxy_laddr4(struct sk_buff *skb, __be32 user_laddr, __be32 daddr);
+
+/**
+ * nf_tproxy_handle_time_wait4 - handle IPv4 TCP TIME_WAIT reopen redirections
+ * @skb:	The skb being processed.
+ * @laddr:	IPv4 address to redirect to or zero.
+ * @lport:	TCP port to redirect to or zero.
+ * @sk:		The TIME_WAIT TCP socket found by the lookup.
+ *
+ * We have to handle SYN packets arriving to TIME_WAIT sockets
+ * differently: instead of reopening the connection we should rather
+ * redirect the new connection to the proxy if there's a listener
+ * socket present.
+ *
+ * nf_tproxy_handle_time_wait4() consumes the socket reference passed in.
+ *
+ * Returns the listener socket if there's one, the TIME_WAIT socket if
+ * no such listener is found, or NULL if the TCP header is incomplete.
+ */
+struct sock *
+nf_tproxy_handle_time_wait4(struct net *net, struct sk_buff *skb,
+			    __be32 laddr, __be16 lport, struct sock *sk);
+
+/*
+ * This is used when the user wants to intercept a connection matching
+ * an explicit iptables rule. In this case the sockets are assumed
+ * matching in preference order:
+ *
+ *   - match: if there's a fully established connection matching the
+ *     _packet_ tuple, it is returned, assuming the redirection
+ *     already took place and we process a packet belonging to an
+ *     established connection
+ *
+ *   - match: if there's a listening socket matching the redirection
+ *     (e.g. on-port & on-ip of the connection), it is returned,
+ *     regardless if it was bound to 0.0.0.0 or an explicit
+ *     address. The reasoning is that if there's an explicit rule, it
+ *     does not really matter if the listener is bound to an interface
+ *     or to 0. The user already stated that he wants redirection
+ *     (since he added the rule).
+ *
+ * Please note that there's an overlap between what a TPROXY target
+ * and a socket match will match. Normally if you have both rules the
+ * "socket" match will be the first one, effectively all packets
+ * belonging to established connections going through that one.
+ */
+struct sock *
+nf_tproxy_get_sock_v4(struct net *net, struct sk_buff *skb, void *hp,
+		      const u8 protocol,
+		      const __be32 saddr, const __be32 daddr,
+		      const __be16 sport, const __be16 dport,
+		      const struct net_device *in,
+		      const enum nf_tproxy_lookup_t lookup_type);
+
+const struct in6_addr *
+nf_tproxy_laddr6(struct sk_buff *skb, const struct in6_addr *user_laddr,
+		 const struct in6_addr *daddr);
+
+/**
+ * nf_tproxy_handle_time_wait6 - handle IPv6 TCP TIME_WAIT reopen redirections
+ * @skb:	The skb being processed.
+ * @tproto:	Transport protocol.
+ * @thoff:	Transport protocol header offset.
+ * @net:	Network namespace.
+ * @laddr:	IPv6 address to redirect to.
+ * @lport:	TCP port to redirect to or zero.
+ * @sk:		The TIME_WAIT TCP socket found by the lookup.
+ *
+ * We have to handle SYN packets arriving to TIME_WAIT sockets
+ * differently: instead of reopening the connection we should rather
+ * redirect the new connection to the proxy if there's a listener
+ * socket present.
+ *
+ * nf_tproxy_handle_time_wait6() consumes the socket reference passed in.
+ *
+ * Returns the listener socket if there's one, the TIME_WAIT socket if
+ * no such listener is found, or NULL if the TCP header is incomplete.
+ */
+struct sock *
+nf_tproxy_handle_time_wait6(struct sk_buff *skb, int tproto, int thoff,
+			    struct net *net,
+			    const struct in6_addr *laddr,
+			    const __be16 lport,
+			    struct sock *sk);
+
+struct sock *
+nf_tproxy_get_sock_v6(struct net *net, struct sk_buff *skb, int thoff, void *hp,
+		      const u8 protocol,
+		      const struct in6_addr *saddr, const struct in6_addr *daddr,
+		      const __be16 sport, const __be16 dport,
+		      const struct net_device *in,
+		      const enum nf_tproxy_lookup_t lookup_type);
+
+#endif /* _NF_TPROXY_H_ */

include/uapi/linux/netfilter/nf_tables.h

@@ -1043,6 +1043,24 @@ enum nft_limit_attributes {
 };
 #define NFTA_LIMIT_MAX		(__NFTA_LIMIT_MAX - 1)
 
+enum nft_connlimit_flags {
+	NFT_CONNLIMIT_F_INV	= (1 << 0),
+};
+
+/**
+ * enum nft_connlimit_attributes - nf_tables connlimit expression netlink attributes
+ *
+ * @NFTA_CONNLIMIT_COUNT: number of connections (NLA_U32)
+ * @NFTA_CONNLIMIT_FLAGS: flags (NLA_U32: enum nft_connlimit_flags)
+ */
+enum nft_connlimit_attributes {
+	NFTA_CONNLIMIT_UNSPEC,
+	NFTA_CONNLIMIT_COUNT,
+	NFTA_CONNLIMIT_FLAGS,
+	__NFTA_CONNLIMIT_MAX
+};
+#define NFTA_CONNLIMIT_MAX	(__NFTA_CONNLIMIT_MAX - 1)
+
 /**
  * enum nft_counter_attributes - nf_tables counter expression netlink attributes
  *
@@ -1357,7 +1375,8 @@ enum nft_ct_helper_attributes {
 #define NFT_OBJECT_QUOTA	2
 #define NFT_OBJECT_CT_HELPER	3
 #define NFT_OBJECT_LIMIT	4
-#define __NFT_OBJECT_MAX	5
+#define NFT_OBJECT_CONNLIMIT	5
+#define __NFT_OBJECT_MAX	6
 #define NFT_OBJECT_MAX		(__NFT_OBJECT_MAX - 1)
 
 /**

net/ipv4/netfilter/Kconfig

@@ -29,7 +29,10 @@ config NF_SOCKET_IPV4
 	tristate "IPv4 socket lookup support"
 	help
 	  This option enables the IPv4 socket lookup infrastructure. This is
-	  is required by the iptables socket match.
+	  is required by the {ip,nf}tables socket match.
+
+config NF_TPROXY_IPV4
+	tristate "IPv4 tproxy support"
 
 if NF_TABLES
 

net/ipv4/netfilter/Makefile

@@ -17,6 +17,7 @@ obj-$(CONFIG_NF_NAT_IPV4) += nf_nat_ipv4.o
 obj-$(CONFIG_NF_DEFRAG_IPV4) += nf_defrag_ipv4.o
 
 obj-$(CONFIG_NF_SOCKET_IPV4) += nf_socket_ipv4.o
+obj-$(CONFIG_NF_TPROXY_IPV4) += nf_tproxy_ipv4.o
 
 # logging
 obj-$(CONFIG_NF_LOG_ARP) += nf_log_arp.o

net/ipv4/netfilter/nf_tproxy_ipv4.c

+/*
+ * Copyright (C) 2007-2008 BalaBit IT Ltd.
+ * Author: Krisztian Kovacs
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <net/netfilter/nf_tproxy.h>
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <net/sock.h>
+#include <net/inet_sock.h>
+#include <linux/ip.h>
+#include <net/checksum.h>
+#include <net/udp.h>
+#include <net/tcp.h>
+#include <linux/inetdevice.h>
+
+struct sock *
+nf_tproxy_handle_time_wait4(struct net *net, struct sk_buff *skb,
+			 __be32 laddr, __be16 lport, struct sock *sk)
+{
+	const struct iphdr *iph = ip_hdr(skb);
+	struct tcphdr _hdr, *hp;
+
+	hp = skb_header_pointer(skb, ip_hdrlen(skb), sizeof(_hdr), &_hdr);
+	if (hp == NULL) {
+		inet_twsk_put(inet_twsk(sk));
+		return NULL;
+	}
+
+	if (hp->syn && !hp->rst && !hp->ack && !hp->fin) {
+		/* SYN to a TIME_WAIT socket, we'd rather redirect it
+		 * to a listener socket if there's one */
+		struct sock *sk2;
+
+		sk2 = nf_tproxy_get_sock_v4(net, skb, hp, iph->protocol,
+					    iph->saddr, laddr ? laddr : iph->daddr,
+					    hp->source, lport ? lport : hp->dest,
+					    skb->dev, NF_TPROXY_LOOKUP_LISTENER);
+		if (sk2) {
+			inet_twsk_deschedule_put(inet_twsk(sk));
+			sk = sk2;
+		}
+	}
+
+	return sk;
+}
+EXPORT_SYMBOL_GPL(nf_tproxy_handle_time_wait4);
+
+__be32 nf_tproxy_laddr4(struct sk_buff *skb, __be32 user_laddr, __be32 daddr)
+{
+	struct in_device *indev;
+	__be32 laddr;
+
+	if (user_laddr)
+		return user_laddr;
+
+	laddr = 0;
+	indev = __in_dev_get_rcu(skb->dev);
+	for_primary_ifa(indev) {
+		laddr = ifa->ifa_local;
+		break;
+	} endfor_ifa(indev);
+
+	return laddr ? laddr : daddr;
+}
+EXPORT_SYMBOL_GPL(nf_tproxy_laddr4);
+
+struct sock *
+nf_tproxy_get_sock_v4(struct net *net, struct sk_buff *skb, void *hp,
+		      const u8 protocol,
+		      const __be32 saddr, const __be32 daddr,
+		      const __be16 sport, const __be16 dport,
+		      const struct net_device *in,
+		      const enum nf_tproxy_lookup_t lookup_type)
+{
+	struct sock *sk;
+	struct tcphdr *tcph;
+
+	switch (protocol) {
+	case IPPROTO_TCP:
+		switch (lookup_type) {
+		case NF_TPROXY_LOOKUP_LISTENER:
+			tcph = hp;
+			sk = inet_lookup_listener(net, &tcp_hashinfo, skb,
+						    ip_hdrlen(skb) +
+						      __tcp_hdrlen(tcph),
+						    saddr, sport,
+						    daddr, dport,
+						    in->ifindex, 0);
+
+			if (sk && !refcount_inc_not_zero(&sk->sk_refcnt))
+				sk = NULL;
+			/* NOTE: we return listeners even if bound to
+			 * 0.0.0.0, those are filtered out in
+			 * xt_socket, since xt_TPROXY needs 0 bound
+			 * listeners too
+			 */
+			break;
+		case NF_TPROXY_LOOKUP_ESTABLISHED:
+			sk = inet_lookup_established(net, &tcp_hashinfo,
+						    saddr, sport, daddr, dport,
+						    in->ifindex);
+			break;
+		default:
+			BUG();
+		}
+		break;
+	case IPPROTO_UDP:
+		sk = udp4_lib_lookup(net, saddr, sport, daddr, dport,
+				     in->ifindex);
+		if (sk) {
+			int connected = (sk->sk_state == TCP_ESTABLISHED);
+			int wildcard = (inet_sk(sk)->inet_rcv_saddr == 0);
+
+			/* NOTE: we return listeners even if bound to
+			 * 0.0.0.0, those are filtered out in
+			 * xt_socket, since xt_TPROXY needs 0 bound
+			 * listeners too
+			 */
+			if ((lookup_type == NF_TPROXY_LOOKUP_ESTABLISHED &&
+			      (!connected || wildcard)) ||
+			    (lookup_type == NF_TPROXY_LOOKUP_LISTENER && connected)) {
+				sock_put(sk);
+				sk = NULL;
+			}
+		}
+		break;
+	default:
+		WARN_ON(1);
+		sk = NULL;
+	}
+
+	pr_debug("tproxy socket lookup: proto %u %08x:%u -> %08x:%u, lookup type: %d, sock %p\n",
+		 protocol, ntohl(saddr), ntohs(sport), ntohl(daddr), ntohs(dport), lookup_type, sk);
+
+	return sk;
+}
+EXPORT_SYMBOL_GPL(nf_tproxy_get_sock_v4);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Balazs Scheidler, Krisztian Kovacs");
+MODULE_DESCRIPTION("Netfilter IPv4 transparent proxy support");

net/ipv6/netfilter/Kconfig

@@ -29,7 +29,10 @@ config NF_SOCKET_IPV6
 	tristate "IPv6 socket lookup support"
 	help
 	  This option enables the IPv6 socket lookup infrastructure. This
-	  is used by the ip6tables socket match.
+	  is used by the {ip6,nf}tables socket match.
+
+config NF_TPROXY_IPV6
+	tristate "IPv6 tproxy support"
 
 if NF_TABLES
 

net/ipv6/netfilter/Makefile

@@ -26,6 +26,7 @@ nf_defrag_ipv6-y := nf_defrag_ipv6_hooks.o nf_conntrack_reasm.o
 obj-$(CONFIG_NF_DEFRAG_IPV6) += nf_defrag_ipv6.o
 
 obj-$(CONFIG_NF_SOCKET_IPV6) += nf_socket_ipv6.o
+obj-$(CONFIG_NF_TPROXY_IPV6) += nf_tproxy_ipv6.o
 
 # logging
 obj-$(CONFIG_NF_LOG_IPV6) += nf_log_ipv6.o

net/ipv6/netfilter/nf_tproxy_ipv6.c

+#include <net/netfilter/nf_tproxy.h>
+#include <linux/module.h>
+#include <net/inet6_hashtables.h>
+#include <net/addrconf.h>
+#include <net/udp.h>
+#include <net/tcp.h>
+
+const struct in6_addr *
+nf_tproxy_laddr6(struct sk_buff *skb, const struct in6_addr *user_laddr,
+	      const struct in6_addr *daddr)
+{
+	struct inet6_dev *indev;
+	struct inet6_ifaddr *ifa;
+	struct in6_addr *laddr;
+
+	if (!ipv6_addr_any(user_laddr))
+		return user_laddr;
+	laddr = NULL;
+
+	indev = __in6_dev_get(skb->dev);
+	if (indev) {
+		read_lock_bh(&indev->lock);
+		list_for_each_entry(ifa, &indev->addr_list, if_list) {
+			if (ifa->flags & (IFA_F_TENTATIVE | IFA_F_DEPRECATED))
+				continue;
+
+			laddr = &ifa->addr;
+			break;
+		}
+		read_unlock_bh(&indev->lock);
+	}
+
+	return laddr ? laddr : daddr;
+}
+EXPORT_SYMBOL_GPL(nf_tproxy_laddr6);
+
+struct sock *
+nf_tproxy_handle_time_wait6(struct sk_buff *skb, int tproto, int thoff,
+			 struct net *net,
+			 const struct in6_addr *laddr,
+			 const __be16 lport,
+			 struct sock *sk)
+{
+	const struct ipv6hdr *iph = ipv6_hdr(skb);
+	struct tcphdr _hdr, *hp;
+
+	hp = skb_header_pointer(skb, thoff, sizeof(_hdr), &_hdr);
+	if (hp == NULL) {
+		inet_twsk_put(inet_twsk(sk));
+		return NULL;
+	}
+
+	if (hp->syn && !hp->rst && !hp->ack && !hp->fin) {
+		/* SYN to a TIME_WAIT socket, we'd rather redirect it
+		 * to a listener socket if there's one */
+		struct sock *sk2;
+
+		sk2 = nf_tproxy_get_sock_v6(net, skb, thoff, hp, tproto,
+					    &iph->saddr,
+					    nf_tproxy_laddr6(skb, laddr, &iph->daddr),
+					    hp->source,
+					    lport ? lport : hp->dest,
+					    skb->dev, NF_TPROXY_LOOKUP_LISTENER);
+		if (sk2) {
+			inet_twsk_deschedule_put(inet_twsk(sk));
+			sk = sk2;
+		}
+	}
+
+	return sk;
+}
+EXPORT_SYMBOL_GPL(nf_tproxy_handle_time_wait6);
+
+struct sock *
+nf_tproxy_get_sock_v6(struct net *net, struct sk_buff *skb, int thoff, void *hp,
+		      const u8 protocol,
+		      const struct in6_addr *saddr, const struct in6_addr *daddr,
+		      const __be16 sport, const __be16 dport,
+		      const struct net_device *in,
+		      const enum nf_tproxy_lookup_t lookup_type)
+{
+	struct sock *sk;
+	struct tcphdr *tcph;
+
+	switch (protocol) {
+	case IPPROTO_TCP:
+		switch (lookup_type) {
+		case NF_TPROXY_LOOKUP_LISTENER:
+			tcph = hp;
+			sk = inet6_lookup_listener(net, &tcp_hashinfo, skb,
+						   thoff + __tcp_hdrlen(tcph),
+						   saddr, sport,
+						   daddr, ntohs(dport),
+						   in->ifindex, 0);
+
+			if (sk && !refcount_inc_not_zero(&sk->sk_refcnt))
+				sk = NULL;
+			/* NOTE: we return listeners even if bound to
+			 * 0.0.0.0, those are filtered out in
+			 * xt_socket, since xt_TPROXY needs 0 bound
+			 * listeners too
+			 */
+			break;
+		case NF_TPROXY_LOOKUP_ESTABLISHED:
+			sk = __inet6_lookup_established(net, &tcp_hashinfo,
+							saddr, sport, daddr, ntohs(dport),
+							in->ifindex, 0);
+			break;
+		default:
+			BUG();
+		}
+		break;
+	case IPPROTO_UDP:
+		sk = udp6_lib_lookup(net, saddr, sport, daddr, dport,
+				     in->ifindex);
+		if (sk) {
+			int connected = (sk->sk_state == TCP_ESTABLISHED);
+			int wildcard = ipv6_addr_any(&sk->sk_v6_rcv_saddr);
+
+			/* NOTE: we return listeners even if bound to
+			 * 0.0.0.0, those are filtered out in
+			 * xt_socket, since xt_TPROXY needs 0 bound
+			 * listeners too
+			 */
+			if ((lookup_type == NF_TPROXY_LOOKUP_ESTABLISHED && (!connected || wildcard)) ||
+			    (lookup_type == NF_TPROXY_LOOKUP_LISTENER && connected)) {
+				sock_put(sk);
+				sk = NULL;
+			}
+		}
+		break;
+	default:
+		WARN_ON(1);
+		sk = NULL;
+	}
+
+	pr_debug("tproxy socket lookup: proto %u %pI6:%u -> %pI6:%u, lookup type: %d, sock %p\n",
+		 protocol, saddr, ntohs(sport), daddr, ntohs(dport), lookup_type, sk);
+
+	return sk;
+}
+EXPORT_SYMBOL_GPL(nf_tproxy_get_sock_v6);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Balazs Scheidler, Krisztian Kovacs");
+MODULE_DESCRIPTION("Netfilter IPv4 transparent proxy support");

net/netfilter/Kconfig

@@ -517,6 +517,15 @@ config NFT_COUNTER
 	  This option adds the "counter" expression that you can use to
 	  include packet and byte counters in a rule.
 
+config NFT_CONNLIMIT
+	tristate "Netfilter nf_tables connlimit module"
+	depends on NF_CONNTRACK
+	depends on NETFILTER_ADVANCED
+	select NETFILTER_CONNCOUNT
+	help
+	  This option adds the "connlimit" expression that you can use to
+	  ratelimit rule matchings per connections.
+
 config NFT_LOG
 	tristate "Netfilter nf_tables log module"
 	help
@@ -989,6 +998,8 @@ config NETFILTER_XT_TARGET_TPROXY
 	depends on IP_NF_MANGLE
 	select NF_DEFRAG_IPV4
 	select NF_DEFRAG_IPV6 if IP6_NF_IPTABLES != n
+	select NF_TPROXY_IPV4
+	select NF_TPROXY_IPV6 if IP6_NF_IPTABLES
 	help
 	  This option adds a `TPROXY' target, which is somewhat similar to
 	  REDIRECT.  It can only be used in the mangle table and is useful

net/netfilter/Makefile

@@ -80,6 +80,7 @@ nf_tables-objs := nf_tables_core.o nf_tables_api.o nft_chain_filter.o \
 
 obj-$(CONFIG_NF_TABLES)		+= nf_tables.o
 obj-$(CONFIG_NFT_COMPAT)	+= nft_compat.o
+obj-$(CONFIG_NFT_CONNLIMIT)	+= nft_connlimit.o
 obj-$(CONFIG_NFT_NUMGEN)	+= nft_numgen.o
 obj-$(CONFIG_NFT_CT)		+= nft_ct.o
 obj-$(CONFIG_NFT_FLOW_OFFLOAD)	+= nft_flow_offload.o

net/netfilter/nf_conncount.c

@@ -79,7 +79,7 @@ static int key_diff(const u32 *a, const u32 *b, unsigned int klen)
 	return memcmp(a, b, klen * sizeof(u32));
 }
 
-static bool add_hlist(struct hlist_head *head,
+bool nf_conncount_add(struct hlist_head *head,
 		      const struct nf_conntrack_tuple *tuple)
 {
 	struct nf_conncount_tuple *conn;
@@ -91,9 +91,9 @@ static bool add_hlist(struct hlist_head *head,
 	hlist_add_head(&conn->node, head);
 	return true;
 }
+EXPORT_SYMBOL_GPL(nf_conncount_add);
 
-static unsigned int check_hlist(struct net *net,
-				struct hlist_head *head,
+unsigned int nf_conncount_lookup(struct net *net, struct hlist_head *head,
 				 const struct nf_conntrack_tuple *tuple,
 				 const struct nf_conntrack_zone *zone,
 				 bool *addit)
@@ -141,6 +141,7 @@ static unsigned int check_hlist(struct net *net,
 
 	return length;
 }
+EXPORT_SYMBOL_GPL(nf_conncount_lookup);
 
 static void tree_nodes_free(struct rb_root *root,
 			    struct nf_conncount_rb *gc_nodes[],
@@ -187,13 +188,15 @@ count_tree(struct net *net, struct rb_root *root,
 		} else {
 			/* same source network -> be counted! */
 			unsigned int count;
-			count = check_hlist(net, &rbconn->hhead, tuple, zone, &addit);
+
+			count = nf_conncount_lookup(net, &rbconn->hhead, tuple,
+						    zone, &addit);
 
 			tree_nodes_free(root, gc_nodes, gc_count);
 			if (!addit)
 				return count;
 
-			if (!add_hlist(&rbconn->hhead, tuple))
+			if (!nf_conncount_add(&rbconn->hhead, tuple))
 				return 0; /* hotdrop */
 
 			return count + 1;
@@ -203,7 +206,7 @@ count_tree(struct net *net, struct rb_root *root,
 			continue;
 
 		/* only used for GC on hhead, retval and 'addit' ignored */
-		check_hlist(net, &rbconn->hhead, tuple, zone, &addit);
+		nf_conncount_lookup(net, &rbconn->hhead, tuple, zone, &addit);
 		if (hlist_empty(&rbconn->hhead))
 			gc_nodes[gc_count++] = rbconn;
 	}
@@ -303,11 +306,19 @@ struct nf_conncount_data *nf_conncount_init(struct net *net, unsigned int family
 }
 EXPORT_SYMBOL_GPL(nf_conncount_init);
 
-static void destroy_tree(struct rb_root *r)
+void nf_conncount_cache_free(struct hlist_head *hhead)
 {
 	struct nf_conncount_tuple *conn;
-	struct nf_conncount_rb *rbconn;
 	struct hlist_node *n;
+
+	hlist_for_each_entry_safe(conn, n, hhead, node)
+		kmem_cache_free(conncount_conn_cachep, conn);
+}
+EXPORT_SYMBOL_GPL(nf_conncount_cache_free);
+
+static void destroy_tree(struct rb_root *r)
+{
+	struct nf_conncount_rb *rbconn;
 	struct rb_node *node;
 
 	while ((node = rb_first(r)) != NULL) {
@@ -315,8 +326,7 @@ static void destroy_tree(struct rb_root *r)
 
 		rb_erase(node, r);
 
-		hlist_for_each_entry_safe(conn, n, &rbconn->hhead, node)
-			kmem_cache_free(conncount_conn_cachep, conn);
+		nf_conncount_cache_free(&rbconn->hhead);
 
 		kmem_cache_free(conncount_rb_cachep, rbconn);
 	}

net/netfilter/nft_connlimit.c

+/* SPDX-License-Identifier: GPL-2.0 */
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/spinlock.h>
+#include <linux/netlink.h>
+#include <linux/netfilter.h>
+#include <linux/netfilter/nf_tables.h>
+#include <net/netfilter/nf_tables.h>
+#include <net/netfilter/nf_conntrack.h>
+#include <net/netfilter/nf_conntrack_count.h>
+#include <net/netfilter/nf_conntrack_core.h>
+#include <net/netfilter/nf_conntrack_tuple.h>
+#include <net/netfilter/nf_conntrack_zones.h>
+
+struct nft_connlimit {
+	spinlock_t		lock;
+	struct hlist_head	hhead;
+	u32			limit;
+	bool			invert;
+};
+
+static inline void nft_connlimit_do_eval(struct nft_connlimit *priv,
+					 struct nft_regs *regs,
+					 const struct nft_pktinfo *pkt,
+					 const struct nft_set_ext *ext)
+{
+	const struct nf_conntrack_zone *zone = &nf_ct_zone_dflt;
+	const struct nf_conntrack_tuple *tuple_ptr;
+	struct nf_conntrack_tuple tuple;
+	enum ip_conntrack_info ctinfo;
+	const struct nf_conn *ct;
+	unsigned int count;
+	bool addit;
+
+	tuple_ptr = &tuple;
+
+	ct = nf_ct_get(pkt->skb, &ctinfo);
+	if (ct != NULL) {
+		tuple_ptr = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple;
+		zone = nf_ct_zone(ct);
+	} else if (!nf_ct_get_tuplepr(pkt->skb, skb_network_offset(pkt->skb),
+				      nft_pf(pkt), nft_net(pkt), &tuple)) {
+		regs->verdict.code = NF_DROP;
+		return;
+	}
+
+	spin_lock_bh(&priv->lock);
+	count = nf_conncount_lookup(nft_net(pkt), &priv->hhead, tuple_ptr, zone,
+				    &addit);
+
+	if (!addit)
+		goto out;
+
+	if (!nf_conncount_add(&priv->hhead, tuple_ptr)) {
+		regs->verdict.code = NF_DROP;
+		spin_unlock_bh(&priv->lock);
+		return;
+	}
+	count++;
+out:
+	spin_unlock_bh(&priv->lock);
+
+	if ((count > priv->limit) ^ priv->invert) {
+		regs->verdict.code = NFT_BREAK;
+		return;
+	}
+}
+
+static int nft_connlimit_do_init(const struct nft_ctx *ctx,
+				 const struct nlattr * const tb[],
+				 struct nft_connlimit *priv)
+{
+	bool invert = false;
+	u32 flags, limit;
+
+	if (!tb[NFTA_CONNLIMIT_COUNT])
+		return -EINVAL;
+
+	limit = ntohl(nla_get_be32(tb[NFTA_CONNLIMIT_COUNT]));
+
+	if (tb[NFTA_CONNLIMIT_FLAGS]) {
+		flags = ntohl(nla_get_be32(tb[NFTA_CONNLIMIT_FLAGS]));
+		if (flags & ~NFT_CONNLIMIT_F_INV)
+			return -EOPNOTSUPP;
+		if (flags & NFT_CONNLIMIT_F_INV)
+			invert = true;
+	}
+
+	spin_lock_init(&priv->lock);
+	INIT_HLIST_HEAD(&priv->hhead);
+	priv->limit	= limit;
+	priv->invert	= invert;
+
+	return nf_ct_netns_get(ctx->net, ctx->family);
+}
+
+static void nft_connlimit_do_destroy(const struct nft_ctx *ctx,
+				     struct nft_connlimit *priv)
+{
+	nf_ct_netns_put(ctx->net, ctx->family);
+	nf_conncount_cache_free(&priv->hhead);
+}
+
+static int nft_connlimit_do_dump(struct sk_buff *skb,
+				 struct nft_connlimit *priv)
+{
+	if (nla_put_be32(skb, NFTA_CONNLIMIT_COUNT, htonl(priv->limit)))
+		goto nla_put_failure;
+	if (priv->invert &&
+	    nla_put_be32(skb, NFTA_CONNLIMIT_FLAGS, htonl(NFT_CONNLIMIT_F_INV)))
+		goto nla_put_failure;
+
+	return 0;
+
+nla_put_failure:
+	return -1;
+}
+
+static inline void nft_connlimit_obj_eval(struct nft_object *obj,
+					struct nft_regs *regs,
+					const struct nft_pktinfo *pkt)
+{
+	struct nft_connlimit *priv = nft_obj_data(obj);
+
+	nft_connlimit_do_eval(priv, regs, pkt, NULL);
+}
+
+static int nft_connlimit_obj_init(const struct nft_ctx *ctx,
+				const struct nlattr * const tb[],
+				struct nft_object *obj)
+{
+	struct nft_connlimit *priv = nft_obj_data(obj);
+
+	return nft_connlimit_do_init(ctx, tb, priv);
+}
+
+static void nft_connlimit_obj_destroy(const struct nft_ctx *ctx,
+				      struct nft_object *obj)
+{
+	struct nft_connlimit *priv = nft_obj_data(obj);
+
+	nft_connlimit_do_destroy(ctx, priv);
+}
+
+static int nft_connlimit_obj_dump(struct sk_buff *skb,
+				  struct nft_object *obj, bool reset)
+{
+	struct nft_connlimit *priv = nft_obj_data(obj);
+
+	return nft_connlimit_do_dump(skb, priv);
+}
+
+static const struct nla_policy nft_connlimit_policy[NFTA_CONNLIMIT_MAX + 1] = {
+	[NFTA_CONNLIMIT_COUNT]	= { .type = NLA_U32 },
+	[NFTA_CONNLIMIT_FLAGS]	= { .type = NLA_U32 },
+};
+
+static struct nft_object_type nft_connlimit_obj_type;
+static const struct nft_object_ops nft_connlimit_obj_ops = {
+	.type		= &nft_connlimit_obj_type,
+	.size		= sizeof(struct nft_connlimit),
+	.eval		= nft_connlimit_obj_eval,
+	.init		= nft_connlimit_obj_init,
+	.destroy	= nft_connlimit_obj_destroy,
+	.dump		= nft_connlimit_obj_dump,
+};
+
+static struct nft_object_type nft_connlimit_obj_type __read_mostly = {
+	.type		= NFT_OBJECT_CONNLIMIT,
+	.ops		= &nft_connlimit_obj_ops,
+	.maxattr	= NFTA_CONNLIMIT_MAX,
+	.policy		= nft_connlimit_policy,
+	.owner		= THIS_MODULE,
+};
+
+static void nft_connlimit_eval(const struct nft_expr *expr,
+			       struct nft_regs *regs,
+			       const struct nft_pktinfo *pkt)
+{
+	struct nft_connlimit *priv = nft_expr_priv(expr);
+
+	nft_connlimit_do_eval(priv, regs, pkt, NULL);
+}
+
+static int nft_connlimit_dump(struct sk_buff *skb, const struct nft_expr *expr)
+{
+	struct nft_connlimit *priv = nft_expr_priv(expr);
+
+	return nft_connlimit_do_dump(skb, priv);
+}
+
+static int nft_connlimit_init(const struct nft_ctx *ctx,
+			      const struct nft_expr *expr,
+			      const struct nlattr * const tb[])
+{
+	struct nft_connlimit *priv = nft_expr_priv(expr);
+
+	return nft_connlimit_do_init(ctx, tb, priv);
+}
+
+static void nft_connlimit_destroy(const struct nft_ctx *ctx,
+				const struct nft_expr *expr)
+{
+	struct nft_connlimit *priv = nft_expr_priv(expr);
+
+	nft_connlimit_do_destroy(ctx, priv);
+}
+
+static int nft_connlimit_clone(struct nft_expr *dst, const struct nft_expr *src)
+{
+	struct nft_connlimit *priv_dst = nft_expr_priv(dst);
+	struct nft_connlimit *priv_src = nft_expr_priv(src);
+
+	spin_lock_init(&priv_dst->lock);
+	INIT_HLIST_HEAD(&priv_dst->hhead);
+	priv_dst->limit	 = priv_src->limit;
+	priv_dst->invert = priv_src->invert;
+
+	return 0;
+}
+
+static void nft_connlimit_destroy_clone(const struct nft_ctx *ctx,
+					const struct nft_expr *expr)
+{
+	struct nft_connlimit *priv = nft_expr_priv(expr);
+
+	nf_conncount_cache_free(&priv->hhead);
+}
+
+static bool nft_connlimit_gc(struct net *net, const struct nft_expr *expr)
+{
+	struct nft_connlimit *priv = nft_expr_priv(expr);
+	bool addit, ret;
+
+	spin_lock_bh(&priv->lock);
+	nf_conncount_lookup(net, &priv->hhead, NULL, &nf_ct_zone_dflt, &addit);
+
+	ret = hlist_empty(&priv->hhead);
+	spin_unlock_bh(&priv->lock);
+
+	return ret;
+}
+
+static struct nft_expr_type nft_connlimit_type;
+static const struct nft_expr_ops nft_connlimit_ops = {
+	.type		= &nft_connlimit_type,
+	.size		= NFT_EXPR_SIZE(sizeof(struct nft_connlimit)),
+	.eval		= nft_connlimit_eval,
+	.init		= nft_connlimit_init,
+	.destroy	= nft_connlimit_destroy,
+	.clone		= nft_connlimit_clone,
+	.destroy_clone	= nft_connlimit_destroy_clone,
+	.dump		= nft_connlimit_dump,
+	.gc		= nft_connlimit_gc,
+};
+
+static struct nft_expr_type nft_connlimit_type __read_mostly = {
+	.name		= "connlimit",
+	.ops		= &nft_connlimit_ops,
+	.policy		= nft_connlimit_policy,
+	.maxattr	= NFTA_CONNLIMIT_MAX,
+	.flags		= NFT_EXPR_STATEFUL | NFT_EXPR_GC,
+	.owner		= THIS_MODULE,
+};
+
+static int __init nft_connlimit_module_init(void)
+{
+	int err;
+
+	err = nft_register_obj(&nft_connlimit_obj_type);
+	if (err < 0)
+		return err;
+
+	err = nft_register_expr(&nft_connlimit_type);
+	if (err < 0)
+		goto err1;
+
+	return 0;
+err1:
+	nft_unregister_obj(&nft_connlimit_obj_type);
+	return err;
+}
+
+static void __exit nft_connlimit_module_exit(void)
+{
+	nft_unregister_expr(&nft_connlimit_type);
+	nft_unregister_obj(&nft_connlimit_obj_type);
+}
+
+module_init(nft_connlimit_module_init);
+module_exit(nft_connlimit_module_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Pablo Neira Ayuso");
+MODULE_ALIAS_NFT_EXPR("connlimit");
+MODULE_ALIAS_NFT_OBJ(NFT_OBJECT_CONNLIMIT);

net/netfilter/nft_counter.c

@@ -96,7 +96,8 @@ static void nft_counter_do_destroy(struct nft_counter_percpu_priv *priv)
 	free_percpu(priv->counter);
 }
 
-static void nft_counter_obj_destroy(struct nft_object *obj)
+static void nft_counter_obj_destroy(const struct nft_ctx *ctx,
+				    struct nft_object *obj)
 {
 	struct nft_counter_percpu_priv *priv = nft_obj_data(obj);
 
@@ -257,6 +258,7 @@ static const struct nft_expr_ops nft_counter_ops = {
 	.eval		= nft_counter_eval,
 	.init		= nft_counter_init,
 	.destroy	= nft_counter_destroy,
+	.destroy_clone	= nft_counter_destroy,
 	.dump		= nft_counter_dump,
 	.clone		= nft_counter_clone,
 };

net/netfilter/nft_ct.c

@@ -826,7 +826,8 @@ static int nft_ct_helper_obj_init(const struct nft_ctx *ctx,
 	return 0;
 }
 
-static void nft_ct_helper_obj_destroy(struct nft_object *obj)
+static void nft_ct_helper_obj_destroy(const struct nft_ctx *ctx,
+				      struct nft_object *obj)
 {
 	struct nft_ct_helper_obj *priv = nft_obj_data(obj);
 

net/netfilter/nft_dynset.c

@@ -195,6 +195,15 @@ static int nft_dynset_init(const struct nft_ctx *ctx,
 		err = -EOPNOTSUPP;
 		if (!(priv->expr->ops->type->flags & NFT_EXPR_STATEFUL))
 			goto err1;
+
+		if (priv->expr->ops->type->flags & NFT_EXPR_GC) {
+			if (set->flags & NFT_SET_TIMEOUT)
+				goto err1;
+			if (!set->ops->gc_init)
+				goto err1;
+			set->ops->gc_init(set);
+		}
+
 	} else if (set->flags & NFT_SET_EVAL)
 		return -EINVAL;
 

net/netfilter/nft_set_hash.c

@@ -311,8 +311,16 @@ static void nft_rhash_gc(struct work_struct *work)
 			continue;
 		}
 
+		if (nft_set_ext_exists(&he->ext, NFT_SET_EXT_EXPR)) {
+			struct nft_expr *expr = nft_set_ext_expr(&he->ext);
+
+			if (expr->ops->gc &&
+			    expr->ops->gc(read_pnet(&set->net), expr))
+				goto gc;
+		}
 		if (!nft_set_elem_expired(&he->ext))
 			continue;
+gc:
 		if (nft_set_elem_mark_busy(&he->ext))
 			continue;
 
@@ -339,6 +347,14 @@ static unsigned int nft_rhash_privsize(const struct nlattr * const nla[],
 	return sizeof(struct nft_rhash);
 }
 
+static void nft_rhash_gc_init(const struct nft_set *set)
+{
+	struct nft_rhash *priv = nft_set_priv(set);
+
+	queue_delayed_work(system_power_efficient_wq, &priv->gc_work,
+			   nft_set_gc_interval(set));
+}
+
 static int nft_rhash_init(const struct nft_set *set,
 			  const struct nft_set_desc *desc,
 			  const struct nlattr * const tb[])
@@ -356,8 +372,8 @@ static int nft_rhash_init(const struct nft_set *set,
 
 	INIT_DEFERRABLE_WORK(&priv->gc_work, nft_rhash_gc);
 	if (set->flags & NFT_SET_TIMEOUT)
-		queue_delayed_work(system_power_efficient_wq, &priv->gc_work,
-				   nft_set_gc_interval(set));
+		nft_rhash_gc_init(set);
+
 	return 0;
 }
 
@@ -647,6 +663,7 @@ static struct nft_set_type nft_rhash_type __read_mostly = {
 		.elemsize	= offsetof(struct nft_rhash_elem, ext),
 		.estimate	= nft_rhash_estimate,
 		.init		= nft_rhash_init,
+		.gc_init	= nft_rhash_gc_init,
 		.destroy	= nft_rhash_destroy,
 		.insert		= nft_rhash_insert,
 		.activate	= nft_rhash_activate,

net/netfilter/nft_socket.c

@@ -5,6 +5,7 @@
 #include <net/netfilter/nf_tables_core.h>
 #include <net/netfilter/nf_socket.h>
 #include <net/inet_sock.h>
+#include <net/tcp.h>
 
 struct nft_socket {
 	enum nft_socket_keys		key:8;
@@ -48,7 +49,7 @@ static void nft_socket_eval(const struct nft_expr *expr,
 
 	switch(priv->key) {
 	case NFT_SOCKET_TRANSPARENT:
-		nft_reg_store8(dest, nf_sk_is_transparent(sk));
+		nft_reg_store8(dest, inet_sk_transparent(sk));
 		break;
 	default:
 		WARN_ON(1);

net/netfilter/xt_socket.c

@@ -73,7 +73,7 @@ socket_match(const struct sk_buff *skb, struct xt_action_param *par,
 		 * if XT_SOCKET_TRANSPARENT is used
 		 */
 		if (info->flags & XT_SOCKET_TRANSPARENT)
-			transparent = nf_sk_is_transparent(sk);
+			transparent = inet_sk_transparent(sk);
 
 		if (info->flags & XT_SOCKET_RESTORESKMARK && !wildcard &&
 		    transparent && sk_fullsock(sk))
@@ -130,7 +130,7 @@ socket_mt6_v1_v2_v3(const struct sk_buff *skb, struct xt_action_param *par)
 		 * if XT_SOCKET_TRANSPARENT is used
 		 */
 		if (info->flags & XT_SOCKET_TRANSPARENT)
-			transparent = nf_sk_is_transparent(sk);
+			transparent = inet_sk_transparent(sk);
 
 		if (info->flags & XT_SOCKET_RESTORESKMARK && !wildcard &&
 		    transparent && sk_fullsock(sk))