From 3c1fece8819ed25257461b71e7c75a1f33eaa61d Mon Sep 17 00:00:00 2001
From: Phil Sutter <phil@nwl.cc>
Date: Mon, 20 Feb 2017 17:52:27 +0100
Subject: [PATCH 01/22] netfilter: nft_exthdr: Allow checking TCP option
 presence, too

Honor NFT_EXTHDR_F_PRESENT flag so we check if the TCP option is
present.

Signed-off-by: Phil Sutter <phil@nwl.cc>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_exthdr.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/net/netfilter/nft_exthdr.c b/net/netfilter/nft_exthdr.c
index c308920b194c..d212a85d2f33 100644
--- a/net/netfilter/nft_exthdr.c
+++ b/net/netfilter/nft_exthdr.c
@@ -98,14 +98,21 @@ static void nft_exthdr_tcp_eval(const struct nft_expr *expr,
 			goto err;
 
 		offset = i + priv->offset;
-		dest[priv->len / NFT_REG32_SIZE] = 0;
-		memcpy(dest, opt + offset, priv->len);
+		if (priv->flags & NFT_EXTHDR_F_PRESENT) {
+			*dest = 1;
+		} else {
+			dest[priv->len / NFT_REG32_SIZE] = 0;
+			memcpy(dest, opt + offset, priv->len);
+		}
 
 		return;
 	}
 
 err:
-	regs->verdict.code = NFT_BREAK;
+	if (priv->flags & NFT_EXTHDR_F_PRESENT)
+		*dest = 0;
+	else
+		regs->verdict.code = NFT_BREAK;
 }
 
 static const struct nla_policy nft_exthdr_policy[NFTA_EXTHDR_MAX + 1] = {

From 511040eea2234d9add3f33ba0e6c2e17944fdfb6 Mon Sep 17 00:00:00 2001
From: Laura Garcia Liebana <nevola@gmail.com>
Date: Thu, 23 Feb 2017 12:10:15 +0100
Subject: [PATCH 02/22] netfilter: nft_hash: rename nft_hash to nft_jhash

This patch renames the local nft_hash structure and functions
to nft_jhash in order to prepare the nft_hash module code to
add new hash functions.

Signed-off-by: Laura Garcia Liebana <laura.garcia@zevenet.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_hash.c | 36 ++++++++++++++++++------------------
 1 file changed, 18 insertions(+), 18 deletions(-)

diff --git a/net/netfilter/nft_hash.c b/net/netfilter/nft_hash.c
index eb2721af898d..ccb834ef049b 100644
--- a/net/netfilter/nft_hash.c
+++ b/net/netfilter/nft_hash.c
@@ -17,7 +17,7 @@
 #include <net/netfilter/nf_tables_core.h>
 #include <linux/jhash.h>
 
-struct nft_hash {
+struct nft_jhash {
 	enum nft_registers      sreg:8;
 	enum nft_registers      dreg:8;
 	u8			len;
@@ -26,11 +26,11 @@ struct nft_hash {
 	u32			offset;
 };
 
-static void nft_hash_eval(const struct nft_expr *expr,
-			  struct nft_regs *regs,
-			  const struct nft_pktinfo *pkt)
+static void nft_jhash_eval(const struct nft_expr *expr,
+			   struct nft_regs *regs,
+			   const struct nft_pktinfo *pkt)
 {
-	struct nft_hash *priv = nft_expr_priv(expr);
+	struct nft_jhash *priv = nft_expr_priv(expr);
 	const void *data = &regs->data[priv->sreg];
 	u32 h;
 
@@ -47,11 +47,11 @@ static const struct nla_policy nft_hash_policy[NFTA_HASH_MAX + 1] = {
 	[NFTA_HASH_OFFSET]	= { .type = NLA_U32 },
 };
 
-static int nft_hash_init(const struct nft_ctx *ctx,
-			 const struct nft_expr *expr,
-			 const struct nlattr * const tb[])
+static int nft_jhash_init(const struct nft_ctx *ctx,
+			  const struct nft_expr *expr,
+			  const struct nlattr * const tb[])
 {
-	struct nft_hash *priv = nft_expr_priv(expr);
+	struct nft_jhash *priv = nft_expr_priv(expr);
 	u32 len;
 	int err;
 
@@ -92,10 +92,10 @@ static int nft_hash_init(const struct nft_ctx *ctx,
 					   NFT_DATA_VALUE, sizeof(u32));
 }
 
-static int nft_hash_dump(struct sk_buff *skb,
-			 const struct nft_expr *expr)
+static int nft_jhash_dump(struct sk_buff *skb,
+			  const struct nft_expr *expr)
 {
-	const struct nft_hash *priv = nft_expr_priv(expr);
+	const struct nft_jhash *priv = nft_expr_priv(expr);
 
 	if (nft_dump_register(skb, NFTA_HASH_SREG, priv->sreg))
 		goto nla_put_failure;
@@ -117,17 +117,17 @@ nla_put_failure:
 }
 
 static struct nft_expr_type nft_hash_type;
-static const struct nft_expr_ops nft_hash_ops = {
+static const struct nft_expr_ops nft_jhash_ops = {
 	.type		= &nft_hash_type,
-	.size		= NFT_EXPR_SIZE(sizeof(struct nft_hash)),
-	.eval		= nft_hash_eval,
-	.init		= nft_hash_init,
-	.dump		= nft_hash_dump,
+	.size		= NFT_EXPR_SIZE(sizeof(struct nft_jhash)),
+	.eval		= nft_jhash_eval,
+	.init		= nft_jhash_init,
+	.dump		= nft_jhash_dump,
 };
 
 static struct nft_expr_type nft_hash_type __read_mostly = {
 	.name		= "hash",
-	.ops		= &nft_hash_ops,
+	.ops		= &nft_jhash_ops,
 	.policy		= nft_hash_policy,
 	.maxattr	= NFTA_HASH_MAX,
 	.owner		= THIS_MODULE,

From 3206caded81ad9bdb2e7ff4c0b94ec5913df8618 Mon Sep 17 00:00:00 2001
From: Laura Garcia Liebana <nevola@gmail.com>
Date: Thu, 2 Mar 2017 17:00:14 +0100
Subject: [PATCH 03/22] netfilter: nft_hash: support of symmetric hash

This patch provides symmetric hash support according to source
ip address and port, and destination ip address and port.

For this purpose, the __skb_get_hash_symmetric() is used to
identify the flow as it uses FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL
flag by default.

The new attribute NFTA_HASH_TYPE has been included to support
different types of hashing functions. Currently supported
NFT_HASH_JENKINS through jhash and NFT_HASH_SYM through symhash.

The main difference between both types are:
 - jhash requires an expression with sreg, symhash doesn't.
 - symhash supports modulus and offset, but not seed.

Examples:

 nft add rule ip nat prerouting ct mark set jhash ip saddr mod 2
 nft add rule ip nat prerouting ct mark set symhash mod 2

By default, jenkins hash will be used if no hash type is
provided for compatibility reasons.

Signed-off-by: Laura Garcia Liebana <laura.garcia@zevenet.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/uapi/linux/netfilter/nf_tables.h | 13 ++++
 net/netfilter/nft_hash.c                 | 99 +++++++++++++++++++++++-
 2 files changed, 111 insertions(+), 1 deletion(-)

diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index 05215d30fe5c..4f7d75682c59 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -815,6 +815,17 @@ enum nft_rt_keys {
 	NFT_RT_NEXTHOP6,
 };
 
+/**
+ * enum nft_hash_types - nf_tables hash expression types
+ *
+ * @NFT_HASH_JENKINS: Jenkins Hash
+ * @NFT_HASH_SYM: Symmetric Hash
+ */
+enum nft_hash_types {
+	NFT_HASH_JENKINS,
+	NFT_HASH_SYM,
+};
+
 /**
  * enum nft_hash_attributes - nf_tables hash expression netlink attributes
  *
@@ -824,6 +835,7 @@ enum nft_rt_keys {
  * @NFTA_HASH_MODULUS: modulus value (NLA_U32)
  * @NFTA_HASH_SEED: seed value (NLA_U32)
  * @NFTA_HASH_OFFSET: add this offset value to hash result (NLA_U32)
+ * @NFTA_HASH_TYPE: hash operation (NLA_U32: nft_hash_types)
  */
 enum nft_hash_attributes {
 	NFTA_HASH_UNSPEC,
@@ -833,6 +845,7 @@ enum nft_hash_attributes {
 	NFTA_HASH_MODULUS,
 	NFTA_HASH_SEED,
 	NFTA_HASH_OFFSET,
+	NFTA_HASH_TYPE,
 	__NFTA_HASH_MAX,
 };
 #define NFTA_HASH_MAX	(__NFTA_HASH_MAX - 1)
diff --git a/net/netfilter/nft_hash.c b/net/netfilter/nft_hash.c
index ccb834ef049b..a6a4633725bb 100644
--- a/net/netfilter/nft_hash.c
+++ b/net/netfilter/nft_hash.c
@@ -38,6 +38,25 @@ static void nft_jhash_eval(const struct nft_expr *expr,
 	regs->data[priv->dreg] = h + priv->offset;
 }
 
+struct nft_symhash {
+	enum nft_registers      dreg:8;
+	u32			modulus;
+	u32			offset;
+};
+
+static void nft_symhash_eval(const struct nft_expr *expr,
+			     struct nft_regs *regs,
+			     const struct nft_pktinfo *pkt)
+{
+	struct nft_symhash *priv = nft_expr_priv(expr);
+	struct sk_buff *skb = pkt->skb;
+	u32 h;
+
+	h = reciprocal_scale(__skb_get_hash_symmetric(skb), priv->modulus);
+
+	regs->data[priv->dreg] = h + priv->offset;
+}
+
 static const struct nla_policy nft_hash_policy[NFTA_HASH_MAX + 1] = {
 	[NFTA_HASH_SREG]	= { .type = NLA_U32 },
 	[NFTA_HASH_DREG]	= { .type = NLA_U32 },
@@ -45,6 +64,7 @@ static const struct nla_policy nft_hash_policy[NFTA_HASH_MAX + 1] = {
 	[NFTA_HASH_MODULUS]	= { .type = NLA_U32 },
 	[NFTA_HASH_SEED]	= { .type = NLA_U32 },
 	[NFTA_HASH_OFFSET]	= { .type = NLA_U32 },
+	[NFTA_HASH_TYPE]	= { .type = NLA_U32 },
 };
 
 static int nft_jhash_init(const struct nft_ctx *ctx,
@@ -92,6 +112,32 @@ static int nft_jhash_init(const struct nft_ctx *ctx,
 					   NFT_DATA_VALUE, sizeof(u32));
 }
 
+static int nft_symhash_init(const struct nft_ctx *ctx,
+			    const struct nft_expr *expr,
+			    const struct nlattr * const tb[])
+{
+	struct nft_symhash *priv = nft_expr_priv(expr);
+
+	if (!tb[NFTA_HASH_DREG]    ||
+	    !tb[NFTA_HASH_MODULUS])
+		return -EINVAL;
+
+	if (tb[NFTA_HASH_OFFSET])
+		priv->offset = ntohl(nla_get_be32(tb[NFTA_HASH_OFFSET]));
+
+	priv->dreg = nft_parse_register(tb[NFTA_HASH_DREG]);
+
+	priv->modulus = ntohl(nla_get_be32(tb[NFTA_HASH_MODULUS]));
+	if (priv->modulus <= 1)
+		return -ERANGE;
+
+	if (priv->offset + priv->modulus - 1 < priv->offset)
+		return -EOVERFLOW;
+
+	return nft_validate_register_store(ctx, priv->dreg, NULL,
+					   NFT_DATA_VALUE, sizeof(u32));
+}
+
 static int nft_jhash_dump(struct sk_buff *skb,
 			  const struct nft_expr *expr)
 {
@@ -110,6 +156,28 @@ static int nft_jhash_dump(struct sk_buff *skb,
 	if (priv->offset != 0)
 		if (nla_put_be32(skb, NFTA_HASH_OFFSET, htonl(priv->offset)))
 			goto nla_put_failure;
+	if (nla_put_be32(skb, NFTA_HASH_TYPE, htonl(NFT_HASH_JENKINS)))
+		goto nla_put_failure;
+	return 0;
+
+nla_put_failure:
+	return -1;
+}
+
+static int nft_symhash_dump(struct sk_buff *skb,
+			    const struct nft_expr *expr)
+{
+	const struct nft_symhash *priv = nft_expr_priv(expr);
+
+	if (nft_dump_register(skb, NFTA_HASH_DREG, priv->dreg))
+		goto nla_put_failure;
+	if (nla_put_be32(skb, NFTA_HASH_MODULUS, htonl(priv->modulus)))
+		goto nla_put_failure;
+	if (priv->offset != 0)
+		if (nla_put_be32(skb, NFTA_HASH_OFFSET, htonl(priv->offset)))
+			goto nla_put_failure;
+	if (nla_put_be32(skb, NFTA_HASH_TYPE, htonl(NFT_HASH_SYM)))
+		goto nla_put_failure;
 	return 0;
 
 nla_put_failure:
@@ -125,9 +193,38 @@ static const struct nft_expr_ops nft_jhash_ops = {
 	.dump		= nft_jhash_dump,
 };
 
+static const struct nft_expr_ops nft_symhash_ops = {
+	.type		= &nft_hash_type,
+	.size		= NFT_EXPR_SIZE(sizeof(struct nft_symhash)),
+	.eval		= nft_symhash_eval,
+	.init		= nft_symhash_init,
+	.dump		= nft_symhash_dump,
+};
+
+static const struct nft_expr_ops *
+nft_hash_select_ops(const struct nft_ctx *ctx,
+		    const struct nlattr * const tb[])
+{
+	u32 type;
+
+	if (!tb[NFTA_HASH_TYPE])
+		return &nft_jhash_ops;
+
+	type = ntohl(nla_get_be32(tb[NFTA_HASH_TYPE]));
+	switch (type) {
+	case NFT_HASH_SYM:
+		return &nft_symhash_ops;
+	case NFT_HASH_JENKINS:
+		return &nft_jhash_ops;
+	default:
+		break;
+	}
+	return ERR_PTR(-EOPNOTSUPP);
+}
+
 static struct nft_expr_type nft_hash_type __read_mostly = {
 	.name		= "hash",
-	.ops		= &nft_jhash_ops,
+	.select_ops	= &nft_hash_select_ops,
 	.policy		= nft_hash_policy,
 	.maxattr	= NFTA_HASH_MAX,
 	.owner		= THIS_MODULE,

From 13fa745da251606fcdbf58acd6b4a551bea6ae99 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Tue, 28 Feb 2017 14:09:24 -0800
Subject: [PATCH 04/22] netfilter: Use pr_cont where appropriate

Logging output was changed when simple printks without KERN_CONT
are now emitted on a new line and KERN_CONT is required to continue
lines so use pr_cont.

Miscellanea:

o realign arguments
o use print_hex_dump instead of a local variant

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/bridge/netfilter/ebt_log.c         | 34 +++++++++++++-------------
 net/ipv4/netfilter/nf_nat_snmp_basic.c | 15 ++----------
 2 files changed, 19 insertions(+), 30 deletions(-)

diff --git a/net/bridge/netfilter/ebt_log.c b/net/bridge/netfilter/ebt_log.c
index 98b9c8e8615e..707caea39743 100644
--- a/net/bridge/netfilter/ebt_log.c
+++ b/net/bridge/netfilter/ebt_log.c
@@ -62,10 +62,10 @@ print_ports(const struct sk_buff *skb, uint8_t protocol, int offset)
 		pptr = skb_header_pointer(skb, offset,
 					  sizeof(_ports), &_ports);
 		if (pptr == NULL) {
-			printk(" INCOMPLETE TCP/UDP header");
+			pr_cont(" INCOMPLETE TCP/UDP header");
 			return;
 		}
-		printk(" SPT=%u DPT=%u", ntohs(pptr->src), ntohs(pptr->dst));
+		pr_cont(" SPT=%u DPT=%u", ntohs(pptr->src), ntohs(pptr->dst));
 	}
 }
 
@@ -100,11 +100,11 @@ ebt_log_packet(struct net *net, u_int8_t pf, unsigned int hooknum,
 
 		ih = skb_header_pointer(skb, 0, sizeof(_iph), &_iph);
 		if (ih == NULL) {
-			printk(" INCOMPLETE IP header");
+			pr_cont(" INCOMPLETE IP header");
 			goto out;
 		}
-		printk(" IP SRC=%pI4 IP DST=%pI4, IP tos=0x%02X, IP proto=%d",
-		       &ih->saddr, &ih->daddr, ih->tos, ih->protocol);
+		pr_cont(" IP SRC=%pI4 IP DST=%pI4, IP tos=0x%02X, IP proto=%d",
+			&ih->saddr, &ih->daddr, ih->tos, ih->protocol);
 		print_ports(skb, ih->protocol, ih->ihl*4);
 		goto out;
 	}
@@ -120,11 +120,11 @@ ebt_log_packet(struct net *net, u_int8_t pf, unsigned int hooknum,
 
 		ih = skb_header_pointer(skb, 0, sizeof(_iph), &_iph);
 		if (ih == NULL) {
-			printk(" INCOMPLETE IPv6 header");
+			pr_cont(" INCOMPLETE IPv6 header");
 			goto out;
 		}
-		printk(" IPv6 SRC=%pI6 IPv6 DST=%pI6, IPv6 priority=0x%01X, Next Header=%d",
-		       &ih->saddr, &ih->daddr, ih->priority, ih->nexthdr);
+		pr_cont(" IPv6 SRC=%pI6 IPv6 DST=%pI6, IPv6 priority=0x%01X, Next Header=%d",
+			&ih->saddr, &ih->daddr, ih->priority, ih->nexthdr);
 		nexthdr = ih->nexthdr;
 		offset_ph = ipv6_skip_exthdr(skb, sizeof(_iph), &nexthdr, &frag_off);
 		if (offset_ph == -1)
@@ -142,12 +142,12 @@ ebt_log_packet(struct net *net, u_int8_t pf, unsigned int hooknum,
 
 		ah = skb_header_pointer(skb, 0, sizeof(_arph), &_arph);
 		if (ah == NULL) {
-			printk(" INCOMPLETE ARP header");
+			pr_cont(" INCOMPLETE ARP header");
 			goto out;
 		}
-		printk(" ARP HTYPE=%d, PTYPE=0x%04x, OPCODE=%d",
-		       ntohs(ah->ar_hrd), ntohs(ah->ar_pro),
-		       ntohs(ah->ar_op));
+		pr_cont(" ARP HTYPE=%d, PTYPE=0x%04x, OPCODE=%d",
+			ntohs(ah->ar_hrd), ntohs(ah->ar_pro),
+			ntohs(ah->ar_op));
 
 		/* If it's for Ethernet and the lengths are OK,
 		 * then log the ARP payload
@@ -161,17 +161,17 @@ ebt_log_packet(struct net *net, u_int8_t pf, unsigned int hooknum,
 			ap = skb_header_pointer(skb, sizeof(_arph),
 						sizeof(_arpp), &_arpp);
 			if (ap == NULL) {
-				printk(" INCOMPLETE ARP payload");
+				pr_cont(" INCOMPLETE ARP payload");
 				goto out;
 			}
-			printk(" ARP MAC SRC=%pM ARP IP SRC=%pI4 ARP MAC DST=%pM ARP IP DST=%pI4",
-					ap->mac_src, ap->ip_src, ap->mac_dst, ap->ip_dst);
+			pr_cont(" ARP MAC SRC=%pM ARP IP SRC=%pI4 ARP MAC DST=%pM ARP IP DST=%pI4",
+				ap->mac_src, ap->ip_src,
+				ap->mac_dst, ap->ip_dst);
 		}
 	}
 out:
-	printk("\n");
+	pr_cont("\n");
 	spin_unlock_bh(&ebt_log_lock);
-
 }
 
 static unsigned int
diff --git a/net/ipv4/netfilter/nf_nat_snmp_basic.c b/net/ipv4/netfilter/nf_nat_snmp_basic.c
index c9b52c361da2..ef49989c93b1 100644
--- a/net/ipv4/netfilter/nf_nat_snmp_basic.c
+++ b/net/ipv4/netfilter/nf_nat_snmp_basic.c
@@ -998,18 +998,6 @@ err_id_free:
  *
  *****************************************************************************/
 
-static void hex_dump(const unsigned char *buf, size_t len)
-{
-	size_t i;
-
-	for (i = 0; i < len; i++) {
-		if (i && !(i % 16))
-			printk("\n");
-		printk("%02x ", *(buf + i));
-	}
-	printk("\n");
-}
-
 /*
  * Parse and mangle SNMP message according to mapping.
  * (And this is the fucking 'basic' method).
@@ -1026,7 +1014,8 @@ static int snmp_parse_mangle(unsigned char *msg,
 	struct snmp_object *obj;
 
 	if (debug > 1)
-		hex_dump(msg, len);
+		print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_NONE, 16, 1,
+			       msg, len, 0);
 
 	asn1_open(&ctx, msg, len);
 

From 74664cf286dc0d2bf6960293cb219e5f741ac92b Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Tue, 28 Feb 2017 11:31:15 +0000
Subject: [PATCH 05/22] netfilter: arp_tables: remove redundant check on ret
 being non-zero

ret is initialized to zero and if it is set to non-zero in the
xt_entry_foreach loop then we exit via the out_free label. Hence
the check for ret being non-zero is redundant and can be removed.

Detected by CoverityScan, CID#1357132 ("Logically Dead Code")

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/ipv4/netfilter/arp_tables.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index 6241a81fd7f5..f17dab1dee6e 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -562,8 +562,6 @@ static int translate_table(struct xt_table_info *newinfo, void *entry0,
 		    XT_ERROR_TARGET) == 0)
 			++newinfo->stacksize;
 	}
-	if (ret != 0)
-		goto out_free;
 
 	ret = -EINVAL;
 	if (i != repl->num_entries)

From c56e3956c17bb24d18470122c0513d963e332205 Mon Sep 17 00:00:00 2001
From: Liping Zhang <zlpnobody@gmail.com>
Date: Sun, 5 Mar 2017 21:02:23 +0800
Subject: [PATCH 06/22] netfilter: nf_tables: validate the expr explicitly
 after init successfully

When we want to validate the expr's dependency or hooks, we must do two
things to accomplish it. First, write a X_validate callback function
and point ->validate to it. Second, call X_validate in init routine.
This is very common, such as fib, nat, reject expr and so on ...

It is a little ugly, since we will call X_validate in the expr's init
routine, it's better to do it in nf_tables_newexpr. So we can avoid to
do this again and again. After doing this, the second step listed above
is not useful anymore, remove them now.

Patch was tested by nftables/tests/py/nft-test.py and
nftables/tests/shell/run-tests.sh.

Signed-off-by: Liping Zhang <zlpnobody@gmail.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/bridge/netfilter/nft_reject_bridge.c |  6 +-----
 net/netfilter/nf_tables_api.c            | 11 +++++++++++
 net/netfilter/nft_compat.c               |  8 --------
 net/netfilter/nft_fib.c                  |  2 +-
 net/netfilter/nft_masq.c                 |  4 ----
 net/netfilter/nft_meta.c                 |  4 ----
 net/netfilter/nft_nat.c                  |  4 ----
 net/netfilter/nft_redir.c                |  4 ----
 net/netfilter/nft_reject.c               |  5 -----
 net/netfilter/nft_reject_inet.c          |  6 +-----
 10 files changed, 14 insertions(+), 40 deletions(-)

diff --git a/net/bridge/netfilter/nft_reject_bridge.c b/net/bridge/netfilter/nft_reject_bridge.c
index 206dc266ecd2..346ef6b00b8f 100644
--- a/net/bridge/netfilter/nft_reject_bridge.c
+++ b/net/bridge/netfilter/nft_reject_bridge.c
@@ -375,11 +375,7 @@ static int nft_reject_bridge_init(const struct nft_ctx *ctx,
 				  const struct nlattr * const tb[])
 {
 	struct nft_reject *priv = nft_expr_priv(expr);
-	int icmp_code, err;
-
-	err = nft_reject_bridge_validate(ctx, expr, NULL);
-	if (err < 0)
-		return err;
+	int icmp_code;
 
 	if (tb[NFTA_REJECT_TYPE] == NULL)
 		return -EINVAL;
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 5e0ccfd5bb37..fd8789eccc92 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -1772,8 +1772,19 @@ static int nf_tables_newexpr(const struct nft_ctx *ctx,
 			goto err1;
 	}
 
+	if (ops->validate) {
+		const struct nft_data *data = NULL;
+
+		err = ops->validate(ctx, expr, &data);
+		if (err < 0)
+			goto err2;
+	}
+
 	return 0;
 
+err2:
+	if (ops->destroy)
+		ops->destroy(ctx, expr);
 err1:
 	expr->ops = NULL;
 	return err;
diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c
index c21e7eb8dce0..fab6bf3f955e 100644
--- a/net/netfilter/nft_compat.c
+++ b/net/netfilter/nft_compat.c
@@ -230,10 +230,6 @@ nft_target_init(const struct nft_ctx *ctx, const struct nft_expr *expr,
 	union nft_entry e = {};
 	int ret;
 
-	ret = nft_compat_chain_validate_dependency(target->table, ctx->chain);
-	if (ret < 0)
-		goto err;
-
 	target_compat_from_user(target, nla_data(tb[NFTA_TARGET_INFO]), info);
 
 	if (ctx->nla[NFTA_RULE_COMPAT]) {
@@ -419,10 +415,6 @@ nft_match_init(const struct nft_ctx *ctx, const struct nft_expr *expr,
 	union nft_entry e = {};
 	int ret;
 
-	ret = nft_compat_chain_validate_dependency(match->table, ctx->chain);
-	if (ret < 0)
-		goto err;
-
 	match_compat_from_user(match, nla_data(tb[NFTA_MATCH_INFO]), info);
 
 	if (ctx->nla[NFTA_RULE_COMPAT]) {
diff --git a/net/netfilter/nft_fib.c b/net/netfilter/nft_fib.c
index 29a4906adc27..fd0b19303b0d 100644
--- a/net/netfilter/nft_fib.c
+++ b/net/netfilter/nft_fib.c
@@ -112,7 +112,7 @@ int nft_fib_init(const struct nft_ctx *ctx, const struct nft_expr *expr,
 	if (err < 0)
 		return err;
 
-	return nft_fib_validate(ctx, expr, NULL);
+	return 0;
 }
 EXPORT_SYMBOL_GPL(nft_fib_init);
 
diff --git a/net/netfilter/nft_masq.c b/net/netfilter/nft_masq.c
index 11ce016cd479..6ac03d4266c9 100644
--- a/net/netfilter/nft_masq.c
+++ b/net/netfilter/nft_masq.c
@@ -46,10 +46,6 @@ int nft_masq_init(const struct nft_ctx *ctx,
 	struct nft_masq *priv = nft_expr_priv(expr);
 	int err;
 
-	err = nft_masq_validate(ctx, expr, NULL);
-	if (err)
-		return err;
-
 	if (tb[NFTA_MASQ_FLAGS]) {
 		priv->flags = ntohl(nla_get_be32(tb[NFTA_MASQ_FLAGS]));
 		if (priv->flags & ~NF_NAT_RANGE_MASK)
diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c
index e1f5ca9b423b..d14417aaf5d4 100644
--- a/net/netfilter/nft_meta.c
+++ b/net/netfilter/nft_meta.c
@@ -370,10 +370,6 @@ int nft_meta_set_init(const struct nft_ctx *ctx,
 		return -EOPNOTSUPP;
 	}
 
-	err = nft_meta_set_validate(ctx, expr, NULL);
-	if (err < 0)
-		return err;
-
 	priv->sreg = nft_parse_register(tb[NFTA_META_SREG]);
 	err = nft_validate_register_load(priv->sreg, len);
 	if (err < 0)
diff --git a/net/netfilter/nft_nat.c b/net/netfilter/nft_nat.c
index 19a7bf3236f9..26a74dfb3b7a 100644
--- a/net/netfilter/nft_nat.c
+++ b/net/netfilter/nft_nat.c
@@ -138,10 +138,6 @@ static int nft_nat_init(const struct nft_ctx *ctx, const struct nft_expr *expr,
 		return -EINVAL;
 	}
 
-	err = nft_nat_validate(ctx, expr, NULL);
-	if (err < 0)
-		return err;
-
 	if (tb[NFTA_NAT_FAMILY] == NULL)
 		return -EINVAL;
 
diff --git a/net/netfilter/nft_redir.c b/net/netfilter/nft_redir.c
index 40dcd05146d5..1e66538bf0ff 100644
--- a/net/netfilter/nft_redir.c
+++ b/net/netfilter/nft_redir.c
@@ -47,10 +47,6 @@ int nft_redir_init(const struct nft_ctx *ctx,
 	unsigned int plen;
 	int err;
 
-	err = nft_redir_validate(ctx, expr, NULL);
-	if (err < 0)
-		return err;
-
 	plen = FIELD_SIZEOF(struct nf_nat_range, min_addr.all);
 	if (tb[NFTA_REDIR_REG_PROTO_MIN]) {
 		priv->sreg_proto_min =
diff --git a/net/netfilter/nft_reject.c b/net/netfilter/nft_reject.c
index c64de3f7379d..29f5bd2377b0 100644
--- a/net/netfilter/nft_reject.c
+++ b/net/netfilter/nft_reject.c
@@ -42,11 +42,6 @@ int nft_reject_init(const struct nft_ctx *ctx,
 		    const struct nlattr * const tb[])
 {
 	struct nft_reject *priv = nft_expr_priv(expr);
-	int err;
-
-	err = nft_reject_validate(ctx, expr, NULL);
-	if (err < 0)
-		return err;
 
 	if (tb[NFTA_REJECT_TYPE] == NULL)
 		return -EINVAL;
diff --git a/net/netfilter/nft_reject_inet.c b/net/netfilter/nft_reject_inet.c
index 9e90a02cb104..5a7fb5ff867d 100644
--- a/net/netfilter/nft_reject_inet.c
+++ b/net/netfilter/nft_reject_inet.c
@@ -66,11 +66,7 @@ static int nft_reject_inet_init(const struct nft_ctx *ctx,
 				const struct nlattr * const tb[])
 {
 	struct nft_reject *priv = nft_expr_priv(expr);
-	int icmp_code, err;
-
-	err = nft_reject_validate(ctx, expr, NULL);
-	if (err < 0)
-		return err;
+	int icmp_code;
 
 	if (tb[NFTA_REJECT_TYPE] == NULL)
 		return -EINVAL;

From c7a72e3fdb5d77486ca3a0ac942c0a2e0d80d5bb Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Mon, 6 Mar 2017 17:46:20 +0100
Subject: [PATCH 07/22] netfilter: nf_tables: add nft_set_lookup()

This new function consolidates set lookup via either name or ID by
introducing a new nft_set_lookup() function. Replace existing spots
where we can use this too.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h |  9 +++++----
 net/netfilter/nf_tables_api.c     | 31 ++++++++++++++++++++++++-------
 net/netfilter/nft_dynset.c        | 14 ++++----------
 net/netfilter/nft_lookup.c        | 14 ++++----------
 net/netfilter/nft_objref.c        | 14 ++++----------
 5 files changed, 41 insertions(+), 41 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index 2aa8a9d80fbe..f0d46726d06e 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -385,10 +385,11 @@ static inline struct nft_set *nft_set_container_of(const void *priv)
 	return (void *)priv - offsetof(struct nft_set, data);
 }
 
-struct nft_set *nf_tables_set_lookup(const struct nft_table *table,
-				     const struct nlattr *nla, u8 genmask);
-struct nft_set *nf_tables_set_lookup_byid(const struct net *net,
-					  const struct nlattr *nla, u8 genmask);
+struct nft_set *nft_set_lookup(const struct net *net,
+			       const struct nft_table *table,
+			       const struct nlattr *nla_set_name,
+			       const struct nlattr *nla_set_id,
+			       u8 genmask);
 
 static inline unsigned long nft_set_gc_interval(const struct nft_set *set)
 {
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index fd8789eccc92..4559f5d66bcc 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -2534,8 +2534,8 @@ static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, struct net *net,
 	return 0;
 }
 
-struct nft_set *nf_tables_set_lookup(const struct nft_table *table,
-				     const struct nlattr *nla, u8 genmask)
+static struct nft_set *nf_tables_set_lookup(const struct nft_table *table,
+					    const struct nlattr *nla, u8 genmask)
 {
 	struct nft_set *set;
 
@@ -2549,11 +2549,10 @@ struct nft_set *nf_tables_set_lookup(const struct nft_table *table,
 	}
 	return ERR_PTR(-ENOENT);
 }
-EXPORT_SYMBOL_GPL(nf_tables_set_lookup);
 
-struct nft_set *nf_tables_set_lookup_byid(const struct net *net,
-					  const struct nlattr *nla,
-					  u8 genmask)
+static struct nft_set *nf_tables_set_lookup_byid(const struct net *net,
+						 const struct nlattr *nla,
+						 u8 genmask)
 {
 	struct nft_trans *trans;
 	u32 id = ntohl(nla_get_be32(nla));
@@ -2568,7 +2567,25 @@ struct nft_set *nf_tables_set_lookup_byid(const struct net *net,
 	}
 	return ERR_PTR(-ENOENT);
 }
-EXPORT_SYMBOL_GPL(nf_tables_set_lookup_byid);
+
+struct nft_set *nft_set_lookup(const struct net *net,
+			       const struct nft_table *table,
+			       const struct nlattr *nla_set_name,
+			       const struct nlattr *nla_set_id,
+			       u8 genmask)
+{
+	struct nft_set *set;
+
+	set = nf_tables_set_lookup(table, nla_set_name, genmask);
+	if (IS_ERR(set)) {
+		if (!nla_set_id)
+			return set;
+
+		set = nf_tables_set_lookup_byid(net, nla_set_id, genmask);
+	}
+	return set;
+}
+EXPORT_SYMBOL_GPL(nft_set_lookup);
 
 static int nf_tables_set_alloc_name(struct nft_ctx *ctx, struct nft_set *set,
 				    const char *name)
diff --git a/net/netfilter/nft_dynset.c b/net/netfilter/nft_dynset.c
index 049ad2d9ee66..3948da380259 100644
--- a/net/netfilter/nft_dynset.c
+++ b/net/netfilter/nft_dynset.c
@@ -133,16 +133,10 @@ static int nft_dynset_init(const struct nft_ctx *ctx,
 			priv->invert = true;
 	}
 
-	set = nf_tables_set_lookup(ctx->table, tb[NFTA_DYNSET_SET_NAME],
-				   genmask);
-	if (IS_ERR(set)) {
-		if (tb[NFTA_DYNSET_SET_ID])
-			set = nf_tables_set_lookup_byid(ctx->net,
-							tb[NFTA_DYNSET_SET_ID],
-							genmask);
-		if (IS_ERR(set))
-			return PTR_ERR(set);
-	}
+	set = nft_set_lookup(ctx->net, ctx->table, tb[NFTA_DYNSET_SET_NAME],
+			     tb[NFTA_DYNSET_SET_ID], genmask);
+	if (IS_ERR(set))
+		return PTR_ERR(set);
 
 	if (set->ops->update == NULL)
 		return -EOPNOTSUPP;
diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c
index e21aea7e5ec8..475570e89ede 100644
--- a/net/netfilter/nft_lookup.c
+++ b/net/netfilter/nft_lookup.c
@@ -71,16 +71,10 @@ static int nft_lookup_init(const struct nft_ctx *ctx,
 	    tb[NFTA_LOOKUP_SREG] == NULL)
 		return -EINVAL;
 
-	set = nf_tables_set_lookup(ctx->table, tb[NFTA_LOOKUP_SET], genmask);
-	if (IS_ERR(set)) {
-		if (tb[NFTA_LOOKUP_SET_ID]) {
-			set = nf_tables_set_lookup_byid(ctx->net,
-							tb[NFTA_LOOKUP_SET_ID],
-							genmask);
-		}
-		if (IS_ERR(set))
-			return PTR_ERR(set);
-	}
+	set = nft_set_lookup(ctx->net, ctx->table, tb[NFTA_LOOKUP_SET],
+			     tb[NFTA_LOOKUP_SET_ID], genmask);
+	if (IS_ERR(set))
+		return PTR_ERR(set);
 
 	if (set->flags & NFT_SET_EVAL)
 		return -EOPNOTSUPP;
diff --git a/net/netfilter/nft_objref.c b/net/netfilter/nft_objref.c
index 1ae8c49ca4a1..1dd428fbaaa3 100644
--- a/net/netfilter/nft_objref.c
+++ b/net/netfilter/nft_objref.c
@@ -116,16 +116,10 @@ static int nft_objref_map_init(const struct nft_ctx *ctx,
 	struct nft_set *set;
 	int err;
 
-	set = nf_tables_set_lookup(ctx->table, tb[NFTA_OBJREF_SET_NAME], genmask);
-	if (IS_ERR(set)) {
-		if (tb[NFTA_OBJREF_SET_ID]) {
-			set = nf_tables_set_lookup_byid(ctx->net,
-							tb[NFTA_OBJREF_SET_ID],
-							genmask);
-		}
-		if (IS_ERR(set))
-			return PTR_ERR(set);
-	}
+	set = nft_set_lookup(ctx->net, ctx->table, tb[NFTA_OBJREF_SET_NAME],
+			     tb[NFTA_OBJREF_SET_ID], genmask);
+	if (IS_ERR(set))
+		return PTR_ERR(set);
 
 	if (!(set->flags & NFT_SET_OBJECT))
 		return -EINVAL;

From efc9b8e33b8b5ef890288758454ce62a1319c94a Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Tue, 7 Mar 2017 12:45:04 +0100
Subject: [PATCH 08/22] netfilter: bridge: remove unneeded rcu_read_lock

as comment says, the function is always called with rcu read lock held.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/bridge/br_netfilter_hooks.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/net/bridge/br_netfilter_hooks.c b/net/bridge/br_netfilter_hooks.c
index 95087e6e8258..52739e6c610e 100644
--- a/net/bridge/br_netfilter_hooks.c
+++ b/net/bridge/br_netfilter_hooks.c
@@ -1016,13 +1016,10 @@ int br_nf_hook_thresh(unsigned int hook, struct net *net,
 	if (!elem)
 		return okfn(net, sk, skb);
 
-	/* We may already have this, but read-locks nest anyway */
-	rcu_read_lock();
 	nf_hook_state_init(&state, hook, NFPROTO_BRIDGE, indev, outdev,
 			   sk, net, okfn);
 
 	ret = nf_hook_slow(skb, &state, elem);
-	rcu_read_unlock();
 	if (ret == 1)
 		ret = okfn(net, sk, skb);
 

From c1183db8853d14df390748aa6baa69734db151aa Mon Sep 17 00:00:00 2001
From: Taehee Yoo <ap420073@gmail.com>
Date: Tue, 7 Mar 2017 22:02:50 +0900
Subject: [PATCH 09/22] netfilter: nf_reject: remove unused variable

variable oiph is not used.

Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/ipv4/netfilter/nf_reject_ipv4.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/net/ipv4/netfilter/nf_reject_ipv4.c b/net/ipv4/netfilter/nf_reject_ipv4.c
index 146d86105183..7cd8d0d918f8 100644
--- a/net/ipv4/netfilter/nf_reject_ipv4.c
+++ b/net/ipv4/netfilter/nf_reject_ipv4.c
@@ -104,7 +104,6 @@ EXPORT_SYMBOL_GPL(nf_reject_ip_tcphdr_put);
 void nf_send_reset(struct net *net, struct sk_buff *oldskb, int hook)
 {
 	struct sk_buff *nskb;
-	const struct iphdr *oiph;
 	struct iphdr *niph;
 	const struct tcphdr *oth;
 	struct tcphdr _oth;
@@ -116,8 +115,6 @@ void nf_send_reset(struct net *net, struct sk_buff *oldskb, int hook)
 	if (skb_rtable(oldskb)->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST))
 		return;
 
-	oiph = ip_hdr(oldskb);
-
 	nskb = alloc_skb(sizeof(struct iphdr) + sizeof(struct tcphdr) +
 			 LL_MAX_HEADER, GFP_ATOMIC);
 	if (!nskb)

From 84fba05511c8e0e1a6e6c8ec2c3c85616c39a8c1 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 8 Mar 2017 16:48:43 +0100
Subject: [PATCH 10/22] netfilter: provide nft_ctx in object init function

this is needed by the upcoming ct helper object type --
we'd like to be able use the table family (ip, ip6, inet) to figure
out which helper has to be requested.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h | 3 ++-
 net/netfilter/nf_tables_api.c     | 7 ++++---
 net/netfilter/nft_counter.c       | 3 ++-
 net/netfilter/nft_quota.c         | 3 ++-
 4 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index f0d46726d06e..49436849d7d7 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1017,7 +1017,8 @@ struct nft_object_type {
 	unsigned int			maxattr;
 	struct module			*owner;
 	const struct nla_policy		*policy;
-	int				(*init)(const struct nlattr * const tb[],
+	int				(*init)(const struct nft_ctx *ctx,
+						const struct nlattr *const tb[],
 						struct nft_object *obj);
 	void				(*destroy)(struct nft_object *obj);
 	int				(*dump)(struct sk_buff *skb,
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 4559f5d66bcc..12cc5218de96 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -4095,7 +4095,8 @@ static const struct nla_policy nft_obj_policy[NFTA_OBJ_MAX + 1] = {
 	[NFTA_OBJ_DATA]		= { .type = NLA_NESTED },
 };
 
-static struct nft_object *nft_obj_init(const struct nft_object_type *type,
+static struct nft_object *nft_obj_init(const struct nft_ctx *ctx,
+				       const struct nft_object_type *type,
 				       const struct nlattr *attr)
 {
 	struct nlattr *tb[type->maxattr + 1];
@@ -4115,7 +4116,7 @@ static struct nft_object *nft_obj_init(const struct nft_object_type *type,
 	if (obj == NULL)
 		goto err1;
 
-	err = type->init((const struct nlattr * const *)tb, obj);
+	err = type->init(ctx, (const struct nlattr * const *)tb, obj);
 	if (err < 0)
 		goto err2;
 
@@ -4223,7 +4224,7 @@ static int nf_tables_newobj(struct net *net, struct sock *nlsk,
 	if (IS_ERR(type))
 		return PTR_ERR(type);
 
-	obj = nft_obj_init(type, nla[NFTA_OBJ_DATA]);
+	obj = nft_obj_init(&ctx, type, nla[NFTA_OBJ_DATA]);
 	if (IS_ERR(obj)) {
 		err = PTR_ERR(obj);
 		goto err1;
diff --git a/net/netfilter/nft_counter.c b/net/netfilter/nft_counter.c
index 7f8422213341..67a710ebde09 100644
--- a/net/netfilter/nft_counter.c
+++ b/net/netfilter/nft_counter.c
@@ -82,7 +82,8 @@ static int nft_counter_do_init(const struct nlattr * const tb[],
 	return 0;
 }
 
-static int nft_counter_obj_init(const struct nlattr * const tb[],
+static int nft_counter_obj_init(const struct nft_ctx *ctx,
+				const struct nlattr * const tb[],
 				struct nft_object *obj)
 {
 	struct nft_counter_percpu_priv *priv = nft_obj_data(obj);
diff --git a/net/netfilter/nft_quota.c b/net/netfilter/nft_quota.c
index 2d6fe3559912..25e33159be57 100644
--- a/net/netfilter/nft_quota.c
+++ b/net/netfilter/nft_quota.c
@@ -99,7 +99,8 @@ static int nft_quota_do_init(const struct nlattr * const tb[],
 	return 0;
 }
 
-static int nft_quota_obj_init(const struct nlattr * const tb[],
+static int nft_quota_obj_init(const struct nft_ctx *ctx,
+			      const struct nlattr * const tb[],
 			      struct nft_object *obj)
 {
 	struct nft_quota *priv = nft_obj_data(obj);

From 1a64edf54f55d7956cf5a0d95898bc1f84f9b818 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 8 Mar 2017 16:48:44 +0100
Subject: [PATCH 11/22] netfilter: nft_ct: add helper set support

this allows to assign connection tracking helpers to
connections via nft objref infrastructure.

The idea is to first specifiy a helper object:

 table ip filter {
    ct helper some-name {
      type "ftp"
      protocol tcp
      l3proto ip
    }
 }

and then assign it via

nft add ... ct helper set "some-name"

helper assignment works for new conntracks only as we cannot expand the
conntrack extension area once it has been committed to the main conntrack
table.

ipv4 and ipv6 protocols are tracked stored separately so
we can also handle families that observe both ipv4 and ipv6 traffic.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/uapi/linux/netfilter/nf_tables.h |  12 +-
 net/netfilter/nft_ct.c                   | 171 +++++++++++++++++++++++
 2 files changed, 182 insertions(+), 1 deletion(-)

diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index 4f7d75682c59..34c8d08b687a 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -1259,10 +1259,20 @@ enum nft_fib_flags {
 	NFTA_FIB_F_OIF		= 1 << 4,	/* restrict to oif */
 };
 
+enum nft_ct_helper_attributes {
+	NFTA_CT_HELPER_UNSPEC,
+	NFTA_CT_HELPER_NAME,
+	NFTA_CT_HELPER_L3PROTO,
+	NFTA_CT_HELPER_L4PROTO,
+	__NFTA_CT_HELPER_MAX,
+};
+#define NFTA_CT_HELPER_MAX	(__NFTA_CT_HELPER_MAX - 1)
+
 #define NFT_OBJECT_UNSPEC	0
 #define NFT_OBJECT_COUNTER	1
 #define NFT_OBJECT_QUOTA	2
-#define __NFT_OBJECT_MAX	3
+#define NFT_OBJECT_CT_HELPER	3
+#define __NFT_OBJECT_MAX	4
 #define NFT_OBJECT_MAX		(__NFT_OBJECT_MAX - 1)
 
 /**
diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c
index bf548a7a71ec..4144ae845bdd 100644
--- a/net/netfilter/nft_ct.c
+++ b/net/netfilter/nft_ct.c
@@ -32,6 +32,12 @@ struct nft_ct {
 	};
 };
 
+struct nft_ct_helper_obj  {
+	struct nf_conntrack_helper *helper4;
+	struct nf_conntrack_helper *helper6;
+	u8 l4proto;
+};
+
 #ifdef CONFIG_NF_CONNTRACK_ZONES
 static DEFINE_PER_CPU(struct nf_conn *, nft_ct_pcpu_template);
 static unsigned int nft_ct_pcpu_template_refcnt __read_mostly;
@@ -730,6 +736,162 @@ static struct nft_expr_type nft_notrack_type __read_mostly = {
 	.owner		= THIS_MODULE,
 };
 
+static int nft_ct_helper_obj_init(const struct nft_ctx *ctx,
+				  const struct nlattr * const tb[],
+				  struct nft_object *obj)
+{
+	struct nft_ct_helper_obj *priv = nft_obj_data(obj);
+	struct nf_conntrack_helper *help4, *help6;
+	char name[NF_CT_HELPER_NAME_LEN];
+	int family = ctx->afi->family;
+
+	if (!tb[NFTA_CT_HELPER_NAME] || !tb[NFTA_CT_HELPER_L4PROTO])
+		return -EINVAL;
+
+	priv->l4proto = nla_get_u8(tb[NFTA_CT_HELPER_L4PROTO]);
+	if (!priv->l4proto)
+		return -ENOENT;
+
+	nla_strlcpy(name, tb[NFTA_CT_HELPER_NAME], sizeof(name));
+
+	if (tb[NFTA_CT_HELPER_L3PROTO])
+		family = ntohs(nla_get_be16(tb[NFTA_CT_HELPER_L3PROTO]));
+
+	help4 = NULL;
+	help6 = NULL;
+
+	switch (family) {
+	case NFPROTO_IPV4:
+		if (ctx->afi->family == NFPROTO_IPV6)
+			return -EINVAL;
+
+		help4 = nf_conntrack_helper_try_module_get(name, family,
+							   priv->l4proto);
+		break;
+	case NFPROTO_IPV6:
+		if (ctx->afi->family == NFPROTO_IPV4)
+			return -EINVAL;
+
+		help6 = nf_conntrack_helper_try_module_get(name, family,
+							   priv->l4proto);
+		break;
+	case NFPROTO_NETDEV: /* fallthrough */
+	case NFPROTO_BRIDGE: /* same */
+	case NFPROTO_INET:
+		help4 = nf_conntrack_helper_try_module_get(name, NFPROTO_IPV4,
+							   priv->l4proto);
+		help6 = nf_conntrack_helper_try_module_get(name, NFPROTO_IPV6,
+							   priv->l4proto);
+		break;
+	default:
+		return -EAFNOSUPPORT;
+	}
+
+	/* && is intentional; only error if INET found neither ipv4 or ipv6 */
+	if (!help4 && !help6)
+		return -ENOENT;
+
+	priv->helper4 = help4;
+	priv->helper6 = help6;
+
+	return 0;
+}
+
+static void nft_ct_helper_obj_destroy(struct nft_object *obj)
+{
+	struct nft_ct_helper_obj *priv = nft_obj_data(obj);
+
+	if (priv->helper4)
+		module_put(priv->helper4->me);
+	if (priv->helper6)
+		module_put(priv->helper6->me);
+}
+
+static void nft_ct_helper_obj_eval(struct nft_object *obj,
+				   struct nft_regs *regs,
+				   const struct nft_pktinfo *pkt)
+{
+	const struct nft_ct_helper_obj *priv = nft_obj_data(obj);
+	struct nf_conn *ct = (struct nf_conn *)skb_nfct(pkt->skb);
+	struct nf_conntrack_helper *to_assign = NULL;
+	struct nf_conn_help *help;
+
+	if (!ct ||
+	    nf_ct_is_confirmed(ct) ||
+	    nf_ct_is_template(ct) ||
+	    priv->l4proto != nf_ct_protonum(ct))
+		return;
+
+	switch (nf_ct_l3num(ct)) {
+	case NFPROTO_IPV4:
+		to_assign = priv->helper4;
+		break;
+	case NFPROTO_IPV6:
+		to_assign = priv->helper6;
+		break;
+	default:
+		WARN_ON_ONCE(1);
+		return;
+	}
+
+	if (!to_assign)
+		return;
+
+	if (test_bit(IPS_HELPER_BIT, &ct->status))
+		return;
+
+	help = nf_ct_helper_ext_add(ct, to_assign, GFP_ATOMIC);
+	if (help) {
+		rcu_assign_pointer(help->helper, to_assign);
+		set_bit(IPS_HELPER_BIT, &ct->status);
+	}
+}
+
+static int nft_ct_helper_obj_dump(struct sk_buff *skb,
+				  struct nft_object *obj, bool reset)
+{
+	const struct nft_ct_helper_obj *priv = nft_obj_data(obj);
+	const struct nf_conntrack_helper *helper = priv->helper4;
+	u16 family;
+
+	if (nla_put_string(skb, NFTA_CT_HELPER_NAME, helper->name))
+		return -1;
+
+	if (nla_put_u8(skb, NFTA_CT_HELPER_L4PROTO, priv->l4proto))
+		return -1;
+
+	if (priv->helper4 && priv->helper6)
+		family = NFPROTO_INET;
+	else if (priv->helper6)
+		family = NFPROTO_IPV6;
+	else
+		family = NFPROTO_IPV4;
+
+	if (nla_put_be16(skb, NFTA_CT_HELPER_L3PROTO, htons(family)))
+		return -1;
+
+	return 0;
+}
+
+static const struct nla_policy nft_ct_helper_policy[NFTA_CT_HELPER_MAX + 1] = {
+	[NFTA_CT_HELPER_NAME] = { .type = NLA_STRING,
+				  .len = NF_CT_HELPER_NAME_LEN - 1 },
+	[NFTA_CT_HELPER_L3PROTO] = { .type = NLA_U16 },
+	[NFTA_CT_HELPER_L4PROTO] = { .type = NLA_U8 },
+};
+
+static struct nft_object_type nft_ct_helper_obj __read_mostly = {
+	.type		= NFT_OBJECT_CT_HELPER,
+	.size		= sizeof(struct nft_ct_helper_obj),
+	.maxattr	= NFTA_CT_HELPER_MAX,
+	.policy		= nft_ct_helper_policy,
+	.eval		= nft_ct_helper_obj_eval,
+	.init		= nft_ct_helper_obj_init,
+	.destroy	= nft_ct_helper_obj_destroy,
+	.dump		= nft_ct_helper_obj_dump,
+	.owner		= THIS_MODULE,
+};
+
 static int __init nft_ct_module_init(void)
 {
 	int err;
@@ -744,7 +906,14 @@ static int __init nft_ct_module_init(void)
 	if (err < 0)
 		goto err1;
 
+	err = nft_register_obj(&nft_ct_helper_obj);
+	if (err < 0)
+		goto err2;
+
 	return 0;
+
+err2:
+	nft_unregister_expr(&nft_notrack_type);
 err1:
 	nft_unregister_expr(&nft_ct_type);
 	return err;
@@ -752,6 +921,7 @@ err1:
 
 static void __exit nft_ct_module_exit(void)
 {
+	nft_unregister_obj(&nft_ct_helper_obj);
 	nft_unregister_expr(&nft_notrack_type);
 	nft_unregister_expr(&nft_ct_type);
 }
@@ -763,3 +933,4 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
 MODULE_ALIAS_NFT_EXPR("ct");
 MODULE_ALIAS_NFT_EXPR("notrack");
+MODULE_ALIAS_NFT_OBJ(NFT_OBJECT_CT_HELPER);

From 055c4b34b94f696d9bd9aad53a11378a0fc409c9 Mon Sep 17 00:00:00 2001
From: Phil Sutter <phil@nwl.cc>
Date: Fri, 10 Mar 2017 18:08:02 +0100
Subject: [PATCH 12/22] netfilter: nft_fib: Support existence check

Instead of the actual interface index or name, set destination register
to just 1 or 0 depending on whether the lookup succeeded or not if
NFTA_FIB_F_PRESENT was set in userspace.

Signed-off-by: Phil Sutter <phil@nwl.cc>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nft_fib.h          |  2 +-
 include/uapi/linux/netfilter/nf_tables.h |  1 +
 net/ipv4/netfilter/nft_fib_ipv4.c        |  4 ++--
 net/ipv6/netfilter/nft_fib_ipv6.c        |  2 +-
 net/netfilter/nft_fib.c                  | 14 +++++++++-----
 5 files changed, 14 insertions(+), 9 deletions(-)

diff --git a/include/net/netfilter/nft_fib.h b/include/net/netfilter/nft_fib.h
index 5ceb2205e4e3..381af9469e6a 100644
--- a/include/net/netfilter/nft_fib.h
+++ b/include/net/netfilter/nft_fib.h
@@ -32,6 +32,6 @@ void nft_fib6_eval_type(const struct nft_expr *expr, struct nft_regs *regs,
 void nft_fib6_eval(const struct nft_expr *expr, struct nft_regs *regs,
 		   const struct nft_pktinfo *pkt);
 
-void nft_fib_store_result(void *reg, enum nft_fib_result r,
+void nft_fib_store_result(void *reg, const struct nft_fib *priv,
 			  const struct nft_pktinfo *pkt, int index);
 #endif
diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index 34c8d08b687a..8f3842690d17 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -1257,6 +1257,7 @@ enum nft_fib_flags {
 	NFTA_FIB_F_MARK		= 1 << 2,	/* use skb->mark */
 	NFTA_FIB_F_IIF		= 1 << 3,	/* restrict to iif */
 	NFTA_FIB_F_OIF		= 1 << 4,	/* restrict to oif */
+	NFTA_FIB_F_PRESENT	= 1 << 5,	/* check existence only */
 };
 
 enum nft_ct_helper_attributes {
diff --git a/net/ipv4/netfilter/nft_fib_ipv4.c b/net/ipv4/netfilter/nft_fib_ipv4.c
index 2981291910dd..f4e4462cb5bb 100644
--- a/net/ipv4/netfilter/nft_fib_ipv4.c
+++ b/net/ipv4/netfilter/nft_fib_ipv4.c
@@ -90,7 +90,7 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs,
 
 	if (nft_hook(pkt) == NF_INET_PRE_ROUTING &&
 	    nft_fib_is_loopback(pkt->skb, nft_in(pkt))) {
-		nft_fib_store_result(dest, priv->result, pkt,
+		nft_fib_store_result(dest, priv, pkt,
 				     nft_in(pkt)->ifindex);
 		return;
 	}
@@ -99,7 +99,7 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs,
 	if (ipv4_is_zeronet(iph->saddr)) {
 		if (ipv4_is_lbcast(iph->daddr) ||
 		    ipv4_is_local_multicast(iph->daddr)) {
-			nft_fib_store_result(dest, priv->result, pkt,
+			nft_fib_store_result(dest, priv, pkt,
 					     get_ifindex(pkt->skb->dev));
 			return;
 		}
diff --git a/net/ipv6/netfilter/nft_fib_ipv6.c b/net/ipv6/netfilter/nft_fib_ipv6.c
index 765facf03d45..e8d88d82636b 100644
--- a/net/ipv6/netfilter/nft_fib_ipv6.c
+++ b/net/ipv6/netfilter/nft_fib_ipv6.c
@@ -159,7 +159,7 @@ void nft_fib6_eval(const struct nft_expr *expr, struct nft_regs *regs,
 
 	if (nft_hook(pkt) == NF_INET_PRE_ROUTING &&
 	    nft_fib_is_loopback(pkt->skb, nft_in(pkt))) {
-		nft_fib_store_result(dest, priv->result, pkt,
+		nft_fib_store_result(dest, priv, pkt,
 				     nft_in(pkt)->ifindex);
 		return;
 	}
diff --git a/net/netfilter/nft_fib.c b/net/netfilter/nft_fib.c
index fd0b19303b0d..21df8cccea65 100644
--- a/net/netfilter/nft_fib.c
+++ b/net/netfilter/nft_fib.c
@@ -24,7 +24,8 @@ const struct nla_policy nft_fib_policy[NFTA_FIB_MAX + 1] = {
 EXPORT_SYMBOL(nft_fib_policy);
 
 #define NFTA_FIB_F_ALL (NFTA_FIB_F_SADDR | NFTA_FIB_F_DADDR | \
-			NFTA_FIB_F_MARK | NFTA_FIB_F_IIF | NFTA_FIB_F_OIF)
+			NFTA_FIB_F_MARK | NFTA_FIB_F_IIF | NFTA_FIB_F_OIF | \
+			NFTA_FIB_F_PRESENT)
 
 int nft_fib_validate(const struct nft_ctx *ctx, const struct nft_expr *expr,
 		     const struct nft_data **data)
@@ -133,19 +134,22 @@ int nft_fib_dump(struct sk_buff *skb, const struct nft_expr *expr)
 }
 EXPORT_SYMBOL_GPL(nft_fib_dump);
 
-void nft_fib_store_result(void *reg, enum nft_fib_result r,
+void nft_fib_store_result(void *reg, const struct nft_fib *priv,
 			  const struct nft_pktinfo *pkt, int index)
 {
 	struct net_device *dev;
 	u32 *dreg = reg;
 
-	switch (r) {
+	switch (priv->result) {
 	case NFT_FIB_RESULT_OIF:
-		*dreg = index;
+		*dreg = (priv->flags & NFTA_FIB_F_PRESENT) ? !!index : index;
 		break;
 	case NFT_FIB_RESULT_OIFNAME:
 		dev = dev_get_by_index_rcu(nft_net(pkt), index);
-		strncpy(reg, dev ? dev->name : "", IFNAMSIZ);
+		if (priv->flags & NFTA_FIB_F_PRESENT)
+			*dreg = !!dev;
+		else
+			strncpy(reg, dev ? dev->name : "", IFNAMSIZ);
 		break;
 	default:
 		WARN_ON_ONCE(1);

From fc09e4a75a12552b9374a5b0c5f5815aecf8c37a Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 9 Mar 2017 12:57:15 +0100
Subject: [PATCH 13/22] netfilter: nf_conntrack: reduce resolve_normal_ct args

also mark init_conntrack noinline, in most cases resolve_normal_ct will
find an existing conntrack entry.

text    data     bss     dec     hex filename
16735    5707     176   22618    585a net/netfilter/nf_conntrack_core.o
16687    5707     176   22570    582a net/netfilter/nf_conntrack_core.o

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_conntrack_core.c | 57 ++++++++++++++-----------------
 1 file changed, 26 insertions(+), 31 deletions(-)

diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 071b97fcbefb..b0f2e8e65084 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -1129,7 +1129,7 @@ EXPORT_SYMBOL_GPL(nf_conntrack_free);
 
 /* Allocate a new conntrack: we return -ENOMEM if classification
    failed due to stress.  Otherwise it really is unclassifiable. */
-static struct nf_conntrack_tuple_hash *
+static noinline struct nf_conntrack_tuple_hash *
 init_conntrack(struct net *net, struct nf_conn *tmpl,
 	       const struct nf_conntrack_tuple *tuple,
 	       struct nf_conntrack_l3proto *l3proto,
@@ -1237,21 +1237,20 @@ init_conntrack(struct net *net, struct nf_conn *tmpl,
 	return &ct->tuplehash[IP_CT_DIR_ORIGINAL];
 }
 
-/* On success, returns conntrack ptr, sets skb->_nfct | ctinfo */
-static inline struct nf_conn *
+/* On success, returns 0, sets skb->_nfct | ctinfo */
+static int
 resolve_normal_ct(struct net *net, struct nf_conn *tmpl,
 		  struct sk_buff *skb,
 		  unsigned int dataoff,
 		  u_int16_t l3num,
 		  u_int8_t protonum,
 		  struct nf_conntrack_l3proto *l3proto,
-		  struct nf_conntrack_l4proto *l4proto,
-		  int *set_reply,
-		  enum ip_conntrack_info *ctinfo)
+		  struct nf_conntrack_l4proto *l4proto)
 {
 	const struct nf_conntrack_zone *zone;
 	struct nf_conntrack_tuple tuple;
 	struct nf_conntrack_tuple_hash *h;
+	enum ip_conntrack_info ctinfo;
 	struct nf_conntrack_zone tmp;
 	struct nf_conn *ct;
 	u32 hash;
@@ -1260,7 +1259,7 @@ resolve_normal_ct(struct net *net, struct nf_conn *tmpl,
 			     dataoff, l3num, protonum, net, &tuple, l3proto,
 			     l4proto)) {
 		pr_debug("Can't get tuple\n");
-		return NULL;
+		return 0;
 	}
 
 	/* look for tuple match */
@@ -1271,33 +1270,30 @@ resolve_normal_ct(struct net *net, struct nf_conn *tmpl,
 		h = init_conntrack(net, tmpl, &tuple, l3proto, l4proto,
 				   skb, dataoff, hash);
 		if (!h)
-			return NULL;
+			return 0;
 		if (IS_ERR(h))
-			return (void *)h;
+			return PTR_ERR(h);
 	}
 	ct = nf_ct_tuplehash_to_ctrack(h);
 
 	/* It exists; we have (non-exclusive) reference. */
 	if (NF_CT_DIRECTION(h) == IP_CT_DIR_REPLY) {
-		*ctinfo = IP_CT_ESTABLISHED_REPLY;
-		/* Please set reply bit if this packet OK */
-		*set_reply = 1;
+		ctinfo = IP_CT_ESTABLISHED_REPLY;
 	} else {
 		/* Once we've had two way comms, always ESTABLISHED. */
 		if (test_bit(IPS_SEEN_REPLY_BIT, &ct->status)) {
 			pr_debug("normal packet for %p\n", ct);
-			*ctinfo = IP_CT_ESTABLISHED;
+			ctinfo = IP_CT_ESTABLISHED;
 		} else if (test_bit(IPS_EXPECTED_BIT, &ct->status)) {
 			pr_debug("related packet for %p\n", ct);
-			*ctinfo = IP_CT_RELATED;
+			ctinfo = IP_CT_RELATED;
 		} else {
 			pr_debug("new packet for %p\n", ct);
-			*ctinfo = IP_CT_NEW;
+			ctinfo = IP_CT_NEW;
 		}
-		*set_reply = 0;
 	}
-	nf_ct_set(skb, ct, *ctinfo);
-	return ct;
+	nf_ct_set(skb, ct, ctinfo);
+	return 0;
 }
 
 unsigned int
@@ -1311,7 +1307,6 @@ nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum,
 	unsigned int *timeouts;
 	unsigned int dataoff;
 	u_int8_t protonum;
-	int set_reply = 0;
 	int ret;
 
 	tmpl = nf_ct_get(skb, &ctinfo);
@@ -1354,23 +1349,22 @@ nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum,
 			goto out;
 	}
 repeat:
-	ct = resolve_normal_ct(net, tmpl, skb, dataoff, pf, protonum,
-			       l3proto, l4proto, &set_reply, &ctinfo);
-	if (!ct) {
-		/* Not valid part of a connection */
-		NF_CT_STAT_INC_ATOMIC(net, invalid);
-		ret = NF_ACCEPT;
-		goto out;
-	}
-
-	if (IS_ERR(ct)) {
+	ret = resolve_normal_ct(net, tmpl, skb, dataoff, pf, protonum,
+				l3proto, l4proto);
+	if (ret < 0) {
 		/* Too stressed to deal. */
 		NF_CT_STAT_INC_ATOMIC(net, drop);
 		ret = NF_DROP;
 		goto out;
 	}
 
-	NF_CT_ASSERT(skb_nfct(skb));
+	ct = nf_ct_get(skb, &ctinfo);
+	if (!ct) {
+		/* Not valid part of a connection */
+		NF_CT_STAT_INC_ATOMIC(net, invalid);
+		ret = NF_ACCEPT;
+		goto out;
+	}
 
 	/* Decide what timeout policy we want to apply to this flow. */
 	timeouts = nf_ct_timeout_lookup(net, ct, l4proto);
@@ -1395,7 +1389,8 @@ repeat:
 		goto out;
 	}
 
-	if (set_reply && !test_and_set_bit(IPS_SEEN_REPLY_BIT, &ct->status))
+	if (ctinfo == IP_CT_ESTABLISHED_REPLY &&
+	    !test_and_set_bit(IPS_SEEN_REPLY_BIT, &ct->status))
 		nf_conntrack_event_cache(IPCT_REPLY, ct);
 out:
 	if (tmpl)

From 2cb4bbd75bdf9d423b9f6c629f81eb66ee312fac Mon Sep 17 00:00:00 2001
From: Liping Zhang <zlpnobody@gmail.com>
Date: Sat, 11 Mar 2017 14:08:09 +0800
Subject: [PATCH 14/22] netfilter: limit: use per-rule spinlock to improve the
 scalability

The limit token is independent between each rules, so there's no
need to use a global spinlock.

Signed-off-by: Liping Zhang <zlpnobody@gmail.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_limit.c | 10 +++++-----
 net/netfilter/xt_limit.c  | 11 ++++++-----
 2 files changed, 11 insertions(+), 10 deletions(-)

diff --git a/net/netfilter/nft_limit.c b/net/netfilter/nft_limit.c
index c6baf412236d..18dd57a52651 100644
--- a/net/netfilter/nft_limit.c
+++ b/net/netfilter/nft_limit.c
@@ -17,9 +17,8 @@
 #include <linux/netfilter/nf_tables.h>
 #include <net/netfilter/nf_tables.h>
 
-static DEFINE_SPINLOCK(limit_lock);
-
 struct nft_limit {
+	spinlock_t	lock;
 	u64		last;
 	u64		tokens;
 	u64		tokens_max;
@@ -34,7 +33,7 @@ static inline bool nft_limit_eval(struct nft_limit *limit, u64 cost)
 	u64 now, tokens;
 	s64 delta;
 
-	spin_lock_bh(&limit_lock);
+	spin_lock_bh(&limit->lock);
 	now = ktime_get_ns();
 	tokens = limit->tokens + now - limit->last;
 	if (tokens > limit->tokens_max)
@@ -44,11 +43,11 @@ static inline bool nft_limit_eval(struct nft_limit *limit, u64 cost)
 	delta = tokens - cost;
 	if (delta >= 0) {
 		limit->tokens = delta;
-		spin_unlock_bh(&limit_lock);
+		spin_unlock_bh(&limit->lock);
 		return limit->invert;
 	}
 	limit->tokens = tokens;
-	spin_unlock_bh(&limit_lock);
+	spin_unlock_bh(&limit->lock);
 	return !limit->invert;
 }
 
@@ -86,6 +85,7 @@ static int nft_limit_init(struct nft_limit *limit,
 			limit->invert = true;
 	}
 	limit->last = ktime_get_ns();
+	spin_lock_init(&limit->lock);
 
 	return 0;
 }
diff --git a/net/netfilter/xt_limit.c b/net/netfilter/xt_limit.c
index dab962df1787..d27b5f1ea619 100644
--- a/net/netfilter/xt_limit.c
+++ b/net/netfilter/xt_limit.c
@@ -18,6 +18,7 @@
 #include <linux/netfilter/xt_limit.h>
 
 struct xt_limit_priv {
+	spinlock_t lock;
 	unsigned long prev;
 	uint32_t credit;
 };
@@ -32,8 +33,6 @@ MODULE_ALIAS("ip6t_limit");
  * see net/sched/sch_tbf.c in the linux source tree
  */
 
-static DEFINE_SPINLOCK(limit_lock);
-
 /* Rusty: This is my (non-mathematically-inclined) understanding of
    this algorithm.  The `average rate' in jiffies becomes your initial
    amount of credit `credit' and the most credit you can ever have
@@ -72,7 +71,7 @@ limit_mt(const struct sk_buff *skb, struct xt_action_param *par)
 	struct xt_limit_priv *priv = r->master;
 	unsigned long now = jiffies;
 
-	spin_lock_bh(&limit_lock);
+	spin_lock_bh(&priv->lock);
 	priv->credit += (now - xchg(&priv->prev, now)) * CREDITS_PER_JIFFY;
 	if (priv->credit > r->credit_cap)
 		priv->credit = r->credit_cap;
@@ -80,11 +79,11 @@ limit_mt(const struct sk_buff *skb, struct xt_action_param *par)
 	if (priv->credit >= r->cost) {
 		/* We're not limited. */
 		priv->credit -= r->cost;
-		spin_unlock_bh(&limit_lock);
+		spin_unlock_bh(&priv->lock);
 		return true;
 	}
 
-	spin_unlock_bh(&limit_lock);
+	spin_unlock_bh(&priv->lock);
 	return false;
 }
 
@@ -126,6 +125,8 @@ static int limit_mt_check(const struct xt_mtchk_param *par)
 		r->credit_cap = priv->credit; /* Credits full. */
 		r->cost = user2credits(r->avg);
 	}
+	spin_lock_init(&priv->lock);
+
 	return 0;
 }
 

From 03e5fd0e9bcc1f34b7a542786b34b8f771e7c260 Mon Sep 17 00:00:00 2001
From: Liping Zhang <zlpnobody@gmail.com>
Date: Sun, 12 Mar 2017 19:38:47 +0800
Subject: [PATCH 15/22] netfilter: nft_set_rbtree: use per-set rwlock to
 improve the scalability

Karel Rericha reported that in his test case, ICMP packets going through
boxes had normally about 5ms latency. But when running nft, actually
listing the sets with interval flags, latency would go up to 30-100ms.
This was observed when router throughput is from 600Mbps to 2Gbps.

This is because we use a single global spinlock to protect the whole
rbtree sets, so "dumping sets" will race with the "key lookup" inevitably.
But actually they are all _readers_, so it's ok to convert the spinlock
to rwlock to avoid competition between them. Also use per-set rwlock since
each set is independent.

Reported-by: Karel Rericha <karel@unitednetworks.cz>
Tested-by: Karel Rericha <karel@unitednetworks.cz>
Signed-off-by: Liping Zhang <zlpnobody@gmail.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nft_set_rbtree.c | 31 ++++++++++++++++---------------
 1 file changed, 16 insertions(+), 15 deletions(-)

diff --git a/net/netfilter/nft_set_rbtree.c b/net/netfilter/nft_set_rbtree.c
index 78dfbf9588b3..e97e2fb53f0a 100644
--- a/net/netfilter/nft_set_rbtree.c
+++ b/net/netfilter/nft_set_rbtree.c
@@ -18,9 +18,8 @@
 #include <linux/netfilter/nf_tables.h>
 #include <net/netfilter/nf_tables.h>
 
-static DEFINE_SPINLOCK(nft_rbtree_lock);
-
 struct nft_rbtree {
+	rwlock_t		lock;
 	struct rb_root		root;
 };
 
@@ -44,14 +43,14 @@ static bool nft_rbtree_equal(const struct nft_set *set, const void *this,
 static bool nft_rbtree_lookup(const struct net *net, const struct nft_set *set,
 			      const u32 *key, const struct nft_set_ext **ext)
 {
-	const struct nft_rbtree *priv = nft_set_priv(set);
+	struct nft_rbtree *priv = nft_set_priv(set);
 	const struct nft_rbtree_elem *rbe, *interval = NULL;
 	u8 genmask = nft_genmask_cur(net);
 	const struct rb_node *parent;
 	const void *this;
 	int d;
 
-	spin_lock_bh(&nft_rbtree_lock);
+	read_lock_bh(&priv->lock);
 	parent = priv->root.rb_node;
 	while (parent != NULL) {
 		rbe = rb_entry(parent, struct nft_rbtree_elem, node);
@@ -75,7 +74,7 @@ static bool nft_rbtree_lookup(const struct net *net, const struct nft_set *set,
 			}
 			if (nft_rbtree_interval_end(rbe))
 				goto out;
-			spin_unlock_bh(&nft_rbtree_lock);
+			read_unlock_bh(&priv->lock);
 
 			*ext = &rbe->ext;
 			return true;
@@ -85,12 +84,12 @@ static bool nft_rbtree_lookup(const struct net *net, const struct nft_set *set,
 	if (set->flags & NFT_SET_INTERVAL && interval != NULL &&
 	    nft_set_elem_active(&interval->ext, genmask) &&
 	    !nft_rbtree_interval_end(interval)) {
-		spin_unlock_bh(&nft_rbtree_lock);
+		read_unlock_bh(&priv->lock);
 		*ext = &interval->ext;
 		return true;
 	}
 out:
-	spin_unlock_bh(&nft_rbtree_lock);
+	read_unlock_bh(&priv->lock);
 	return false;
 }
 
@@ -140,12 +139,13 @@ static int nft_rbtree_insert(const struct net *net, const struct nft_set *set,
 			     const struct nft_set_elem *elem,
 			     struct nft_set_ext **ext)
 {
+	struct nft_rbtree *priv = nft_set_priv(set);
 	struct nft_rbtree_elem *rbe = elem->priv;
 	int err;
 
-	spin_lock_bh(&nft_rbtree_lock);
+	write_lock_bh(&priv->lock);
 	err = __nft_rbtree_insert(net, set, rbe, ext);
-	spin_unlock_bh(&nft_rbtree_lock);
+	write_unlock_bh(&priv->lock);
 
 	return err;
 }
@@ -157,9 +157,9 @@ static void nft_rbtree_remove(const struct net *net,
 	struct nft_rbtree *priv = nft_set_priv(set);
 	struct nft_rbtree_elem *rbe = elem->priv;
 
-	spin_lock_bh(&nft_rbtree_lock);
+	write_lock_bh(&priv->lock);
 	rb_erase(&rbe->node, &priv->root);
-	spin_unlock_bh(&nft_rbtree_lock);
+	write_unlock_bh(&priv->lock);
 }
 
 static void nft_rbtree_activate(const struct net *net,
@@ -224,12 +224,12 @@ static void nft_rbtree_walk(const struct nft_ctx *ctx,
 			    struct nft_set *set,
 			    struct nft_set_iter *iter)
 {
-	const struct nft_rbtree *priv = nft_set_priv(set);
+	struct nft_rbtree *priv = nft_set_priv(set);
 	struct nft_rbtree_elem *rbe;
 	struct nft_set_elem elem;
 	struct rb_node *node;
 
-	spin_lock_bh(&nft_rbtree_lock);
+	read_lock_bh(&priv->lock);
 	for (node = rb_first(&priv->root); node != NULL; node = rb_next(node)) {
 		rbe = rb_entry(node, struct nft_rbtree_elem, node);
 
@@ -242,13 +242,13 @@ static void nft_rbtree_walk(const struct nft_ctx *ctx,
 
 		iter->err = iter->fn(ctx, set, iter, &elem);
 		if (iter->err < 0) {
-			spin_unlock_bh(&nft_rbtree_lock);
+			read_unlock_bh(&priv->lock);
 			return;
 		}
 cont:
 		iter->count++;
 	}
-	spin_unlock_bh(&nft_rbtree_lock);
+	read_unlock_bh(&priv->lock);
 }
 
 static unsigned int nft_rbtree_privsize(const struct nlattr * const nla[])
@@ -262,6 +262,7 @@ static int nft_rbtree_init(const struct nft_set *set,
 {
 	struct nft_rbtree *priv = nft_set_priv(set);
 
+	rwlock_init(&priv->lock);
 	priv->root = RB_ROOT;
 	return 0;
 }

From 864e91ca981ea1d18f05a8f5063520d103373266 Mon Sep 17 00:00:00 2001
From: Cong Wang <xiyou.wangcong@gmail.com>
Date: Fri, 9 Dec 2016 21:09:59 -0800
Subject: [PATCH 16/22] ipvs: remove an annoying printk in netns init

At most it is used for debugging purpose, but I don't think
it is even useful for debugging, just remove it.

Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
---
 net/netfilter/ipvs/ip_vs_core.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c
index db40050f8785..9aaa49025cdc 100644
--- a/net/netfilter/ipvs/ip_vs_core.c
+++ b/net/netfilter/ipvs/ip_vs_core.c
@@ -2231,8 +2231,6 @@ static int __net_init __ip_vs_init(struct net *net)
 	if (ip_vs_sync_net_init(ipvs) < 0)
 		goto sync_fail;
 
-	printk(KERN_INFO "IPVS: Creating netns size=%zu id=%d\n",
-			 sizeof(struct netns_ipvs), ipvs->gen);
 	return 0;
 /*
  * Error handling

From a2f346d82bcab1927a199f475dc299c00413ec39 Mon Sep 17 00:00:00 2001
From: Hangbin Liu <liuhangbin@gmail.com>
Date: Mon, 20 Feb 2017 16:31:35 +0800
Subject: [PATCH 17/22] ipvs: fix sync_threshold description and add
 sync_refresh_period, sync_retries

Fix sync_threshold description which should have two values. Also add
sync_refresh_period and sync_retries based on commit 749c42b620a9
("ipvs: reduce sync rate with time thresholds").

Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
---
 Documentation/networking/ipvs-sysctl.txt | 38 +++++++++++++++++++-----
 1 file changed, 30 insertions(+), 8 deletions(-)

diff --git a/Documentation/networking/ipvs-sysctl.txt b/Documentation/networking/ipvs-sysctl.txt
index e6b1c025fdd8..7acaaa65451e 100644
--- a/Documentation/networking/ipvs-sysctl.txt
+++ b/Documentation/networking/ipvs-sysctl.txt
@@ -185,15 +185,37 @@ secure_tcp - INTEGER
         The value definition is the same as that of drop_entry and
         drop_packet.
 
-sync_threshold - INTEGER
-        default 3
+sync_threshold - vector of 2 INTEGERs: sync_threshold, sync_period
+	default 3 50
 
-        It sets synchronization threshold, which is the minimum number
-        of incoming packets that a connection needs to receive before
-        the connection will be synchronized. A connection will be
-        synchronized, every time the number of its incoming packets
-        modulus 50 equals the threshold. The range of the threshold is
-        from 0 to 49.
+	It sets synchronization threshold, which is the minimum number
+	of incoming packets that a connection needs to receive before
+	the connection will be synchronized. A connection will be
+	synchronized, every time the number of its incoming packets
+	modulus sync_period equals the threshold. The range of the
+	threshold is from 0 to sync_period.
+
+	When sync_period and sync_refresh_period are 0, send sync only
+	for state changes or only once when pkts matches sync_threshold
+
+sync_refresh_period - UNSIGNED INTEGER
+	default 0
+
+	In seconds, difference in reported connection timer that triggers
+	new sync message. It can be used to avoid sync messages for the
+	specified period (or half of the connection timeout if it is lower)
+	if connection state is not changed since last sync.
+
+	This is useful for normal connections with high traffic to reduce
+	sync rate. Additionally, retry sync_retries times with period of
+	sync_refresh_period/8.
+
+sync_retries - INTEGER
+	default 0
+
+	Defines sync retries with period of sync_refresh_period/8. Useful
+	to protect against loss of sync messages. The range of the
+	sync_retries is from 0 to 3.
 
 snat_reroute - BOOLEAN
 	0 - disabled

From 237e5722bc615c3440d8476a694448074380fa01 Mon Sep 17 00:00:00 2001
From: Hangbin Liu <liuhangbin@gmail.com>
Date: Mon, 20 Feb 2017 16:31:36 +0800
Subject: [PATCH 18/22] ipvs: Document sysctl sync_qlen_max and sync_sock_size

Document sysctl sync_qlen_max and sync_sock_size based on
commit 1c003b1580e2 ("ipvs: wakeup master thread").

Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
---
 Documentation/networking/ipvs-sysctl.txt | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/Documentation/networking/ipvs-sysctl.txt b/Documentation/networking/ipvs-sysctl.txt
index 7acaaa65451e..159d70b6dff3 100644
--- a/Documentation/networking/ipvs-sysctl.txt
+++ b/Documentation/networking/ipvs-sysctl.txt
@@ -217,6 +217,20 @@ sync_retries - INTEGER
 	to protect against loss of sync messages. The range of the
 	sync_retries is from 0 to 3.
 
+sync_qlen_max - UNSIGNED LONG
+
+	Hard limit for queued sync messages that are not sent yet. It
+	defaults to 1/32 of the memory pages but actually represents
+	number of messages. It will protect us from allocating large
+	parts of memory when the sending rate is lower than the queuing
+	rate.
+
+sync_sock_size - INTEGER
+	default 0
+
+	Configuration of SNDBUF (master) or RCVBUF (slave) socket limit.
+	Default value is 0 (preserve system defaults).
+
 snat_reroute - BOOLEAN
 	0 - disabled
 	not 0 - enabled (default)

From 24b444155b1e237accb2571ac656cf40c941a2a7 Mon Sep 17 00:00:00 2001
From: Hangbin Liu <liuhangbin@gmail.com>
Date: Mon, 20 Feb 2017 16:31:37 +0800
Subject: [PATCH 19/22] ipvs: Document sysctl sync_ports

Document sysctl sync_ports based on commit f73181c8288f ("ipvs: add support
for sync threads").

Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
---
 Documentation/networking/ipvs-sysctl.txt | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/Documentation/networking/ipvs-sysctl.txt b/Documentation/networking/ipvs-sysctl.txt
index 159d70b6dff3..a6feecd467cd 100644
--- a/Documentation/networking/ipvs-sysctl.txt
+++ b/Documentation/networking/ipvs-sysctl.txt
@@ -231,6 +231,14 @@ sync_sock_size - INTEGER
 	Configuration of SNDBUF (master) or RCVBUF (slave) socket limit.
 	Default value is 0 (preserve system defaults).
 
+sync_ports - INTEGER
+	default 1
+
+	The number of threads that master and backup servers can use for
+	sync traffic. Every thread will use single UDP port, thread 0 will
+	use the default port 8848 while last thread will use port
+	8848+sync_ports-1.
+
 snat_reroute - BOOLEAN
 	0 - disabled
 	not 0 - enabled (default)

From 3c679cba588a46ba81a264673e192bbd3c92455b Mon Sep 17 00:00:00 2001
From: Hangbin Liu <liuhangbin@gmail.com>
Date: Mon, 20 Feb 2017 16:31:38 +0800
Subject: [PATCH 20/22] ipvs: Document sysctl pmtu_disc

Document sysctl pmtu_disc based on commit 3654e61137db ("ipvs: add
pmtu_disc option to disable IP DF for TUN packets").

Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
---
 Documentation/networking/ipvs-sysctl.txt | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/Documentation/networking/ipvs-sysctl.txt b/Documentation/networking/ipvs-sysctl.txt
index a6feecd467cd..056898685d40 100644
--- a/Documentation/networking/ipvs-sysctl.txt
+++ b/Documentation/networking/ipvs-sysctl.txt
@@ -175,6 +175,14 @@ nat_icmp_send - BOOLEAN
         for VS/NAT when the load balancer receives packets from real
         servers but the connection entries don't exist.
 
+pmtu_disc - BOOLEAN
+	0 - disabled
+	not 0 - enabled (default)
+
+	By default, reject with FRAG_NEEDED all DF packets that exceed
+	the PMTU, irrespective of the forwarding method. For TUN method
+	the flag can be disabled to fragment such packets.
+
 secure_tcp - INTEGER
         0  - disabled (default)
 

From b54ab92b84b6161f91b1ad9160199422b3699009 Mon Sep 17 00:00:00 2001
From: "Reshetova, Elena" <elena.reshetova@intel.com>
Date: Thu, 16 Mar 2017 10:03:34 +0200
Subject: [PATCH 21/22] netfilter: refcounter conversions

refcount_t type and corresponding API (see include/linux/refcount.h)
should be used instead of atomic_t when the variable is used as
a reference counter. This allows to avoid accidental
refcounter overflows that might lead to use-after-free
situations.

Signed-off-by: Elena Reshetova <elena.reshetova@intel.com>
Signed-off-by: Hans Liljestrand <ishkamiel@gmail.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: David Windsor <dwindsor@gmail.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/ip_vs.h                          | 16 +++++++------
 include/net/netfilter/nf_conntrack_expect.h  |  4 +++-
 include/net/netfilter/nf_conntrack_timeout.h |  3 ++-
 net/ipv4/netfilter/ipt_CLUSTERIP.c           | 19 ++++++++--------
 net/netfilter/ipvs/ip_vs_conn.c              | 24 ++++++++++----------
 net/netfilter/ipvs/ip_vs_core.c              |  4 ++--
 net/netfilter/ipvs/ip_vs_ctl.c               | 12 +++++-----
 net/netfilter/ipvs/ip_vs_lblc.c              |  2 +-
 net/netfilter/ipvs/ip_vs_lblcr.c             |  6 ++---
 net/netfilter/ipvs/ip_vs_nq.c                |  2 +-
 net/netfilter/ipvs/ip_vs_proto_sctp.c        |  2 +-
 net/netfilter/ipvs/ip_vs_proto_tcp.c         |  2 +-
 net/netfilter/ipvs/ip_vs_rr.c                |  2 +-
 net/netfilter/ipvs/ip_vs_sed.c               |  2 +-
 net/netfilter/ipvs/ip_vs_wlc.c               |  2 +-
 net/netfilter/ipvs/ip_vs_wrr.c               |  2 +-
 net/netfilter/nf_conntrack_expect.c          | 10 ++++----
 net/netfilter/nf_conntrack_netlink.c         |  4 ++--
 net/netfilter/nfnetlink_acct.c               | 16 +++++++------
 net/netfilter/nfnetlink_cttimeout.c          | 12 +++++-----
 net/netfilter/nfnetlink_log.c                | 14 +++++++-----
 21 files changed, 85 insertions(+), 75 deletions(-)

diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index 7bdfa7d78363..8a4a57b887fb 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -12,6 +12,8 @@
 #include <linux/list.h>                 /* for struct list_head */
 #include <linux/spinlock.h>             /* for struct rwlock_t */
 #include <linux/atomic.h>               /* for struct atomic_t */
+#include <linux/refcount.h>             /* for struct refcount_t */
+
 #include <linux/compiler.h>
 #include <linux/timer.h>
 #include <linux/bug.h>
@@ -525,7 +527,7 @@ struct ip_vs_conn {
 	struct netns_ipvs	*ipvs;
 
 	/* counter and timer */
-	atomic_t		refcnt;		/* reference count */
+	refcount_t		refcnt;		/* reference count */
 	struct timer_list	timer;		/* Expiration timer */
 	volatile unsigned long	timeout;	/* timeout */
 
@@ -667,7 +669,7 @@ struct ip_vs_dest {
 	atomic_t		conn_flags;	/* flags to copy to conn */
 	atomic_t		weight;		/* server weight */
 
-	atomic_t		refcnt;		/* reference counter */
+	refcount_t		refcnt;		/* reference counter */
 	struct ip_vs_stats      stats;          /* statistics */
 	unsigned long		idle_start;	/* start time, jiffies */
 
@@ -1211,14 +1213,14 @@ struct ip_vs_conn * ip_vs_conn_out_get_proto(struct netns_ipvs *ipvs, int af,
  */
 static inline bool __ip_vs_conn_get(struct ip_vs_conn *cp)
 {
-	return atomic_inc_not_zero(&cp->refcnt);
+	return refcount_inc_not_zero(&cp->refcnt);
 }
 
 /* put back the conn without restarting its timer */
 static inline void __ip_vs_conn_put(struct ip_vs_conn *cp)
 {
 	smp_mb__before_atomic();
-	atomic_dec(&cp->refcnt);
+	refcount_dec(&cp->refcnt);
 }
 void ip_vs_conn_put(struct ip_vs_conn *cp);
 void ip_vs_conn_fill_cport(struct ip_vs_conn *cp, __be16 cport);
@@ -1410,18 +1412,18 @@ void ip_vs_try_bind_dest(struct ip_vs_conn *cp);
 
 static inline void ip_vs_dest_hold(struct ip_vs_dest *dest)
 {
-	atomic_inc(&dest->refcnt);
+	refcount_inc(&dest->refcnt);
 }
 
 static inline void ip_vs_dest_put(struct ip_vs_dest *dest)
 {
 	smp_mb__before_atomic();
-	atomic_dec(&dest->refcnt);
+	refcount_dec(&dest->refcnt);
 }
 
 static inline void ip_vs_dest_put_and_free(struct ip_vs_dest *dest)
 {
-	if (atomic_dec_and_test(&dest->refcnt))
+	if (refcount_dec_and_test(&dest->refcnt))
 		kfree(dest);
 }
 
diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h
index 5ed33ea4718e..65cc2cb005d9 100644
--- a/include/net/netfilter/nf_conntrack_expect.h
+++ b/include/net/netfilter/nf_conntrack_expect.h
@@ -5,6 +5,8 @@
 #ifndef _NF_CONNTRACK_EXPECT_H
 #define _NF_CONNTRACK_EXPECT_H
 
+#include <linux/refcount.h>
+
 #include <net/netfilter/nf_conntrack.h>
 #include <net/netfilter/nf_conntrack_zones.h>
 
@@ -37,7 +39,7 @@ struct nf_conntrack_expect {
 	struct timer_list timeout;
 
 	/* Usage count. */
-	atomic_t use;
+	refcount_t use;
 
 	/* Flags */
 	unsigned int flags;
diff --git a/include/net/netfilter/nf_conntrack_timeout.h b/include/net/netfilter/nf_conntrack_timeout.h
index 5cc5e9e6171a..d40b89355fdd 100644
--- a/include/net/netfilter/nf_conntrack_timeout.h
+++ b/include/net/netfilter/nf_conntrack_timeout.h
@@ -4,6 +4,7 @@
 #include <net/net_namespace.h>
 #include <linux/netfilter/nf_conntrack_common.h>
 #include <linux/netfilter/nf_conntrack_tuple_common.h>
+#include <linux/refcount.h>
 #include <net/netfilter/nf_conntrack.h>
 #include <net/netfilter/nf_conntrack_extend.h>
 
@@ -12,7 +13,7 @@
 struct ctnl_timeout {
 	struct list_head	head;
 	struct rcu_head		rcu_head;
-	atomic_t		refcnt;
+	refcount_t		refcnt;
 	char			name[CTNL_TIMEOUT_NAME_MAX];
 	__u16			l3num;
 	struct nf_conntrack_l4proto *l4proto;
diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c
index 52f26459efc3..fcbdc0c49b0e 100644
--- a/net/ipv4/netfilter/ipt_CLUSTERIP.c
+++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c
@@ -22,6 +22,7 @@
 #include <linux/icmp.h>
 #include <linux/if_arp.h>
 #include <linux/seq_file.h>
+#include <linux/refcount.h>
 #include <linux/netfilter_arp.h>
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter_ipv4/ip_tables.h>
@@ -40,8 +41,8 @@ MODULE_DESCRIPTION("Xtables: CLUSTERIP target");
 
 struct clusterip_config {
 	struct list_head list;			/* list of all configs */
-	atomic_t refcount;			/* reference count */
-	atomic_t entries;			/* number of entries/rules
+	refcount_t refcount;			/* reference count */
+	refcount_t entries;			/* number of entries/rules
 						 * referencing us */
 
 	__be32 clusterip;			/* the IP address */
@@ -77,7 +78,7 @@ struct clusterip_net {
 static inline void
 clusterip_config_get(struct clusterip_config *c)
 {
-	atomic_inc(&c->refcount);
+	refcount_inc(&c->refcount);
 }
 
 
@@ -89,7 +90,7 @@ static void clusterip_config_rcu_free(struct rcu_head *head)
 static inline void
 clusterip_config_put(struct clusterip_config *c)
 {
-	if (atomic_dec_and_test(&c->refcount))
+	if (refcount_dec_and_test(&c->refcount))
 		call_rcu_bh(&c->rcu, clusterip_config_rcu_free);
 }
 
@@ -103,7 +104,7 @@ clusterip_config_entry_put(struct clusterip_config *c)
 	struct clusterip_net *cn = net_generic(net, clusterip_net_id);
 
 	local_bh_disable();
-	if (atomic_dec_and_lock(&c->entries, &cn->lock)) {
+	if (refcount_dec_and_lock(&c->entries, &cn->lock)) {
 		list_del_rcu(&c->list);
 		spin_unlock(&cn->lock);
 		local_bh_enable();
@@ -149,10 +150,10 @@ clusterip_config_find_get(struct net *net, __be32 clusterip, int entry)
 			c = NULL;
 		else
 #endif
-		if (unlikely(!atomic_inc_not_zero(&c->refcount)))
+		if (unlikely(!refcount_inc_not_zero(&c->refcount)))
 			c = NULL;
 		else if (entry)
-			atomic_inc(&c->entries);
+			refcount_inc(&c->entries);
 	}
 	rcu_read_unlock_bh();
 
@@ -188,8 +189,8 @@ clusterip_config_init(const struct ipt_clusterip_tgt_info *i, __be32 ip,
 	clusterip_config_init_nodelist(c, i);
 	c->hash_mode = i->hash_mode;
 	c->hash_initval = i->hash_initval;
-	atomic_set(&c->refcount, 1);
-	atomic_set(&c->entries, 1);
+	refcount_set(&c->refcount, 1);
+	refcount_set(&c->entries, 1);
 
 	spin_lock_bh(&cn->lock);
 	if (__clusterip_config_find(net, ip)) {
diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c
index e6a2753dff9e..3d2ac71a83ec 100644
--- a/net/netfilter/ipvs/ip_vs_conn.c
+++ b/net/netfilter/ipvs/ip_vs_conn.c
@@ -181,7 +181,7 @@ static inline int ip_vs_conn_hash(struct ip_vs_conn *cp)
 
 	if (!(cp->flags & IP_VS_CONN_F_HASHED)) {
 		cp->flags |= IP_VS_CONN_F_HASHED;
-		atomic_inc(&cp->refcnt);
+		refcount_inc(&cp->refcnt);
 		hlist_add_head_rcu(&cp->c_list, &ip_vs_conn_tab[hash]);
 		ret = 1;
 	} else {
@@ -215,7 +215,7 @@ static inline int ip_vs_conn_unhash(struct ip_vs_conn *cp)
 	if (cp->flags & IP_VS_CONN_F_HASHED) {
 		hlist_del_rcu(&cp->c_list);
 		cp->flags &= ~IP_VS_CONN_F_HASHED;
-		atomic_dec(&cp->refcnt);
+		refcount_dec(&cp->refcnt);
 		ret = 1;
 	} else
 		ret = 0;
@@ -242,13 +242,13 @@ static inline bool ip_vs_conn_unlink(struct ip_vs_conn *cp)
 	if (cp->flags & IP_VS_CONN_F_HASHED) {
 		ret = false;
 		/* Decrease refcnt and unlink conn only if we are last user */
-		if (atomic_cmpxchg(&cp->refcnt, 1, 0) == 1) {
+		if (refcount_dec_if_one(&cp->refcnt)) {
 			hlist_del_rcu(&cp->c_list);
 			cp->flags &= ~IP_VS_CONN_F_HASHED;
 			ret = true;
 		}
 	} else
-		ret = atomic_read(&cp->refcnt) ? false : true;
+		ret = refcount_read(&cp->refcnt) ? false : true;
 
 	spin_unlock(&cp->lock);
 	ct_write_unlock_bh(hash);
@@ -475,7 +475,7 @@ static void __ip_vs_conn_put_timer(struct ip_vs_conn *cp)
 void ip_vs_conn_put(struct ip_vs_conn *cp)
 {
 	if ((cp->flags & IP_VS_CONN_F_ONE_PACKET) &&
-	    (atomic_read(&cp->refcnt) == 1) &&
+	    (refcount_read(&cp->refcnt) == 1) &&
 	    !timer_pending(&cp->timer))
 		/* expire connection immediately */
 		__ip_vs_conn_put_notimer(cp);
@@ -617,8 +617,8 @@ ip_vs_bind_dest(struct ip_vs_conn *cp, struct ip_vs_dest *dest)
 		      IP_VS_DBG_ADDR(cp->af, &cp->vaddr), ntohs(cp->vport),
 		      IP_VS_DBG_ADDR(cp->daf, &cp->daddr), ntohs(cp->dport),
 		      ip_vs_fwd_tag(cp), cp->state,
-		      cp->flags, atomic_read(&cp->refcnt),
-		      atomic_read(&dest->refcnt));
+		      cp->flags, refcount_read(&cp->refcnt),
+		      refcount_read(&dest->refcnt));
 
 	/* Update the connection counters */
 	if (!(flags & IP_VS_CONN_F_TEMPLATE)) {
@@ -714,8 +714,8 @@ static inline void ip_vs_unbind_dest(struct ip_vs_conn *cp)
 		      IP_VS_DBG_ADDR(cp->af, &cp->vaddr), ntohs(cp->vport),
 		      IP_VS_DBG_ADDR(cp->daf, &cp->daddr), ntohs(cp->dport),
 		      ip_vs_fwd_tag(cp), cp->state,
-		      cp->flags, atomic_read(&cp->refcnt),
-		      atomic_read(&dest->refcnt));
+		      cp->flags, refcount_read(&cp->refcnt),
+		      refcount_read(&dest->refcnt));
 
 	/* Update the connection counters */
 	if (!(cp->flags & IP_VS_CONN_F_TEMPLATE)) {
@@ -863,10 +863,10 @@ static void ip_vs_conn_expire(unsigned long data)
 
   expire_later:
 	IP_VS_DBG(7, "delayed: conn->refcnt=%d conn->n_control=%d\n",
-		  atomic_read(&cp->refcnt),
+		  refcount_read(&cp->refcnt),
 		  atomic_read(&cp->n_control));
 
-	atomic_inc(&cp->refcnt);
+	refcount_inc(&cp->refcnt);
 	cp->timeout = 60*HZ;
 
 	if (ipvs->sync_state & IP_VS_STATE_MASTER)
@@ -941,7 +941,7 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p, int dest_af,
 	 * it in the table, so that other thread run ip_vs_random_dropentry
 	 * but cannot drop this entry.
 	 */
-	atomic_set(&cp->refcnt, 1);
+	refcount_set(&cp->refcnt, 1);
 
 	cp->control = NULL;
 	atomic_set(&cp->n_control, 0);
diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c
index 9aaa49025cdc..b4a746d0e39b 100644
--- a/net/netfilter/ipvs/ip_vs_core.c
+++ b/net/netfilter/ipvs/ip_vs_core.c
@@ -542,7 +542,7 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb,
 		      IP_VS_DBG_ADDR(cp->af, &cp->caddr), ntohs(cp->cport),
 		      IP_VS_DBG_ADDR(cp->af, &cp->vaddr), ntohs(cp->vport),
 		      IP_VS_DBG_ADDR(cp->daf, &cp->daddr), ntohs(cp->dport),
-		      cp->flags, atomic_read(&cp->refcnt));
+		      cp->flags, refcount_read(&cp->refcnt));
 
 	ip_vs_conn_stats(cp, svc);
 	return cp;
@@ -1193,7 +1193,7 @@ struct ip_vs_conn *ip_vs_new_conn_out(struct ip_vs_service *svc,
 		      IP_VS_DBG_ADDR(cp->af, &cp->caddr), ntohs(cp->cport),
 		      IP_VS_DBG_ADDR(cp->af, &cp->vaddr), ntohs(cp->vport),
 		      IP_VS_DBG_ADDR(cp->af, &cp->daddr), ntohs(cp->dport),
-		      cp->flags, atomic_read(&cp->refcnt));
+		      cp->flags, refcount_read(&cp->refcnt));
 	LeaveFunction(12);
 	return cp;
 }
diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c
index 5aeb0dde6ccc..541aa7694775 100644
--- a/net/netfilter/ipvs/ip_vs_ctl.c
+++ b/net/netfilter/ipvs/ip_vs_ctl.c
@@ -699,7 +699,7 @@ ip_vs_trash_get_dest(struct ip_vs_service *svc, int dest_af,
 			      dest->vfwmark,
 			      IP_VS_DBG_ADDR(dest->af, &dest->addr),
 			      ntohs(dest->port),
-			      atomic_read(&dest->refcnt));
+			      refcount_read(&dest->refcnt));
 		if (dest->af == dest_af &&
 		    ip_vs_addr_equal(dest_af, &dest->addr, daddr) &&
 		    dest->port == dport &&
@@ -934,7 +934,7 @@ ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest,
 	atomic_set(&dest->activeconns, 0);
 	atomic_set(&dest->inactconns, 0);
 	atomic_set(&dest->persistconns, 0);
-	atomic_set(&dest->refcnt, 1);
+	refcount_set(&dest->refcnt, 1);
 
 	INIT_HLIST_NODE(&dest->d_list);
 	spin_lock_init(&dest->dst_lock);
@@ -998,7 +998,7 @@ ip_vs_add_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest)
 		IP_VS_DBG_BUF(3, "Get destination %s:%u from trash, "
 			      "dest->refcnt=%d, service %u/%s:%u\n",
 			      IP_VS_DBG_ADDR(udest->af, &daddr), ntohs(dport),
-			      atomic_read(&dest->refcnt),
+			      refcount_read(&dest->refcnt),
 			      dest->vfwmark,
 			      IP_VS_DBG_ADDR(svc->af, &dest->vaddr),
 			      ntohs(dest->vport));
@@ -1074,7 +1074,7 @@ static void __ip_vs_del_dest(struct netns_ipvs *ipvs, struct ip_vs_dest *dest,
 	spin_lock_bh(&ipvs->dest_trash_lock);
 	IP_VS_DBG_BUF(3, "Moving dest %s:%u into trash, dest->refcnt=%d\n",
 		      IP_VS_DBG_ADDR(dest->af, &dest->addr), ntohs(dest->port),
-		      atomic_read(&dest->refcnt));
+		      refcount_read(&dest->refcnt));
 	if (list_empty(&ipvs->dest_trash) && !cleanup)
 		mod_timer(&ipvs->dest_trash_timer,
 			  jiffies + (IP_VS_DEST_TRASH_PERIOD >> 1));
@@ -1157,7 +1157,7 @@ static void ip_vs_dest_trash_expire(unsigned long data)
 
 	spin_lock(&ipvs->dest_trash_lock);
 	list_for_each_entry_safe(dest, next, &ipvs->dest_trash, t_list) {
-		if (atomic_read(&dest->refcnt) > 1)
+		if (refcount_read(&dest->refcnt) > 1)
 			continue;
 		if (dest->idle_start) {
 			if (time_before(now, dest->idle_start +
@@ -1545,7 +1545,7 @@ ip_vs_forget_dev(struct ip_vs_dest *dest, struct net_device *dev)
 			      dev->name,
 			      IP_VS_DBG_ADDR(dest->af, &dest->addr),
 			      ntohs(dest->port),
-			      atomic_read(&dest->refcnt));
+			      refcount_read(&dest->refcnt));
 		__ip_vs_dst_cache_reset(dest);
 	}
 	spin_unlock_bh(&dest->dst_lock);
diff --git a/net/netfilter/ipvs/ip_vs_lblc.c b/net/netfilter/ipvs/ip_vs_lblc.c
index 5824927cf8e0..b6aa4a970c6e 100644
--- a/net/netfilter/ipvs/ip_vs_lblc.c
+++ b/net/netfilter/ipvs/ip_vs_lblc.c
@@ -448,7 +448,7 @@ __ip_vs_lblc_schedule(struct ip_vs_service *svc)
 		      IP_VS_DBG_ADDR(least->af, &least->addr),
 		      ntohs(least->port),
 		      atomic_read(&least->activeconns),
-		      atomic_read(&least->refcnt),
+		      refcount_read(&least->refcnt),
 		      atomic_read(&least->weight), loh);
 
 	return least;
diff --git a/net/netfilter/ipvs/ip_vs_lblcr.c b/net/netfilter/ipvs/ip_vs_lblcr.c
index 703f11877bee..c13ff575f9f7 100644
--- a/net/netfilter/ipvs/ip_vs_lblcr.c
+++ b/net/netfilter/ipvs/ip_vs_lblcr.c
@@ -204,7 +204,7 @@ static inline struct ip_vs_dest *ip_vs_dest_set_min(struct ip_vs_dest_set *set)
 		      IP_VS_DBG_ADDR(least->af, &least->addr),
 		      ntohs(least->port),
 		      atomic_read(&least->activeconns),
-		      atomic_read(&least->refcnt),
+		      refcount_read(&least->refcnt),
 		      atomic_read(&least->weight), loh);
 	return least;
 }
@@ -249,7 +249,7 @@ static inline struct ip_vs_dest *ip_vs_dest_set_max(struct ip_vs_dest_set *set)
 		      __func__,
 		      IP_VS_DBG_ADDR(most->af, &most->addr), ntohs(most->port),
 		      atomic_read(&most->activeconns),
-		      atomic_read(&most->refcnt),
+		      refcount_read(&most->refcnt),
 		      atomic_read(&most->weight), moh);
 	return most;
 }
@@ -612,7 +612,7 @@ __ip_vs_lblcr_schedule(struct ip_vs_service *svc)
 		      IP_VS_DBG_ADDR(least->af, &least->addr),
 		      ntohs(least->port),
 		      atomic_read(&least->activeconns),
-		      atomic_read(&least->refcnt),
+		      refcount_read(&least->refcnt),
 		      atomic_read(&least->weight), loh);
 
 	return least;
diff --git a/net/netfilter/ipvs/ip_vs_nq.c b/net/netfilter/ipvs/ip_vs_nq.c
index a8b63401e773..7d9d4ac596ca 100644
--- a/net/netfilter/ipvs/ip_vs_nq.c
+++ b/net/netfilter/ipvs/ip_vs_nq.c
@@ -110,7 +110,7 @@ ip_vs_nq_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
 		      IP_VS_DBG_ADDR(least->af, &least->addr),
 		      ntohs(least->port),
 		      atomic_read(&least->activeconns),
-		      atomic_read(&least->refcnt),
+		      refcount_read(&least->refcnt),
 		      atomic_read(&least->weight), loh);
 
 	return least;
diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c
index d952d67f904d..56f8e4b204ff 100644
--- a/net/netfilter/ipvs/ip_vs_proto_sctp.c
+++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c
@@ -447,7 +447,7 @@ set_sctp_state(struct ip_vs_proto_data *pd, struct ip_vs_conn *cp,
 				ntohs(cp->cport),
 				sctp_state_name(cp->state),
 				sctp_state_name(next_state),
-				atomic_read(&cp->refcnt));
+				refcount_read(&cp->refcnt));
 		if (dest) {
 			if (!(cp->flags & IP_VS_CONN_F_INACTIVE) &&
 				(next_state != IP_VS_SCTP_S_ESTABLISHED)) {
diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c
index 5117bcb7d2f0..12dc8d5bc37d 100644
--- a/net/netfilter/ipvs/ip_vs_proto_tcp.c
+++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c
@@ -557,7 +557,7 @@ set_tcp_state(struct ip_vs_proto_data *pd, struct ip_vs_conn *cp,
 			      ntohs(cp->cport),
 			      tcp_state_name(cp->state),
 			      tcp_state_name(new_state),
-			      atomic_read(&cp->refcnt));
+			      refcount_read(&cp->refcnt));
 
 		if (dest) {
 			if (!(cp->flags & IP_VS_CONN_F_INACTIVE) &&
diff --git a/net/netfilter/ipvs/ip_vs_rr.c b/net/netfilter/ipvs/ip_vs_rr.c
index 58bacfc461ee..ee0530d14c5f 100644
--- a/net/netfilter/ipvs/ip_vs_rr.c
+++ b/net/netfilter/ipvs/ip_vs_rr.c
@@ -97,7 +97,7 @@ stop:
 		      "activeconns %d refcnt %d weight %d\n",
 		      IP_VS_DBG_ADDR(dest->af, &dest->addr), ntohs(dest->port),
 		      atomic_read(&dest->activeconns),
-		      atomic_read(&dest->refcnt), atomic_read(&dest->weight));
+		      refcount_read(&dest->refcnt), atomic_read(&dest->weight));
 
 	return dest;
 }
diff --git a/net/netfilter/ipvs/ip_vs_sed.c b/net/netfilter/ipvs/ip_vs_sed.c
index f8e2d00f528b..ab23cf203437 100644
--- a/net/netfilter/ipvs/ip_vs_sed.c
+++ b/net/netfilter/ipvs/ip_vs_sed.c
@@ -111,7 +111,7 @@ ip_vs_sed_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
 		      IP_VS_DBG_ADDR(least->af, &least->addr),
 		      ntohs(least->port),
 		      atomic_read(&least->activeconns),
-		      atomic_read(&least->refcnt),
+		      refcount_read(&least->refcnt),
 		      atomic_read(&least->weight), loh);
 
 	return least;
diff --git a/net/netfilter/ipvs/ip_vs_wlc.c b/net/netfilter/ipvs/ip_vs_wlc.c
index 6b366fd90554..6add39e0ec20 100644
--- a/net/netfilter/ipvs/ip_vs_wlc.c
+++ b/net/netfilter/ipvs/ip_vs_wlc.c
@@ -83,7 +83,7 @@ ip_vs_wlc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
 		      IP_VS_DBG_ADDR(least->af, &least->addr),
 		      ntohs(least->port),
 		      atomic_read(&least->activeconns),
-		      atomic_read(&least->refcnt),
+		      refcount_read(&least->refcnt),
 		      atomic_read(&least->weight), loh);
 
 	return least;
diff --git a/net/netfilter/ipvs/ip_vs_wrr.c b/net/netfilter/ipvs/ip_vs_wrr.c
index 17e6d4406ca7..62258dd457ac 100644
--- a/net/netfilter/ipvs/ip_vs_wrr.c
+++ b/net/netfilter/ipvs/ip_vs_wrr.c
@@ -218,7 +218,7 @@ found:
 		      "activeconns %d refcnt %d weight %d\n",
 		      IP_VS_DBG_ADDR(dest->af, &dest->addr), ntohs(dest->port),
 		      atomic_read(&dest->activeconns),
-		      atomic_read(&dest->refcnt),
+		      refcount_read(&dest->refcnt),
 		      atomic_read(&dest->weight));
 	mark->cl = dest;
 
diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c
index 4b2e1fb28bb4..cb29e598605f 100644
--- a/net/netfilter/nf_conntrack_expect.c
+++ b/net/netfilter/nf_conntrack_expect.c
@@ -133,7 +133,7 @@ nf_ct_expect_find_get(struct net *net,
 
 	rcu_read_lock();
 	i = __nf_ct_expect_find(net, zone, tuple);
-	if (i && !atomic_inc_not_zero(&i->use))
+	if (i && !refcount_inc_not_zero(&i->use))
 		i = NULL;
 	rcu_read_unlock();
 
@@ -186,7 +186,7 @@ nf_ct_find_expectation(struct net *net,
 		return NULL;
 
 	if (exp->flags & NF_CT_EXPECT_PERMANENT) {
-		atomic_inc(&exp->use);
+		refcount_inc(&exp->use);
 		return exp;
 	} else if (del_timer(&exp->timeout)) {
 		nf_ct_unlink_expect(exp);
@@ -275,7 +275,7 @@ struct nf_conntrack_expect *nf_ct_expect_alloc(struct nf_conn *me)
 		return NULL;
 
 	new->master = me;
-	atomic_set(&new->use, 1);
+	refcount_set(&new->use, 1);
 	return new;
 }
 EXPORT_SYMBOL_GPL(nf_ct_expect_alloc);
@@ -348,7 +348,7 @@ static void nf_ct_expect_free_rcu(struct rcu_head *head)
 
 void nf_ct_expect_put(struct nf_conntrack_expect *exp)
 {
-	if (atomic_dec_and_test(&exp->use))
+	if (refcount_dec_and_test(&exp->use))
 		call_rcu(&exp->rcu, nf_ct_expect_free_rcu);
 }
 EXPORT_SYMBOL_GPL(nf_ct_expect_put);
@@ -361,7 +361,7 @@ static void nf_ct_expect_insert(struct nf_conntrack_expect *exp)
 	unsigned int h = nf_ct_expect_dst_hash(net, &exp->tuple);
 
 	/* two references : one for hash insert, one for the timer */
-	atomic_add(2, &exp->use);
+	refcount_add(2, &exp->use);
 
 	hlist_add_head(&exp->lnode, &master_help->expectations);
 	master_help->expecting[exp->class]++;
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 6806b5e73567..d49cc1e03c5b 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -2693,7 +2693,7 @@ restart:
 						    cb->nlh->nlmsg_seq,
 						    IPCTNL_MSG_EXP_NEW,
 						    exp) < 0) {
-				if (!atomic_inc_not_zero(&exp->use))
+				if (!refcount_inc_not_zero(&exp->use))
 					continue;
 				cb->args[1] = (unsigned long)exp;
 				goto out;
@@ -2739,7 +2739,7 @@ restart:
 					    cb->nlh->nlmsg_seq,
 					    IPCTNL_MSG_EXP_NEW,
 					    exp) < 0) {
-			if (!atomic_inc_not_zero(&exp->use))
+			if (!refcount_inc_not_zero(&exp->use))
 				continue;
 			cb->args[1] = (unsigned long)exp;
 			goto out;
diff --git a/net/netfilter/nfnetlink_acct.c b/net/netfilter/nfnetlink_acct.c
index d44d89b56127..f44cbd35357f 100644
--- a/net/netfilter/nfnetlink_acct.c
+++ b/net/netfilter/nfnetlink_acct.c
@@ -11,6 +11,7 @@
 #include <linux/kernel.h>
 #include <linux/skbuff.h>
 #include <linux/atomic.h>
+#include <linux/refcount.h>
 #include <linux/netlink.h>
 #include <linux/rculist.h>
 #include <linux/slab.h>
@@ -32,7 +33,7 @@ struct nf_acct {
 	atomic64_t		bytes;
 	unsigned long		flags;
 	struct list_head	head;
-	atomic_t		refcnt;
+	refcount_t		refcnt;
 	char			name[NFACCT_NAME_MAX];
 	struct rcu_head		rcu_head;
 	char			data[0];
@@ -123,7 +124,7 @@ static int nfnl_acct_new(struct net *net, struct sock *nfnl,
 		atomic64_set(&nfacct->pkts,
 			     be64_to_cpu(nla_get_be64(tb[NFACCT_PKTS])));
 	}
-	atomic_set(&nfacct->refcnt, 1);
+	refcount_set(&nfacct->refcnt, 1);
 	list_add_tail_rcu(&nfacct->head, &net->nfnl_acct_list);
 	return 0;
 }
@@ -166,7 +167,7 @@ nfnl_acct_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type,
 			 NFACCT_PAD) ||
 	    nla_put_be64(skb, NFACCT_BYTES, cpu_to_be64(bytes),
 			 NFACCT_PAD) ||
-	    nla_put_be32(skb, NFACCT_USE, htonl(atomic_read(&acct->refcnt))))
+	    nla_put_be32(skb, NFACCT_USE, htonl(refcount_read(&acct->refcnt))))
 		goto nla_put_failure;
 	if (acct->flags & NFACCT_F_QUOTA) {
 		u64 *quota = (u64 *)acct->data;
@@ -325,11 +326,12 @@ static int nfnl_acct_get(struct net *net, struct sock *nfnl,
 static int nfnl_acct_try_del(struct nf_acct *cur)
 {
 	int ret = 0;
+	unsigned int refcount;
 
 	/* We want to avoid races with nfnl_acct_put. So only when the current
 	 * refcnt is 1, we decrease it to 0.
 	 */
-	if (atomic_cmpxchg(&cur->refcnt, 1, 0) == 1) {
+	if (refcount_dec_if_one(&cur->refcnt)) {
 		/* We are protected by nfnl mutex. */
 		list_del_rcu(&cur->head);
 		kfree_rcu(cur, rcu_head);
@@ -413,7 +415,7 @@ struct nf_acct *nfnl_acct_find_get(struct net *net, const char *acct_name)
 		if (!try_module_get(THIS_MODULE))
 			goto err;
 
-		if (!atomic_inc_not_zero(&cur->refcnt)) {
+		if (!refcount_inc_not_zero(&cur->refcnt)) {
 			module_put(THIS_MODULE);
 			goto err;
 		}
@@ -429,7 +431,7 @@ EXPORT_SYMBOL_GPL(nfnl_acct_find_get);
 
 void nfnl_acct_put(struct nf_acct *acct)
 {
-	if (atomic_dec_and_test(&acct->refcnt))
+	if (refcount_dec_and_test(&acct->refcnt))
 		kfree_rcu(acct, rcu_head);
 
 	module_put(THIS_MODULE);
@@ -502,7 +504,7 @@ static void __net_exit nfnl_acct_net_exit(struct net *net)
 	list_for_each_entry_safe(cur, tmp, &net->nfnl_acct_list, head) {
 		list_del_rcu(&cur->head);
 
-		if (atomic_dec_and_test(&cur->refcnt))
+		if (refcount_dec_and_test(&cur->refcnt))
 			kfree_rcu(cur, rcu_head);
 	}
 }
diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c
index 139e0867e56e..baa75f3ab7e7 100644
--- a/net/netfilter/nfnetlink_cttimeout.c
+++ b/net/netfilter/nfnetlink_cttimeout.c
@@ -138,7 +138,7 @@ static int cttimeout_new_timeout(struct net *net, struct sock *ctnl,
 	strcpy(timeout->name, nla_data(cda[CTA_TIMEOUT_NAME]));
 	timeout->l3num = l3num;
 	timeout->l4proto = l4proto;
-	atomic_set(&timeout->refcnt, 1);
+	refcount_set(&timeout->refcnt, 1);
 	list_add_tail_rcu(&timeout->head, &net->nfct_timeout_list);
 
 	return 0;
@@ -172,7 +172,7 @@ ctnl_timeout_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type,
 	    nla_put_be16(skb, CTA_TIMEOUT_L3PROTO, htons(timeout->l3num)) ||
 	    nla_put_u8(skb, CTA_TIMEOUT_L4PROTO, timeout->l4proto->l4proto) ||
 	    nla_put_be32(skb, CTA_TIMEOUT_USE,
-			 htonl(atomic_read(&timeout->refcnt))))
+			 htonl(refcount_read(&timeout->refcnt))))
 		goto nla_put_failure;
 
 	if (likely(l4proto->ctnl_timeout.obj_to_nlattr)) {
@@ -339,7 +339,7 @@ static int ctnl_timeout_try_del(struct net *net, struct ctnl_timeout *timeout)
 	/* We want to avoid races with ctnl_timeout_put. So only when the
 	 * current refcnt is 1, we decrease it to 0.
 	 */
-	if (atomic_cmpxchg(&timeout->refcnt, 1, 0) == 1) {
+	if (refcount_dec_if_one(&timeout->refcnt)) {
 		/* We are protected by nfnl mutex. */
 		list_del_rcu(&timeout->head);
 		nf_ct_l4proto_put(timeout->l4proto);
@@ -536,7 +536,7 @@ ctnl_timeout_find_get(struct net *net, const char *name)
 		if (!try_module_get(THIS_MODULE))
 			goto err;
 
-		if (!atomic_inc_not_zero(&timeout->refcnt)) {
+		if (!refcount_inc_not_zero(&timeout->refcnt)) {
 			module_put(THIS_MODULE);
 			goto err;
 		}
@@ -550,7 +550,7 @@ err:
 
 static void ctnl_timeout_put(struct ctnl_timeout *timeout)
 {
-	if (atomic_dec_and_test(&timeout->refcnt))
+	if (refcount_dec_and_test(&timeout->refcnt))
 		kfree_rcu(timeout, rcu_head);
 
 	module_put(THIS_MODULE);
@@ -601,7 +601,7 @@ static void __net_exit cttimeout_net_exit(struct net *net)
 		list_del_rcu(&cur->head);
 		nf_ct_l4proto_put(cur->l4proto);
 
-		if (atomic_dec_and_test(&cur->refcnt))
+		if (refcount_dec_and_test(&cur->refcnt))
 			kfree_rcu(cur, rcu_head);
 	}
 }
diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c
index 08247bf7d7b8..ecd857b75ffe 100644
--- a/net/netfilter/nfnetlink_log.c
+++ b/net/netfilter/nfnetlink_log.c
@@ -40,6 +40,8 @@
 #include <net/netfilter/nfnetlink_log.h>
 
 #include <linux/atomic.h>
+#include <linux/refcount.h>
+
 
 #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
 #include "../bridge/br_private.h"
@@ -57,7 +59,7 @@
 struct nfulnl_instance {
 	struct hlist_node hlist;	/* global list of instances */
 	spinlock_t lock;
-	atomic_t use;			/* use count */
+	refcount_t use;			/* use count */
 
 	unsigned int qlen;		/* number of nlmsgs in skb */
 	struct sk_buff *skb;		/* pre-allocatd skb */
@@ -115,7 +117,7 @@ __instance_lookup(struct nfnl_log_net *log, u_int16_t group_num)
 static inline void
 instance_get(struct nfulnl_instance *inst)
 {
-	atomic_inc(&inst->use);
+	refcount_inc(&inst->use);
 }
 
 static struct nfulnl_instance *
@@ -125,7 +127,7 @@ instance_lookup_get(struct nfnl_log_net *log, u_int16_t group_num)
 
 	rcu_read_lock_bh();
 	inst = __instance_lookup(log, group_num);
-	if (inst && !atomic_inc_not_zero(&inst->use))
+	if (inst && !refcount_inc_not_zero(&inst->use))
 		inst = NULL;
 	rcu_read_unlock_bh();
 
@@ -145,7 +147,7 @@ static void nfulnl_instance_free_rcu(struct rcu_head *head)
 static void
 instance_put(struct nfulnl_instance *inst)
 {
-	if (inst && atomic_dec_and_test(&inst->use))
+	if (inst && refcount_dec_and_test(&inst->use))
 		call_rcu_bh(&inst->rcu, nfulnl_instance_free_rcu);
 }
 
@@ -180,7 +182,7 @@ instance_create(struct net *net, u_int16_t group_num,
 	INIT_HLIST_NODE(&inst->hlist);
 	spin_lock_init(&inst->lock);
 	/* needs to be two, since we _put() after creation */
-	atomic_set(&inst->use, 2);
+	refcount_set(&inst->use, 2);
 
 	setup_timer(&inst->timer, nfulnl_timer, (unsigned long)inst);
 
@@ -1031,7 +1033,7 @@ static int seq_show(struct seq_file *s, void *v)
 		   inst->group_num,
 		   inst->peer_portid, inst->qlen,
 		   inst->copy_mode, inst->copy_range,
-		   inst->flushtimeout, atomic_read(&inst->use));
+		   inst->flushtimeout, refcount_read(&inst->use));
 
 	return 0;
 }

From 4485a841be171dbd8d3f0701b00f59d389e94ce6 Mon Sep 17 00:00:00 2001
From: "Reshetova, Elena" <elena.reshetova@intel.com>
Date: Mon, 20 Mar 2017 11:43:28 +0200
Subject: [PATCH 22/22] netfilter: fix the warning on unused refcount variable

net/netfilter/nfnetlink_acct.c: In function 'nfnl_acct_try_del':
net/netfilter/nfnetlink_acct.c:329:15: warning: unused variable 'refcount' [-Wunused-variable]
unsigned int refcount;
             ^

Fixes: b54ab92b84b6 ("netfilter: refcounter conversions")
Signed-off-by: Elena Reshetova <elena.reshetova@intel.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nfnetlink_acct.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/netfilter/nfnetlink_acct.c b/net/netfilter/nfnetlink_acct.c
index f44cbd35357f..c86da174a5fc 100644
--- a/net/netfilter/nfnetlink_acct.c
+++ b/net/netfilter/nfnetlink_acct.c
@@ -326,7 +326,6 @@ static int nfnl_acct_get(struct net *net, struct sock *nfnl,
 static int nfnl_acct_try_del(struct nf_acct *cur)
 {
 	int ret = 0;
-	unsigned int refcount;
 
 	/* We want to avoid races with nfnl_acct_put. So only when the current
 	 * refcnt is 1, we decrease it to 0.