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

Pablo Neira Ayuso says:

====================
Netfilter/IPVS updates for net-next

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

1) Add abstracted ICMP codes to the nf_tables reject expression. We
   introduce four reasons to reject using ICMP that overlap in IPv4
   and IPv6 from the semantic point of view. This should simplify the
   maintainance of dual stack rule-sets through the inet table.

2) Move nf_send_reset() functions from header files to per-family
   nf_reject modules, suggested by Patrick McHardy.

3) We have to use IS_ENABLED(CONFIG_BRIDGE_NETFILTER) everywhere in the
   code now that br_netfilter can be modularized. Convert remaining spots
   in the network stack code.

4) Use rcu_barrier() in the nf_tables module removal path to ensure that
   we don't leave object that are still pending to be released via
   call_rcu (that may likely result in a crash).

5) Remove incomplete arch 32/64 compat from nft_compat. The original (bad)
   idea was to probe the word size based on the xtables match/target info
   size, but this assumption is wrong when you have to dump the information
   back to userspace.

6) Allow to filter from prerouting and postrouting in the nf_tables bridge.
   In order to emulate the ebtables NAT chains (which are actually simple
   filter chains with no special semantics), we have support filtering from
   this hooks too.

7) Add explicit module dependency between xt_physdev and br_netfilter.
   This provides a way to detect if the user needs br_netfilter from
   the configuration path. This should reduce the breakage of the
   br_netfilter modularization.

8) Cleanup coding style in ip_vs.h, from Simon Horman.

9) Fix crash in the recently added nf_tables masq expression. We have
   to register/unregister the notifiers to clean up the conntrack table
   entries from the module init/exit path, not from the rule addition /
   deletion path. From Arturo Borrero.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/netfilter/br_netfilter.h

+#ifndef _BR_NETFILTER_H_
+#define _BR_NETFILTER_H_
+
+void br_netfilter_enable(void);
+
+#endif /* _BR_NETFILTER_H_ */

include/net/netfilter/ipv4/nf_reject.h

 #ifndef _IPV4_NF_REJECT_H
 #define _IPV4_NF_REJECT_H
 
-#include <net/ip.h>
-#include <net/tcp.h>
-#include <net/route.h>
-#include <net/dst.h>
+#include <net/icmp.h>
 
 static inline void nf_send_unreach(struct sk_buff *skb_in, int code)
 {
 	icmp_send(skb_in, ICMP_DEST_UNREACH, code, 0);
 }
 
-/* Send RST reply */
-static void nf_send_reset(struct sk_buff *oldskb, int hook)
-{
-	struct sk_buff *nskb;
-	const struct iphdr *oiph;
-	struct iphdr *niph;
-	const struct tcphdr *oth;
-	struct tcphdr _otcph, *tcph;
-
-	/* IP header checks: fragment. */
-	if (ip_hdr(oldskb)->frag_off & htons(IP_OFFSET))
-		return;
-
-	oth = skb_header_pointer(oldskb, ip_hdrlen(oldskb),
-				 sizeof(_otcph), &_otcph);
-	if (oth == NULL)
-		return;
-
-	/* No RST for RST. */
-	if (oth->rst)
-		return;
-
-	if (skb_rtable(oldskb)->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST))
-		return;
-
-	/* Check checksum */
-	if (nf_ip_checksum(oldskb, hook, ip_hdrlen(oldskb), IPPROTO_TCP))
-		return;
-	oiph = ip_hdr(oldskb);
-
-	nskb = alloc_skb(sizeof(struct iphdr) + sizeof(struct tcphdr) +
-			 LL_MAX_HEADER, GFP_ATOMIC);
-	if (!nskb)
-		return;
-
-	skb_reserve(nskb, LL_MAX_HEADER);
-
-	skb_reset_network_header(nskb);
-	niph = (struct iphdr *)skb_put(nskb, sizeof(struct iphdr));
-	niph->version	= 4;
-	niph->ihl	= sizeof(struct iphdr) / 4;
-	niph->tos	= 0;
-	niph->id	= 0;
-	niph->frag_off	= htons(IP_DF);
-	niph->protocol	= IPPROTO_TCP;
-	niph->check	= 0;
-	niph->saddr	= oiph->daddr;
-	niph->daddr	= oiph->saddr;
-
-	skb_reset_transport_header(nskb);
-	tcph = (struct tcphdr *)skb_put(nskb, sizeof(struct tcphdr));
-	memset(tcph, 0, sizeof(*tcph));
-	tcph->source	= oth->dest;
-	tcph->dest	= oth->source;
-	tcph->doff	= sizeof(struct tcphdr) / 4;
-
-	if (oth->ack)
-		tcph->seq = oth->ack_seq;
-	else {
-		tcph->ack_seq = htonl(ntohl(oth->seq) + oth->syn + oth->fin +
-				      oldskb->len - ip_hdrlen(oldskb) -
-				      (oth->doff << 2));
-		tcph->ack = 1;
-	}
-
-	tcph->rst	= 1;
-	tcph->check = ~tcp_v4_check(sizeof(struct tcphdr), niph->saddr,
-				    niph->daddr, 0);
-	nskb->ip_summed = CHECKSUM_PARTIAL;
-	nskb->csum_start = (unsigned char *)tcph - nskb->head;
-	nskb->csum_offset = offsetof(struct tcphdr, check);
-
-	/* ip_route_me_harder expects skb->dst to be set */
-	skb_dst_set_noref(nskb, skb_dst(oldskb));
-
-	nskb->protocol = htons(ETH_P_IP);
-	if (ip_route_me_harder(nskb, RTN_UNSPEC))
-		goto free_nskb;
-
-	niph->ttl	= ip4_dst_hoplimit(skb_dst(nskb));
-
-	/* "Never happens" */
-	if (nskb->len > dst_mtu(skb_dst(nskb)))
-		goto free_nskb;
-
-	nf_ct_attach(nskb, oldskb);
-
-#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
-	/* If we use ip_local_out for bridged traffic, the MAC source on
-	 * the RST will be ours, instead of the destination's.  This confuses
-	 * some routers/firewalls, and they drop the packet.  So we need to
-	 * build the eth header using the original destination's MAC as the
-	 * source, and send the RST packet directly.
-	 */
-	if (oldskb->nf_bridge) {
-		struct ethhdr *oeth = eth_hdr(oldskb);
-		nskb->dev = oldskb->nf_bridge->physindev;
-		niph->tot_len = htons(nskb->len);
-		ip_send_check(niph);
-		if (dev_hard_header(nskb, nskb->dev, ntohs(nskb->protocol),
-				    oeth->h_source, oeth->h_dest, nskb->len) < 0)
-			goto free_nskb;
-		dev_queue_xmit(nskb);
-	} else
-#endif
-		ip_local_out(nskb);
-
-	return;
-
- free_nskb:
-	kfree_skb(nskb);
-}
-
+void nf_send_reset(struct sk_buff *oldskb, int hook);
 
 #endif /* _IPV4_NF_REJECT_H */

include/net/netfilter/nft_reject.h

@@ -14,12 +14,7 @@ int nft_reject_init(const struct nft_ctx *ctx,
 
 int nft_reject_dump(struct sk_buff *skb, const struct nft_expr *expr);
 
-void nft_reject_ipv4_eval(const struct nft_expr *expr,
-			  struct nft_data data[NFT_REG_MAX + 1],
-			  const struct nft_pktinfo *pkt);
-
-void nft_reject_ipv6_eval(const struct nft_expr *expr,
-			  struct nft_data data[NFT_REG_MAX + 1],
-			  const struct nft_pktinfo *pkt);
+int nft_reject_icmp_code(u8 code);
+int nft_reject_icmpv6_code(u8 code);
 
 #endif

include/uapi/linux/netfilter/nf_tables.h

@@ -749,12 +749,33 @@ enum nft_queue_attributes {
  *
  * @NFT_REJECT_ICMP_UNREACH: reject using ICMP unreachable
  * @NFT_REJECT_TCP_RST: reject using TCP RST
+ * @NFT_REJECT_ICMPX_UNREACH: abstracted ICMP unreachable for bridge and inet
  */
 enum nft_reject_types {
 	NFT_REJECT_ICMP_UNREACH,
 	NFT_REJECT_TCP_RST,
+	NFT_REJECT_ICMPX_UNREACH,
 };
 
+/**
+ * enum nft_reject_code - Generic reject codes for IPv4/IPv6
+ *
+ * @NFT_REJECT_ICMPX_NO_ROUTE: no route to host / network unreachable
+ * @NFT_REJECT_ICMPX_PORT_UNREACH: port unreachable
+ * @NFT_REJECT_ICMPX_HOST_UNREACH: host unreachable
+ * @NFT_REJECT_ICMPX_ADMIN_PROHIBITED: administratively prohibited
+ *
+ * These codes are mapped to real ICMP and ICMPv6 codes.
+ */
+enum nft_reject_inet_code {
+	NFT_REJECT_ICMPX_NO_ROUTE	= 0,
+	NFT_REJECT_ICMPX_PORT_UNREACH,
+	NFT_REJECT_ICMPX_HOST_UNREACH,
+	NFT_REJECT_ICMPX_ADMIN_PROHIBITED,
+	__NFT_REJECT_ICMPX_MAX
+};
+#define NFT_REJECT_ICMPX_MAX	(__NFT_REJECT_ICMPX_MAX + 1)
+
 /**
  * enum nft_reject_attributes - nf_tables reject expression netlink attributes
  *

net/bridge/br_netfilter.c

@@ -856,6 +856,11 @@ static unsigned int ip_sabotage_in(const struct nf_hook_ops *ops,
 	return NF_ACCEPT;
 }
 
+void br_netfilter_enable(void)
+{
+}
+EXPORT_SYMBOL_GPL(br_netfilter_enable);
+
 /* For br_nf_post_routing, we need (prio = NF_BR_PRI_LAST), because
  * br_dev_queue_push_xmit is called afterwards */
 static struct nf_hook_ops br_nf_ops[] __read_mostly = {

net/bridge/netfilter/nf_tables_bridge.c

@@ -34,9 +34,11 @@ static struct nft_af_info nft_af_bridge __read_mostly = {
 	.owner		= THIS_MODULE,
 	.nops		= 1,
 	.hooks		= {
+		[NF_BR_PRE_ROUTING]	= nft_do_chain_bridge,
 		[NF_BR_LOCAL_IN]	= nft_do_chain_bridge,
 		[NF_BR_FORWARD]		= nft_do_chain_bridge,
 		[NF_BR_LOCAL_OUT]	= nft_do_chain_bridge,
+		[NF_BR_POST_ROUTING]	= nft_do_chain_bridge,
 	},
 };
 

net/bridge/netfilter/nft_reject_bridge.c

@@ -14,21 +14,106 @@
 #include <linux/netfilter/nf_tables.h>
 #include <net/netfilter/nf_tables.h>
 #include <net/netfilter/nft_reject.h>
+#include <net/netfilter/ipv4/nf_reject.h>
+#include <net/netfilter/ipv6/nf_reject.h>
 
 static void nft_reject_bridge_eval(const struct nft_expr *expr,
 				 struct nft_data data[NFT_REG_MAX + 1],
 				 const struct nft_pktinfo *pkt)
 {
+	struct nft_reject *priv = nft_expr_priv(expr);
+	struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out);
+
 	switch (eth_hdr(pkt->skb)->h_proto) {
 	case htons(ETH_P_IP):
-		return nft_reject_ipv4_eval(expr, data, pkt);
+		switch (priv->type) {
+		case NFT_REJECT_ICMP_UNREACH:
+			nf_send_unreach(pkt->skb, priv->icmp_code);
+			break;
+		case NFT_REJECT_TCP_RST:
+			nf_send_reset(pkt->skb, pkt->ops->hooknum);
+			break;
+		case NFT_REJECT_ICMPX_UNREACH:
+			nf_send_unreach(pkt->skb,
+					nft_reject_icmp_code(priv->icmp_code));
+			break;
+		}
+		break;
 	case htons(ETH_P_IPV6):
-		return nft_reject_ipv6_eval(expr, data, pkt);
+		switch (priv->type) {
+		case NFT_REJECT_ICMP_UNREACH:
+			nf_send_unreach6(net, pkt->skb, priv->icmp_code,
+					 pkt->ops->hooknum);
+			break;
+		case NFT_REJECT_TCP_RST:
+			nf_send_reset6(net, pkt->skb, pkt->ops->hooknum);
+			break;
+		case NFT_REJECT_ICMPX_UNREACH:
+			nf_send_unreach6(net, pkt->skb,
+					 nft_reject_icmpv6_code(priv->icmp_code),
+					 pkt->ops->hooknum);
+			break;
+		}
+		break;
 	default:
 		/* No explicit way to reject this protocol, drop it. */
+		break;
+	}
 	data[NFT_REG_VERDICT].verdict = NF_DROP;
+}
+
+static int nft_reject_bridge_init(const struct nft_ctx *ctx,
+				  const struct nft_expr *expr,
+				  const struct nlattr * const tb[])
+{
+	struct nft_reject *priv = nft_expr_priv(expr);
+	int icmp_code;
+
+	if (tb[NFTA_REJECT_TYPE] == NULL)
+		return -EINVAL;
+
+	priv->type = ntohl(nla_get_be32(tb[NFTA_REJECT_TYPE]));
+	switch (priv->type) {
+	case NFT_REJECT_ICMP_UNREACH:
+	case NFT_REJECT_ICMPX_UNREACH:
+		if (tb[NFTA_REJECT_ICMP_CODE] == NULL)
+			return -EINVAL;
+
+		icmp_code = nla_get_u8(tb[NFTA_REJECT_ICMP_CODE]);
+		if (priv->type == NFT_REJECT_ICMPX_UNREACH &&
+		    icmp_code > NFT_REJECT_ICMPX_MAX)
+			return -EINVAL;
+
+		priv->icmp_code = icmp_code;
 		break;
+	case NFT_REJECT_TCP_RST:
+		break;
+	default:
+		return -EINVAL;
 	}
+	return 0;
+}
+
+static int nft_reject_bridge_dump(struct sk_buff *skb,
+				  const struct nft_expr *expr)
+{
+	const struct nft_reject *priv = nft_expr_priv(expr);
+
+	if (nla_put_be32(skb, NFTA_REJECT_TYPE, htonl(priv->type)))
+		goto nla_put_failure;
+
+	switch (priv->type) {
+	case NFT_REJECT_ICMP_UNREACH:
+	case NFT_REJECT_ICMPX_UNREACH:
+		if (nla_put_u8(skb, NFTA_REJECT_ICMP_CODE, priv->icmp_code))
+			goto nla_put_failure;
+		break;
+	}
+
+	return 0;
+
+nla_put_failure:
+	return -1;
 }
 
 static struct nft_expr_type nft_reject_bridge_type;
@@ -36,8 +121,8 @@ static const struct nft_expr_ops nft_reject_bridge_ops = {
 	.type		= &nft_reject_bridge_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_reject)),
 	.eval		= nft_reject_bridge_eval,
-	.init		= nft_reject_init,
-	.dump		= nft_reject_dump,
+	.init		= nft_reject_bridge_init,
+	.dump		= nft_reject_bridge_dump,
 };
 
 static struct nft_expr_type nft_reject_bridge_type __read_mostly = {

net/core/skbuff.c

@@ -572,7 +572,7 @@ static void skb_release_head_state(struct sk_buff *skb)
 #if IS_ENABLED(CONFIG_NF_CONNTRACK)
 	nf_conntrack_put(skb->nfct);
 #endif
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 	nf_bridge_put(skb->nf_bridge);
 #endif
 /* XXX: IS this still necessary? - JHS */

net/ipv4/ip_output.c

@@ -516,7 +516,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
 
 	hlen = iph->ihl * 4;
 	mtu = mtu - hlen;	/* Size of data space */
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 	if (skb->nf_bridge)
 		mtu -= nf_bridge_mtu_reduction(skb);
 #endif

net/ipv4/netfilter/Kconfig

@@ -61,8 +61,13 @@ config NFT_CHAIN_ROUTE_IPV4
 	  fields such as the source, destination, type of service and
 	  the packet mark.
 
+config NF_REJECT_IPV4
+	tristate "IPv4 packet rejection"
+	default m if NETFILTER_ADVANCED=n
+
 config NFT_REJECT_IPV4
 	depends on NF_TABLES_IPV4
+	select NF_REJECT_IPV4
 	default NFT_REJECT
 	tristate
 
@@ -208,6 +213,7 @@ config IP_NF_FILTER
 config IP_NF_TARGET_REJECT
 	tristate "REJECT target support"
 	depends on IP_NF_FILTER
+	select NF_REJECT_IPV4
 	default m if NETFILTER_ADVANCED=n
 	help
 	  The REJECT target allows a filtering rule to specify that an ICMP

net/ipv4/netfilter/Makefile

@@ -23,6 +23,9 @@ obj-$(CONFIG_NF_DEFRAG_IPV4) += nf_defrag_ipv4.o
 obj-$(CONFIG_NF_LOG_ARP) += nf_log_arp.o
 obj-$(CONFIG_NF_LOG_IPV4) += nf_log_ipv4.o
 
+# reject
+obj-$(CONFIG_NF_REJECT_IPV4) += nf_reject_ipv4.o
+
 # NAT helpers (nf_conntrack)
 obj-$(CONFIG_NF_NAT_H323) += nf_nat_h323.o
 obj-$(CONFIG_NF_NAT_PPTP) += nf_nat_pptp.o

net/ipv4/netfilter/ipt_REJECT.c

@@ -20,7 +20,7 @@
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter_ipv4/ip_tables.h>
 #include <linux/netfilter_ipv4/ipt_REJECT.h>
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 #include <linux/netfilter_bridge.h>
 #endif
 

net/ipv4/netfilter/nf_defrag_ipv4.c

@@ -50,7 +50,7 @@ static enum ip_defrag_users nf_ct_defrag_user(unsigned int hooknum,
 		zone = nf_ct_zone((struct nf_conn *)skb->nfct);
 #endif
 
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 	if (skb->nf_bridge &&
 	    skb->nf_bridge->mask & BRNF_NF_BRIDGE_PREROUTING)
 		return IP_DEFRAG_CONNTRACK_BRIDGE_IN + zone;

net/ipv4/netfilter/nf_reject_ipv4.c

+/* (C) 1999-2001 Paul `Rusty' Russell
+ * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org>
+ *
+ * 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/ip.h>
+#include <net/tcp.h>
+#include <net/route.h>
+#include <net/dst.h>
+#include <linux/netfilter_ipv4.h>
+
+/* Send RST reply */
+void nf_send_reset(struct sk_buff *oldskb, int hook)
+{
+	struct sk_buff *nskb;
+	const struct iphdr *oiph;
+	struct iphdr *niph;
+	const struct tcphdr *oth;
+	struct tcphdr _otcph, *tcph;
+
+	/* IP header checks: fragment. */
+	if (ip_hdr(oldskb)->frag_off & htons(IP_OFFSET))
+		return;
+
+	oth = skb_header_pointer(oldskb, ip_hdrlen(oldskb),
+				 sizeof(_otcph), &_otcph);
+	if (oth == NULL)
+		return;
+
+	/* No RST for RST. */
+	if (oth->rst)
+		return;
+
+	if (skb_rtable(oldskb)->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST))
+		return;
+
+	/* Check checksum */
+	if (nf_ip_checksum(oldskb, hook, ip_hdrlen(oldskb), IPPROTO_TCP))
+		return;
+	oiph = ip_hdr(oldskb);
+
+	nskb = alloc_skb(sizeof(struct iphdr) + sizeof(struct tcphdr) +
+			 LL_MAX_HEADER, GFP_ATOMIC);
+	if (!nskb)
+		return;
+
+	skb_reserve(nskb, LL_MAX_HEADER);
+
+	skb_reset_network_header(nskb);
+	niph = (struct iphdr *)skb_put(nskb, sizeof(struct iphdr));
+	niph->version	= 4;
+	niph->ihl	= sizeof(struct iphdr) / 4;
+	niph->tos	= 0;
+	niph->id	= 0;
+	niph->frag_off	= htons(IP_DF);
+	niph->protocol	= IPPROTO_TCP;
+	niph->check	= 0;
+	niph->saddr	= oiph->daddr;
+	niph->daddr	= oiph->saddr;
+
+	skb_reset_transport_header(nskb);
+	tcph = (struct tcphdr *)skb_put(nskb, sizeof(struct tcphdr));
+	memset(tcph, 0, sizeof(*tcph));
+	tcph->source	= oth->dest;
+	tcph->dest	= oth->source;
+	tcph->doff	= sizeof(struct tcphdr) / 4;
+
+	if (oth->ack)
+		tcph->seq = oth->ack_seq;
+	else {
+		tcph->ack_seq = htonl(ntohl(oth->seq) + oth->syn + oth->fin +
+				      oldskb->len - ip_hdrlen(oldskb) -
+				      (oth->doff << 2));
+		tcph->ack = 1;
+	}
+
+	tcph->rst	= 1;
+	tcph->check = ~tcp_v4_check(sizeof(struct tcphdr), niph->saddr,
+				    niph->daddr, 0);
+	nskb->ip_summed = CHECKSUM_PARTIAL;
+	nskb->csum_start = (unsigned char *)tcph - nskb->head;
+	nskb->csum_offset = offsetof(struct tcphdr, check);
+
+	/* ip_route_me_harder expects skb->dst to be set */
+	skb_dst_set_noref(nskb, skb_dst(oldskb));
+
+	nskb->protocol = htons(ETH_P_IP);
+	if (ip_route_me_harder(nskb, RTN_UNSPEC))
+		goto free_nskb;
+
+	niph->ttl	= ip4_dst_hoplimit(skb_dst(nskb));
+
+	/* "Never happens" */
+	if (nskb->len > dst_mtu(skb_dst(nskb)))
+		goto free_nskb;
+
+	nf_ct_attach(nskb, oldskb);
+
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
+	/* If we use ip_local_out for bridged traffic, the MAC source on
+	 * the RST will be ours, instead of the destination's.  This confuses
+	 * some routers/firewalls, and they drop the packet.  So we need to
+	 * build the eth header using the original destination's MAC as the
+	 * source, and send the RST packet directly.
+	 */
+	if (oldskb->nf_bridge) {
+		struct ethhdr *oeth = eth_hdr(oldskb);
+		nskb->dev = oldskb->nf_bridge->physindev;
+		niph->tot_len = htons(nskb->len);
+		ip_send_check(niph);
+		if (dev_hard_header(nskb, nskb->dev, ntohs(nskb->protocol),
+				    oeth->h_source, oeth->h_dest, nskb->len) < 0)
+			goto free_nskb;
+		dev_queue_xmit(nskb);
+	} else
+#endif
+		ip_local_out(nskb);
+
+	return;
+
+ free_nskb:
+	kfree_skb(nskb);
+}
+EXPORT_SYMBOL_GPL(nf_send_reset);

net/ipv4/netfilter/nft_masq_ipv4.c

@@ -32,33 +32,12 @@ static void nft_masq_ipv4_eval(const struct nft_expr *expr,
 	data[NFT_REG_VERDICT].verdict = verdict;
 }
 
-static int nft_masq_ipv4_init(const struct nft_ctx *ctx,
-			      const struct nft_expr *expr,
-			      const struct nlattr * const tb[])
-{
-	int err;
-
-	err = nft_masq_init(ctx, expr, tb);
-	if (err < 0)
-		return err;
-
-	nf_nat_masquerade_ipv4_register_notifier();
-	return 0;
-}
-
-static void nft_masq_ipv4_destroy(const struct nft_ctx *ctx,
-				  const struct nft_expr *expr)
-{
-	nf_nat_masquerade_ipv4_unregister_notifier();
-}
-
 static struct nft_expr_type nft_masq_ipv4_type;
 static const struct nft_expr_ops nft_masq_ipv4_ops = {
 	.type		= &nft_masq_ipv4_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_masq)),
 	.eval		= nft_masq_ipv4_eval,
-	.init		= nft_masq_ipv4_init,
-	.destroy	= nft_masq_ipv4_destroy,
+	.init		= nft_masq_init,
 	.dump		= nft_masq_dump,
 };
 
@@ -73,12 +52,21 @@ static struct nft_expr_type nft_masq_ipv4_type __read_mostly = {
 
 static int __init nft_masq_ipv4_module_init(void)
 {
-	return nft_register_expr(&nft_masq_ipv4_type);
+	int ret;
+
+	ret = nft_register_expr(&nft_masq_ipv4_type);
+	if (ret < 0)
+		return ret;
+
+	nf_nat_masquerade_ipv4_register_notifier();
+
+	return ret;
 }
 
 static void __exit nft_masq_ipv4_module_exit(void)
 {
 	nft_unregister_expr(&nft_masq_ipv4_type);
+	nf_nat_masquerade_ipv4_unregister_notifier();
 }
 
 module_init(nft_masq_ipv4_module_init);

net/ipv4/netfilter/nft_reject_ipv4.c

@@ -16,7 +16,6 @@
 #include <linux/netfilter.h>
 #include <linux/netfilter/nf_tables.h>
 #include <net/netfilter/nf_tables.h>
-#include <net/icmp.h>
 #include <net/netfilter/ipv4/nf_reject.h>
 #include <net/netfilter/nft_reject.h>
 

net/ipv6/netfilter/Kconfig

@@ -40,8 +40,13 @@ config NFT_CHAIN_ROUTE_IPV6
 	  fields such as the source, destination, flowlabel, hop-limit and
 	  the packet mark.
 
+config NF_REJECT_IPV6
+	tristate "IPv6 packet rejection"
+	default m if NETFILTER_ADVANCED=n
+
 config NFT_REJECT_IPV6
 	depends on NF_TABLES_IPV6
+	select NF_REJECT_IPV6
 	default NFT_REJECT
 	tristate
 
@@ -208,6 +213,7 @@ config IP6_NF_FILTER
 config IP6_NF_TARGET_REJECT
 	tristate "REJECT target support"
 	depends on IP6_NF_FILTER
+	select NF_REJECT_IPV6
 	default m if NETFILTER_ADVANCED=n
 	help
 	  The REJECT target allows a filtering rule to specify that an ICMPv6

net/ipv6/netfilter/Makefile

@@ -27,6 +27,9 @@ obj-$(CONFIG_NF_DEFRAG_IPV6) += nf_defrag_ipv6.o
 # logging
 obj-$(CONFIG_NF_LOG_IPV6) += nf_log_ipv6.o
 
+# reject
+obj-$(CONFIG_NF_REJECT_IPV6) += nf_reject_ipv6.o
+
 # nf_tables
 obj-$(CONFIG_NF_TABLES_IPV6) += nf_tables_ipv6.o
 obj-$(CONFIG_NFT_CHAIN_ROUTE_IPV6) += nft_chain_route_ipv6.o

net/ipv6/netfilter/nf_defrag_ipv6_hooks.c

@@ -40,7 +40,7 @@ static enum ip6_defrag_users nf_ct6_defrag_user(unsigned int hooknum,
 		zone = nf_ct_zone((struct nf_conn *)skb->nfct);
 #endif
 
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 	if (skb->nf_bridge &&
 	    skb->nf_bridge->mask & BRNF_NF_BRIDGE_PREROUTING)
 		return IP6_DEFRAG_CONNTRACK_BRIDGE_IN + zone;

net/ipv6/netfilter/nf_reject_ipv6.c

+/* (C) 1999-2001 Paul `Rusty' Russell
+ * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org>
+ *
+ * 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/ipv6.h>
+#include <net/ip6_route.h>
+#include <net/ip6_fib.h>
+#include <net/ip6_checksum.h>
+#include <linux/netfilter_ipv6.h>
+
+void nf_send_reset6(struct net *net, struct sk_buff *oldskb, int hook)
+{
+	struct sk_buff *nskb;
+	struct tcphdr otcph, *tcph;
+	unsigned int otcplen, hh_len;
+	int tcphoff, needs_ack;
+	const struct ipv6hdr *oip6h = ipv6_hdr(oldskb);
+	struct ipv6hdr *ip6h;
+#define DEFAULT_TOS_VALUE	0x0U
+	const __u8 tclass = DEFAULT_TOS_VALUE;
+	struct dst_entry *dst = NULL;
+	u8 proto;
+	__be16 frag_off;
+	struct flowi6 fl6;
+
+	if ((!(ipv6_addr_type(&oip6h->saddr) & IPV6_ADDR_UNICAST)) ||
+	    (!(ipv6_addr_type(&oip6h->daddr) & IPV6_ADDR_UNICAST))) {
+		pr_debug("addr is not unicast.\n");
+		return;
+	}
+
+	proto = oip6h->nexthdr;
+	tcphoff = ipv6_skip_exthdr(oldskb, ((u8*)(oip6h+1) - oldskb->data), &proto, &frag_off);
+
+	if ((tcphoff < 0) || (tcphoff > oldskb->len)) {
+		pr_debug("Cannot get TCP header.\n");
+		return;
+	}
+
+	otcplen = oldskb->len - tcphoff;
+
+	/* IP header checks: fragment, too short. */
+	if (proto != IPPROTO_TCP || otcplen < sizeof(struct tcphdr)) {
+		pr_debug("proto(%d) != IPPROTO_TCP, "
+			 "or too short. otcplen = %d\n",
+			 proto, otcplen);
+		return;
+	}
+
+	if (skb_copy_bits(oldskb, tcphoff, &otcph, sizeof(struct tcphdr)))
+		BUG();
+
+	/* No RST for RST. */
+	if (otcph.rst) {
+		pr_debug("RST is set\n");
+		return;
+	}
+
+	/* Check checksum. */
+	if (nf_ip6_checksum(oldskb, hook, tcphoff, IPPROTO_TCP)) {
+		pr_debug("TCP checksum is invalid\n");
+		return;
+	}
+
+	memset(&fl6, 0, sizeof(fl6));
+	fl6.flowi6_proto = IPPROTO_TCP;
+	fl6.saddr = oip6h->daddr;
+	fl6.daddr = oip6h->saddr;
+	fl6.fl6_sport = otcph.dest;
+	fl6.fl6_dport = otcph.source;
+	security_skb_classify_flow(oldskb, flowi6_to_flowi(&fl6));
+	dst = ip6_route_output(net, NULL, &fl6);
+	if (dst == NULL || dst->error) {
+		dst_release(dst);
+		return;
+	}
+	dst = xfrm_lookup(net, dst, flowi6_to_flowi(&fl6), NULL, 0);
+	if (IS_ERR(dst))
+		return;
+
+	hh_len = (dst->dev->hard_header_len + 15)&~15;
+	nskb = alloc_skb(hh_len + 15 + dst->header_len + sizeof(struct ipv6hdr)
+			 + sizeof(struct tcphdr) + dst->trailer_len,
+			 GFP_ATOMIC);
+
+	if (!nskb) {
+		net_dbg_ratelimited("cannot alloc skb\n");
+		dst_release(dst);
+		return;
+	}
+
+	skb_dst_set(nskb, dst);
+
+	skb_reserve(nskb, hh_len + dst->header_len);
+
+	skb_put(nskb, sizeof(struct ipv6hdr));
+	skb_reset_network_header(nskb);
+	ip6h = ipv6_hdr(nskb);
+	ip6_flow_hdr(ip6h, tclass, 0);
+	ip6h->hop_limit = ip6_dst_hoplimit(dst);
+	ip6h->nexthdr = IPPROTO_TCP;
+	ip6h->saddr = oip6h->daddr;
+	ip6h->daddr = oip6h->saddr;
+
+	skb_reset_transport_header(nskb);
+	tcph = (struct tcphdr *)skb_put(nskb, sizeof(struct tcphdr));
+	/* Truncate to length (no data) */
+	tcph->doff = sizeof(struct tcphdr)/4;
+	tcph->source = otcph.dest;
+	tcph->dest = otcph.source;
+
+	if (otcph.ack) {
+		needs_ack = 0;
+		tcph->seq = otcph.ack_seq;
+		tcph->ack_seq = 0;
+	} else {
+		needs_ack = 1;
+		tcph->ack_seq = htonl(ntohl(otcph.seq) + otcph.syn + otcph.fin
+				      + otcplen - (otcph.doff<<2));
+		tcph->seq = 0;
+	}
+
+	/* Reset flags */
+	((u_int8_t *)tcph)[13] = 0;
+	tcph->rst = 1;
+	tcph->ack = needs_ack;
+	tcph->window = 0;
+	tcph->urg_ptr = 0;
+	tcph->check = 0;
+
+	/* Adjust TCP checksum */
+	tcph->check = csum_ipv6_magic(&ipv6_hdr(nskb)->saddr,
+				      &ipv6_hdr(nskb)->daddr,
+				      sizeof(struct tcphdr), IPPROTO_TCP,
+				      csum_partial(tcph,
+						   sizeof(struct tcphdr), 0));
+
+	nf_ct_attach(nskb, oldskb);
+
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
+	/* If we use ip6_local_out for bridged traffic, the MAC source on
+	 * the RST will be ours, instead of the destination's.  This confuses
+	 * some routers/firewalls, and they drop the packet.  So we need to
+	 * build the eth header using the original destination's MAC as the
+	 * source, and send the RST packet directly.
+	 */
+	if (oldskb->nf_bridge) {
+		struct ethhdr *oeth = eth_hdr(oldskb);
+		nskb->dev = oldskb->nf_bridge->physindev;
+		nskb->protocol = htons(ETH_P_IPV6);
+		ip6h->payload_len = htons(sizeof(struct tcphdr));
+		if (dev_hard_header(nskb, nskb->dev, ntohs(nskb->protocol),
+				    oeth->h_source, oeth->h_dest, nskb->len) < 0)
+			return;
+		dev_queue_xmit(nskb);
+	} else
+#endif
+		ip6_local_out(nskb);
+}
+EXPORT_SYMBOL_GPL(nf_send_reset6);

net/ipv6/netfilter/nft_masq_ipv6.c

@@ -32,33 +32,12 @@ static void nft_masq_ipv6_eval(const struct nft_expr *expr,
 	data[NFT_REG_VERDICT].verdict = verdict;
 }
 
-static int nft_masq_ipv6_init(const struct nft_ctx *ctx,
-			      const struct nft_expr *expr,
-			      const struct nlattr * const tb[])
-{
-	int err;
-
-	err = nft_masq_init(ctx, expr, tb);
-	if (err < 0)
-		return err;
-
-	nf_nat_masquerade_ipv6_register_notifier();
-	return 0;
-}
-
-static void nft_masq_ipv6_destroy(const struct nft_ctx *ctx,
-				  const struct nft_expr *expr)
-{
-	nf_nat_masquerade_ipv6_unregister_notifier();
-}
-
 static struct nft_expr_type nft_masq_ipv6_type;
 static const struct nft_expr_ops nft_masq_ipv6_ops = {
 	.type		= &nft_masq_ipv6_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_masq)),
 	.eval		= nft_masq_ipv6_eval,
-	.init		= nft_masq_ipv6_init,
-	.destroy	= nft_masq_ipv6_destroy,
+	.init		= nft_masq_init,
 	.dump		= nft_masq_dump,
 };
 
@@ -73,12 +52,21 @@ static struct nft_expr_type nft_masq_ipv6_type __read_mostly = {
 
 static int __init nft_masq_ipv6_module_init(void)
 {
-	return nft_register_expr(&nft_masq_ipv6_type);
+	int ret;
+
+	ret = nft_register_expr(&nft_masq_ipv6_type);
+	if (ret < 0)
+		return ret;
+
+	nf_nat_masquerade_ipv6_register_notifier();
+
+	return ret;
 }
 
 static void __exit nft_masq_ipv6_module_exit(void)
 {
 	nft_unregister_expr(&nft_masq_ipv6_type);
+	nf_nat_masquerade_ipv6_unregister_notifier();
 }
 
 module_init(nft_masq_ipv6_module_init);

net/netfilter/ipset/ip_set_hash_netiface.c

@@ -237,7 +237,7 @@ hash_netiface4_kadt(struct ip_set *set, const struct sk_buff *skb,
 #define SRCDIR		(opt->flags & IPSET_DIM_TWO_SRC)
 
 	if (opt->cmdflags & IPSET_FLAG_PHYSDEV) {
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 		const struct nf_bridge_info *nf_bridge = skb->nf_bridge;
 
 		if (!nf_bridge)
@@ -474,7 +474,7 @@ hash_netiface6_kadt(struct ip_set *set, const struct sk_buff *skb,
 	ip6_netmask(&e.ip, e.cidr);
 
 	if (opt->cmdflags & IPSET_FLAG_PHYSDEV) {
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 		const struct nf_bridge_info *nf_bridge = skb->nf_bridge;
 
 		if (!nf_bridge)

net/netfilter/nf_log_common.c

@@ -158,7 +158,7 @@ nf_log_dump_packet_common(struct nf_log_buf *m, u_int8_t pf,
 	       '0' + loginfo->u.log.level, prefix,
 	       in ? in->name : "",
 	       out ? out->name : "");
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 	if (skb->nf_bridge) {
 		const struct net_device *physindev;
 		const struct net_device *physoutdev;

net/netfilter/nf_queue.c

@@ -52,7 +52,7 @@ void nf_queue_entry_release_refs(struct nf_queue_entry *entry)
 		dev_put(entry->indev);
 	if (entry->outdev)
 		dev_put(entry->outdev);
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 	if (entry->skb->nf_bridge) {
 		struct nf_bridge_info *nf_bridge = entry->skb->nf_bridge;
 
@@ -77,7 +77,7 @@ bool nf_queue_entry_get_refs(struct nf_queue_entry *entry)
 		dev_hold(entry->indev);
 	if (entry->outdev)
 		dev_hold(entry->outdev);
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 	if (entry->skb->nf_bridge) {
 		struct nf_bridge_info *nf_bridge = entry->skb->nf_bridge;
 		struct net_device *physdev;

net/netfilter/nf_tables_api.c

@@ -4163,6 +4163,7 @@ static void __exit nf_tables_module_exit(void)
 {
 	unregister_pernet_subsys(&nf_tables_net_ops);
 	nfnetlink_subsys_unregister(&nf_tables_subsys);
+	rcu_barrier();
 	nf_tables_core_module_exit();
 	kfree(info);
 }

net/netfilter/nfnetlink_log.c

@@ -36,7 +36,7 @@
 
 #include <linux/atomic.h>
 
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 #include "../bridge/br_private.h"
 #endif
 
@@ -429,7 +429,7 @@ __build_packet_message(struct nfnl_log_net *log,
 		goto nla_put_failure;
 
 	if (indev) {
-#ifndef CONFIG_BRIDGE_NETFILTER
+#if !IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 		if (nla_put_be32(inst->skb, NFULA_IFINDEX_INDEV,
 				 htonl(indev->ifindex)))
 			goto nla_put_failure;
@@ -460,7 +460,7 @@ __build_packet_message(struct nfnl_log_net *log,
 	}
 
 	if (outdev) {
-#ifndef CONFIG_BRIDGE_NETFILTER
+#if !IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 		if (nla_put_be32(inst->skb, NFULA_IFINDEX_OUTDEV,
 				 htonl(outdev->ifindex)))
 			goto nla_put_failure;
@@ -640,7 +640,7 @@ nfulnl_log_packet(struct net *net,
 		+ nla_total_size(sizeof(struct nfulnl_msg_packet_hdr))
 		+ nla_total_size(sizeof(u_int32_t))	/* ifindex */
 		+ nla_total_size(sizeof(u_int32_t))	/* ifindex */
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 		+ nla_total_size(sizeof(u_int32_t))	/* ifindex */
 		+ nla_total_size(sizeof(u_int32_t))	/* ifindex */
 #endif

net/netfilter/nfnetlink_queue_core.c

@@ -36,7 +36,7 @@
 
 #include <linux/atomic.h>
 
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 #include "../bridge/br_private.h"
 #endif
 
@@ -302,7 +302,7 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue,
 		+ nla_total_size(sizeof(struct nfqnl_msg_packet_hdr))
 		+ nla_total_size(sizeof(u_int32_t))	/* ifindex */
 		+ nla_total_size(sizeof(u_int32_t))	/* ifindex */
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 		+ nla_total_size(sizeof(u_int32_t))	/* ifindex */
 		+ nla_total_size(sizeof(u_int32_t))	/* ifindex */
 #endif
@@ -380,7 +380,7 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue,
 
 	indev = entry->indev;
 	if (indev) {
-#ifndef CONFIG_BRIDGE_NETFILTER
+#if !IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 		if (nla_put_be32(skb, NFQA_IFINDEX_INDEV, htonl(indev->ifindex)))
 			goto nla_put_failure;
 #else
@@ -410,7 +410,7 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue,
 	}
 
 	if (outdev) {
-#ifndef CONFIG_BRIDGE_NETFILTER
+#if !IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 		if (nla_put_be32(skb, NFQA_IFINDEX_OUTDEV, htonl(outdev->ifindex)))
 			goto nla_put_failure;
 #else
@@ -569,7 +569,7 @@ nf_queue_entry_dup(struct nf_queue_entry *e)
 	return NULL;
 }
 
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 /* When called from bridge netfilter, skb->data must point to MAC header
  * before calling skb_gso_segment(). Else, original MAC header is lost
  * and segmented skbs will be sent to wrong destination.
@@ -763,7 +763,7 @@ dev_cmp(struct nf_queue_entry *entry, unsigned long ifindex)
 	if (entry->outdev)
 		if (entry->outdev->ifindex == ifindex)
 			return 1;
-#ifdef CONFIG_BRIDGE_NETFILTER
+#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 	if (entry->skb->nf_bridge) {
 		if (entry->skb->nf_bridge->physindev &&
 		    entry->skb->nf_bridge->physindev->ifindex == ifindex)

net/netfilter/nft_compat.c

@@ -101,26 +101,12 @@ nft_target_set_tgchk_param(struct xt_tgchk_param *par,
 
 static void target_compat_from_user(struct xt_target *t, void *in, void *out)
 {
-#ifdef CONFIG_COMPAT
-	if (t->compat_from_user) {
 	int pad;
 
-		t->compat_from_user(out, in);
+	memcpy(out, in, t->targetsize);
 	pad = XT_ALIGN(t->targetsize) - t->targetsize;
 	if (pad > 0)
 		memset(out + t->targetsize, 0, pad);
-	} else
-#endif
-		memcpy(out, in, XT_ALIGN(t->targetsize));
-}
-
-static inline int nft_compat_target_offset(struct xt_target *target)
-{
-#ifdef CONFIG_COMPAT
-	return xt_compat_target_offset(target);
-#else
-	return 0;
-#endif
 }
 
 static const struct nla_policy nft_rule_compat_policy[NFTA_RULE_COMPAT_MAX + 1] = {
@@ -208,34 +194,6 @@ nft_target_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr)
 	module_put(target->me);
 }
 
-static int
-target_dump_info(struct sk_buff *skb, const struct xt_target *t, const void *in)
-{
-	int ret;
-
-#ifdef CONFIG_COMPAT
-	if (t->compat_to_user) {
-		mm_segment_t old_fs;
-		void *out;
-
-		out = kmalloc(XT_ALIGN(t->targetsize), GFP_ATOMIC);
-		if (out == NULL)
-			return -ENOMEM;
-
-		/* We want to reuse existing compat_to_user */
-		old_fs = get_fs();
-		set_fs(KERNEL_DS);
-		t->compat_to_user(out, in);
-		set_fs(old_fs);
-		ret = nla_put(skb, NFTA_TARGET_INFO, XT_ALIGN(t->targetsize), out);
-		kfree(out);
-	} else
-#endif
-		ret = nla_put(skb, NFTA_TARGET_INFO, XT_ALIGN(t->targetsize), in);
-
-	return ret;
-}
-
 static int nft_target_dump(struct sk_buff *skb, const struct nft_expr *expr)
 {
 	const struct xt_target *target = expr->ops->data;
@@ -243,7 +201,7 @@ static int nft_target_dump(struct sk_buff *skb, const struct nft_expr *expr)
 
 	if (nla_put_string(skb, NFTA_TARGET_NAME, target->name) ||
 	    nla_put_be32(skb, NFTA_TARGET_REV, htonl(target->revision)) ||
-	    target_dump_info(skb, target, info))
+	    nla_put(skb, NFTA_TARGET_INFO, XT_ALIGN(target->targetsize), info))
 		goto nla_put_failure;
 
 	return 0;
@@ -341,17 +299,12 @@ nft_match_set_mtchk_param(struct xt_mtchk_param *par, const struct nft_ctx *ctx,
 
 static void match_compat_from_user(struct xt_match *m, void *in, void *out)
 {
-#ifdef CONFIG_COMPAT
-	if (m->compat_from_user) {
 	int pad;
 
-		m->compat_from_user(out, in);
+	memcpy(out, in, m->matchsize);
 	pad = XT_ALIGN(m->matchsize) - m->matchsize;
 	if (pad > 0)
 		memset(out + m->matchsize, 0, pad);
-	} else
-#endif
-		memcpy(out, in, XT_ALIGN(m->matchsize));
 }
 
 static int
@@ -404,43 +357,6 @@ nft_match_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr)
 	module_put(match->me);
 }
 
-static int
-match_dump_info(struct sk_buff *skb, const struct xt_match *m, const void *in)
-{
-	int ret;
-
-#ifdef CONFIG_COMPAT
-	if (m->compat_to_user) {
-		mm_segment_t old_fs;
-		void *out;
-
-		out = kmalloc(XT_ALIGN(m->matchsize), GFP_ATOMIC);
-		if (out == NULL)
-			return -ENOMEM;
-
-		/* We want to reuse existing compat_to_user */
-		old_fs = get_fs();
-		set_fs(KERNEL_DS);
-		m->compat_to_user(out, in);
-		set_fs(old_fs);
-		ret = nla_put(skb, NFTA_MATCH_INFO, XT_ALIGN(m->matchsize), out);
-		kfree(out);
-	} else
-#endif
-		ret = nla_put(skb, NFTA_MATCH_INFO, XT_ALIGN(m->matchsize), in);
-
-	return ret;
-}
-
-static inline int nft_compat_match_offset(struct xt_match *match)
-{
-#ifdef CONFIG_COMPAT
-	return xt_compat_match_offset(match);
-#else
-	return 0;
-#endif
-}
-
 static int nft_match_dump(struct sk_buff *skb, const struct nft_expr *expr)
 {
 	void *info = nft_expr_priv(expr);
@@ -448,7 +364,7 @@ static int nft_match_dump(struct sk_buff *skb, const struct nft_expr *expr)
 
 	if (nla_put_string(skb, NFTA_MATCH_NAME, match->name) ||
 	    nla_put_be32(skb, NFTA_MATCH_REV, htonl(match->revision)) ||
-	    match_dump_info(skb, match, info))
+	    nla_put(skb, NFTA_MATCH_INFO, XT_ALIGN(match->matchsize), info))
 		goto nla_put_failure;
 
 	return 0;
@@ -643,8 +559,7 @@ nft_match_select_ops(const struct nft_ctx *ctx,
 		return ERR_PTR(-ENOMEM);
 
 	nft_match->ops.type = &nft_match_type;
-	nft_match->ops.size = NFT_EXPR_SIZE(XT_ALIGN(match->matchsize) +
-					    nft_compat_match_offset(match));
+	nft_match->ops.size = NFT_EXPR_SIZE(XT_ALIGN(match->matchsize));
 	nft_match->ops.eval = nft_match_eval;
 	nft_match->ops.init = nft_match_init;
 	nft_match->ops.destroy = nft_match_destroy;
@@ -714,8 +629,7 @@ nft_target_select_ops(const struct nft_ctx *ctx,
 		return ERR_PTR(-ENOMEM);
 
 	nft_target->ops.type = &nft_target_type;
-	nft_target->ops.size = NFT_EXPR_SIZE(XT_ALIGN(target->targetsize) +
-					     nft_compat_target_offset(target));
+	nft_target->ops.size = NFT_EXPR_SIZE(XT_ALIGN(target->targetsize));
 	nft_target->ops.eval = nft_target_eval;
 	nft_target->ops.init = nft_target_init;
 	nft_target->ops.destroy = nft_target_destroy;

net/netfilter/nft_reject.c

@@ -17,6 +17,8 @@
 #include <linux/netfilter/nf_tables.h>
 #include <net/netfilter/nf_tables.h>
 #include <net/netfilter/nft_reject.h>
+#include <linux/icmp.h>
+#include <linux/icmpv6.h>
 
 const struct nla_policy nft_reject_policy[NFTA_REJECT_MAX + 1] = {
 	[NFTA_REJECT_TYPE]		= { .type = NLA_U32 },
@@ -70,5 +72,40 @@ int nft_reject_dump(struct sk_buff *skb, const struct nft_expr *expr)
 }
 EXPORT_SYMBOL_GPL(nft_reject_dump);
 
+static u8 icmp_code_v4[NFT_REJECT_ICMPX_MAX] = {
+	[NFT_REJECT_ICMPX_NO_ROUTE]		= ICMP_NET_UNREACH,
+	[NFT_REJECT_ICMPX_PORT_UNREACH]		= ICMP_PORT_UNREACH,
+	[NFT_REJECT_ICMPX_HOST_UNREACH]		= ICMP_HOST_UNREACH,
+	[NFT_REJECT_ICMPX_ADMIN_PROHIBITED]	= ICMP_PKT_FILTERED,
+};
+
+int nft_reject_icmp_code(u8 code)
+{
+	if (code > NFT_REJECT_ICMPX_MAX)
+		return -EINVAL;
+
+	return icmp_code_v4[code];
+}
+
+EXPORT_SYMBOL_GPL(nft_reject_icmp_code);
+
+
+static u8 icmp_code_v6[NFT_REJECT_ICMPX_MAX] = {
+	[NFT_REJECT_ICMPX_NO_ROUTE]		= ICMPV6_NOROUTE,
+	[NFT_REJECT_ICMPX_PORT_UNREACH]		= ICMPV6_PORT_UNREACH,
+	[NFT_REJECT_ICMPX_HOST_UNREACH]		= ICMPV6_ADDR_UNREACH,
+	[NFT_REJECT_ICMPX_ADMIN_PROHIBITED]	= ICMPV6_ADM_PROHIBITED,
+};
+
+int nft_reject_icmpv6_code(u8 code)
+{
+	if (code > NFT_REJECT_ICMPX_MAX)
+		return -EINVAL;
+
+	return icmp_code_v6[code];
+}
+
+EXPORT_SYMBOL_GPL(nft_reject_icmpv6_code);
+
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");

net/netfilter/nft_reject_inet.c

@@ -14,17 +14,103 @@
 #include <linux/netfilter/nf_tables.h>
 #include <net/netfilter/nf_tables.h>
 #include <net/netfilter/nft_reject.h>
+#include <net/netfilter/ipv4/nf_reject.h>
+#include <net/netfilter/ipv6/nf_reject.h>
 
 static void nft_reject_inet_eval(const struct nft_expr *expr,
 				 struct nft_data data[NFT_REG_MAX + 1],
 				 const struct nft_pktinfo *pkt)
 {
+	struct nft_reject *priv = nft_expr_priv(expr);
+	struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out);
+
 	switch (pkt->ops->pf) {
 	case NFPROTO_IPV4:
-		return nft_reject_ipv4_eval(expr, data, pkt);
+		switch (priv->type) {
+		case NFT_REJECT_ICMP_UNREACH:
+			nf_send_unreach(pkt->skb, priv->icmp_code);
+			break;
+		case NFT_REJECT_TCP_RST:
+			nf_send_reset(pkt->skb, pkt->ops->hooknum);
+			break;
+		case NFT_REJECT_ICMPX_UNREACH:
+			nf_send_unreach(pkt->skb,
+					nft_reject_icmp_code(priv->icmp_code));
+			break;
+		}
+		break;
 	case NFPROTO_IPV6:
-		return nft_reject_ipv6_eval(expr, data, pkt);
+		switch (priv->type) {
+		case NFT_REJECT_ICMP_UNREACH:
+			nf_send_unreach6(net, pkt->skb, priv->icmp_code,
+					 pkt->ops->hooknum);
+			break;
+		case NFT_REJECT_TCP_RST:
+			nf_send_reset6(net, pkt->skb, pkt->ops->hooknum);
+			break;
+		case NFT_REJECT_ICMPX_UNREACH:
+			nf_send_unreach6(net, pkt->skb,
+					 nft_reject_icmpv6_code(priv->icmp_code),
+					 pkt->ops->hooknum);
+			break;
+		}
+		break;
 	}
+	data[NFT_REG_VERDICT].verdict = NF_DROP;
+}
+
+static int nft_reject_inet_init(const struct nft_ctx *ctx,
+				const struct nft_expr *expr,
+				const struct nlattr * const tb[])
+{
+	struct nft_reject *priv = nft_expr_priv(expr);
+	int icmp_code;
+
+	if (tb[NFTA_REJECT_TYPE] == NULL)
+		return -EINVAL;
+
+	priv->type = ntohl(nla_get_be32(tb[NFTA_REJECT_TYPE]));
+	switch (priv->type) {
+	case NFT_REJECT_ICMP_UNREACH:
+	case NFT_REJECT_ICMPX_UNREACH:
+		if (tb[NFTA_REJECT_ICMP_CODE] == NULL)
+			return -EINVAL;
+
+		icmp_code = nla_get_u8(tb[NFTA_REJECT_ICMP_CODE]);
+		if (priv->type == NFT_REJECT_ICMPX_UNREACH &&
+		    icmp_code > NFT_REJECT_ICMPX_MAX)
+			return -EINVAL;
+
+		priv->icmp_code = icmp_code;
+		break;
+	case NFT_REJECT_TCP_RST:
+		break;
+	default:
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int nft_reject_inet_dump(struct sk_buff *skb,
+				const struct nft_expr *expr)
+{
+	const struct nft_reject *priv = nft_expr_priv(expr);
+
+	if (nla_put_be32(skb, NFTA_REJECT_TYPE, htonl(priv->type)))
+		goto nla_put_failure;
+
+	switch (priv->type) {
+	case NFT_REJECT_ICMP_UNREACH:
+	case NFT_REJECT_ICMPX_UNREACH:
+		if (nla_put_u8(skb, NFTA_REJECT_ICMP_CODE, priv->icmp_code))
+			goto nla_put_failure;
+		break;
+	}
+
+	return 0;
+
+nla_put_failure:
+	return -1;
 }
 
 static struct nft_expr_type nft_reject_inet_type;
@@ -32,8 +118,8 @@ static const struct nft_expr_ops nft_reject_inet_ops = {
 	.type		= &nft_reject_inet_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_reject)),
 	.eval		= nft_reject_inet_eval,
-	.init		= nft_reject_init,
-	.dump		= nft_reject_dump,
+	.init		= nft_reject_inet_init,
+	.dump		= nft_reject_inet_dump,
 };
 
 static struct nft_expr_type nft_reject_inet_type __read_mostly = {

net/netfilter/xt_physdev.c

@@ -13,6 +13,7 @@
 #include <linux/netfilter_bridge.h>
 #include <linux/netfilter/xt_physdev.h>
 #include <linux/netfilter/x_tables.h>
+#include <net/netfilter/br_netfilter.h>
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Bart De Schuymer <bdschuym@pandora.be>");
@@ -87,6 +88,8 @@ static int physdev_mt_check(const struct xt_mtchk_param *par)
 {
 	const struct xt_physdev_info *info = par->matchinfo;
 
+	br_netfilter_enable();
+
 	if (!(info->bitmask & XT_PHYSDEV_OP_MASK) ||
 	    info->bitmask & ~XT_PHYSDEV_OP_MASK)
 		return -EINVAL;