From c4f3db15958277c03d1c324894255ea3ecbf86e1 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 11 Oct 2017 10:47:40 +0200
Subject: netfilter: conntrack: add and use nf_l4proto_log_invalid

We currently pass down the l4 protocol to the conntrack ->packet()
function, but the only user of this is the debug info decision.

Same information can be derived from struct nf_conn.
As a first step, add and use a new log function for this, similar to
nf_ct_helper_log().

Add __cold annotation -- invalid packets should be infrequent so
gcc can consider all call paths that lead to such a function as
unlikely.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_conntrack_l4proto.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

(limited to 'include/net/netfilter')

diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h
index 738a0307a96b..6d79a061d360 100644
--- a/include/net/netfilter/nf_conntrack_l4proto.h
+++ b/include/net/netfilter/nf_conntrack_l4proto.h
@@ -152,8 +152,18 @@ extern const struct nla_policy nf_ct_port_nla_policy[];
 #define LOG_INVALID(net, proto)				\
 	((net)->ct.sysctl_log_invalid == (proto) ||	\
 	 (net)->ct.sysctl_log_invalid == IPPROTO_RAW)
+
+__printf(5, 6) __cold
+void nf_l4proto_log_invalid(const struct sk_buff *skb,
+			    struct net *net,
+			    u16 pf, u8 protonum,
+			    const char *fmt, ...);
 #else
 static inline int LOG_INVALID(struct net *net, int proto) { return 0; }
+
+static inline __printf(5, 6) __cold
+void nf_l4proto_log_invalid(const struct sk_buff *skb, struct net *net,
+			    u16 pf, u8 protonum, const char *fmt, ...) {}
 #endif /* CONFIG_SYSCTL */
 
 #endif /*_NF_CONNTRACK_PROTOCOL_H*/
-- 
cgit v1.2.3-59-g8ed1b


From 3d0b527bc9dc0e8c4428eb1a98d4cd27bd1114c7 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 11 Oct 2017 10:47:41 +0200
Subject: netfilter: conntrack: add and use nf_ct_l4proto_log_invalid

We currently pass down the l4 protocol to the conntrack ->packet()
function, but the only user of this is the debug info decision.

Same information can be derived from struct nf_conn.
Add a wrapper for the previous patch that extracs the information
from nf_conn and passes it to nf_l4proto_log_invalid().

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_conntrack_l4proto.h | 14 ++++++++------
 net/netfilter/nf_conntrack_proto.c           | 23 +++++++++++++++++++++++
 net/netfilter/nf_conntrack_proto_dccp.c      | 17 +++++------------
 net/netfilter/nf_conntrack_proto_tcp.c       | 25 +++++++++----------------
 4 files changed, 45 insertions(+), 34 deletions(-)

(limited to 'include/net/netfilter')

diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h
index 6d79a061d360..5d51255b5bfb 100644
--- a/include/net/netfilter/nf_conntrack_l4proto.h
+++ b/include/net/netfilter/nf_conntrack_l4proto.h
@@ -149,21 +149,23 @@ int nf_ct_port_nlattr_tuple_size(void);
 extern const struct nla_policy nf_ct_port_nla_policy[];
 
 #ifdef CONFIG_SYSCTL
-#define LOG_INVALID(net, proto)				\
-	((net)->ct.sysctl_log_invalid == (proto) ||	\
-	 (net)->ct.sysctl_log_invalid == IPPROTO_RAW)
-
+__printf(3, 4) __cold
+void nf_ct_l4proto_log_invalid(const struct sk_buff *skb,
+			       const struct nf_conn *ct,
+			       const char *fmt, ...);
 __printf(5, 6) __cold
 void nf_l4proto_log_invalid(const struct sk_buff *skb,
 			    struct net *net,
 			    u16 pf, u8 protonum,
 			    const char *fmt, ...);
 #else
-static inline int LOG_INVALID(struct net *net, int proto) { return 0; }
-
 static inline __printf(5, 6) __cold
 void nf_l4proto_log_invalid(const struct sk_buff *skb, struct net *net,
 			    u16 pf, u8 protonum, const char *fmt, ...) {}
+static inline __printf(3, 4) __cold
+void nf_ct_l4proto_log_invalid(const struct sk_buff *skb,
+			       const struct nf_conn *ct,
+			       const char *fmt, ...) { }
 #endif /* CONFIG_SYSCTL */
 
 #endif /*_NF_CONNTRACK_PROTOCOL_H*/
diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c
index bcd3ee270d75..83f739e9dc08 100644
--- a/net/netfilter/nf_conntrack_proto.c
+++ b/net/netfilter/nf_conntrack_proto.c
@@ -87,6 +87,29 @@ void nf_l4proto_log_invalid(const struct sk_buff *skb,
 	va_end(args);
 }
 EXPORT_SYMBOL_GPL(nf_l4proto_log_invalid);
+
+__printf(3, 4)
+void nf_ct_l4proto_log_invalid(const struct sk_buff *skb,
+			       const struct nf_conn *ct,
+			       const char *fmt, ...)
+{
+	struct va_format vaf;
+	struct net *net;
+	va_list args;
+
+	net = nf_ct_net(ct);
+	if (likely(net->ct.sysctl_log_invalid == 0))
+		return;
+
+	va_start(args, fmt);
+	vaf.fmt = fmt;
+	vaf.va = &args;
+
+	nf_l4proto_log_invalid(skb, net, nf_ct_l3num(ct),
+			       nf_ct_protonum(ct), "%pV", &vaf);
+	va_end(args);
+}
+EXPORT_SYMBOL_GPL(nf_ct_l4proto_log_invalid);
 #endif
 
 const struct nf_conntrack_l4proto *
diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c
index ef501c7edb96..49e0abcdc6f4 100644
--- a/net/netfilter/nf_conntrack_proto_dccp.c
+++ b/net/netfilter/nf_conntrack_proto_dccp.c
@@ -428,13 +428,13 @@ static bool dccp_new(struct nf_conn *ct, const struct sk_buff *skb,
 	default:
 		dn = dccp_pernet(net);
 		if (dn->dccp_loose == 0) {
-			msg = "nf_ct_dccp: not picking up existing connection ";
+			msg = "not picking up existing connection ";
 			goto out_invalid;
 		}
 	case CT_DCCP_REQUEST:
 		break;
 	case CT_DCCP_INVALID:
-		msg = "nf_ct_dccp: invalid state transition ";
+		msg = "invalid state transition ";
 		goto out_invalid;
 	}
 
@@ -447,9 +447,7 @@ static bool dccp_new(struct nf_conn *ct, const struct sk_buff *skb,
 	return true;
 
 out_invalid:
-	if (LOG_INVALID(net, IPPROTO_DCCP))
-		nf_log_packet(net, nf_ct_l3num(ct), 0, skb, NULL, NULL,
-			      NULL, "%s", msg);
+	nf_ct_l4proto_log_invalid(skb, ct, "%s", msg);
 	return false;
 }
 
@@ -472,7 +470,6 @@ static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb,
 		       u_int8_t pf,
 		       unsigned int *timeouts)
 {
-	struct net *net = nf_ct_net(ct);
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
 	struct dccp_hdr _dh, *dh;
 	u_int8_t type, old_state, new_state;
@@ -534,15 +531,11 @@ static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb,
 		ct->proto.dccp.last_pkt = type;
 
 		spin_unlock_bh(&ct->lock);
-		if (LOG_INVALID(net, IPPROTO_DCCP))
-			nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL,
-				      "nf_ct_dccp: invalid packet ignored ");
+		nf_ct_l4proto_log_invalid(skb, ct, "%s", "invalid packet");
 		return NF_ACCEPT;
 	case CT_DCCP_INVALID:
 		spin_unlock_bh(&ct->lock);
-		if (LOG_INVALID(net, IPPROTO_DCCP))
-			nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL,
-				      "nf_ct_dccp: invalid state transition ");
+		nf_ct_l4proto_log_invalid(skb, ct, "%s", "invalid state transition");
 		return -NF_ACCEPT;
 	}
 
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index 14198b2a2e2c..dced574f6006 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -702,9 +702,9 @@ static bool tcp_in_window(const struct nf_conn *ct,
 		if (sender->flags & IP_CT_TCP_FLAG_BE_LIBERAL ||
 		    tn->tcp_be_liberal)
 			res = true;
-		if (!res && LOG_INVALID(net, IPPROTO_TCP))
-			nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL,
-			"nf_ct_tcp: %s ",
+		if (!res) {
+			nf_ct_l4proto_log_invalid(skb, ct,
+			"%s",
 			before(seq, sender->td_maxend + 1) ?
 			in_recv_win ?
 			before(sack, receiver->td_end + 1) ?
@@ -713,6 +713,7 @@ static bool tcp_in_window(const struct nf_conn *ct,
 			: "ACK is over the upper bound (ACKed data not seen yet)"
 			: "SEQ is under the lower bound (already ACKed data retransmitted)"
 			: "SEQ is over the upper bound (over the window of the receiver)");
+		}
 	}
 
 	pr_debug("tcp_in_window: res=%u sender end=%u maxend=%u maxwin=%u "
@@ -937,10 +938,8 @@ static int tcp_packet(struct nf_conn *ct,
 					IP_CT_EXP_CHALLENGE_ACK;
 		}
 		spin_unlock_bh(&ct->lock);
-		if (LOG_INVALID(net, IPPROTO_TCP))
-			nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL,
-				  "nf_ct_tcp: invalid packet ignored in "
-				  "state %s ", tcp_conntrack_names[old_state]);
+		nf_ct_l4proto_log_invalid(skb, ct, "invalid packet ignored in "
+					  "state %s ", tcp_conntrack_names[old_state]);
 		return NF_ACCEPT;
 	case TCP_CONNTRACK_MAX:
 		/* Special case for SYN proxy: when the SYN to the server or
@@ -962,9 +961,7 @@ static int tcp_packet(struct nf_conn *ct,
 		pr_debug("nf_ct_tcp: Invalid dir=%i index=%u ostate=%u\n",
 			 dir, get_conntrack_index(th), old_state);
 		spin_unlock_bh(&ct->lock);
-		if (LOG_INVALID(net, IPPROTO_TCP))
-			nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL,
-				  "nf_ct_tcp: invalid state ");
+		nf_ct_l4proto_log_invalid(skb, ct, "invalid state");
 		return -NF_ACCEPT;
 	case TCP_CONNTRACK_TIME_WAIT:
 		/* RFC5961 compliance cause stack to send "challenge-ACK"
@@ -979,9 +976,7 @@ static int tcp_packet(struct nf_conn *ct,
 			/* Detected RFC5961 challenge ACK */
 			ct->proto.tcp.last_flags &= ~IP_CT_EXP_CHALLENGE_ACK;
 			spin_unlock_bh(&ct->lock);
-			if (LOG_INVALID(net, IPPROTO_TCP))
-				nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL,
-				      "nf_ct_tcp: challenge-ACK ignored ");
+			nf_ct_l4proto_log_invalid(skb, ct, "challenge-ack ignored");
 			return NF_ACCEPT; /* Don't change state */
 		}
 		break;
@@ -991,9 +986,7 @@ static int tcp_packet(struct nf_conn *ct,
 		    && before(ntohl(th->seq), ct->proto.tcp.seen[!dir].td_maxack)) {
 			/* Invalid RST  */
 			spin_unlock_bh(&ct->lock);
-			if (LOG_INVALID(net, IPPROTO_TCP))
-				nf_log_packet(net, pf, 0, skb, NULL, NULL,
-					      NULL, "nf_ct_tcp: invalid RST ");
+			nf_ct_l4proto_log_invalid(skb, ct, "invalid rst");
 			return -NF_ACCEPT;
 		}
 		if (index == TCP_RST_SET
-- 
cgit v1.2.3-59-g8ed1b


From eb6fad5a4a328b85d3faa8b301b522e3f316b49d Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 11 Oct 2017 10:47:42 +0200
Subject: netfilter: conntrack: remove pf argument from l4 packet functions

not needed/used anymore.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_conntrack_l4proto.h   | 1 -
 net/ipv4/netfilter/nf_conntrack_proto_icmp.c   | 1 -
 net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 1 -
 net/netfilter/nf_conntrack_core.c              | 2 +-
 net/netfilter/nf_conntrack_proto_dccp.c        | 1 -
 net/netfilter/nf_conntrack_proto_generic.c     | 1 -
 net/netfilter/nf_conntrack_proto_gre.c         | 1 -
 net/netfilter/nf_conntrack_proto_sctp.c        | 1 -
 net/netfilter/nf_conntrack_proto_tcp.c         | 6 ++----
 net/netfilter/nf_conntrack_proto_udp.c         | 1 -
 10 files changed, 3 insertions(+), 13 deletions(-)

(limited to 'include/net/netfilter')

diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h
index 5d51255b5bfb..e06518874144 100644
--- a/include/net/netfilter/nf_conntrack_l4proto.h
+++ b/include/net/netfilter/nf_conntrack_l4proto.h
@@ -42,7 +42,6 @@ struct nf_conntrack_l4proto {
 		      const struct sk_buff *skb,
 		      unsigned int dataoff,
 		      enum ip_conntrack_info ctinfo,
-		      u_int8_t pf,
 		      unsigned int *timeouts);
 
 	/* Called when a new connection for this protocol found;
diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
index 7281a7b77a0e..8969420cecc3 100644
--- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
+++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
@@ -81,7 +81,6 @@ static int icmp_packet(struct nf_conn *ct,
 		       const struct sk_buff *skb,
 		       unsigned int dataoff,
 		       enum ip_conntrack_info ctinfo,
-		       u_int8_t pf,
 		       unsigned int *timeout)
 {
 	/* Do not immediately delete the connection after the first
diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
index 0f227ca4a5a2..dca921df28e1 100644
--- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
@@ -94,7 +94,6 @@ static int icmpv6_packet(struct nf_conn *ct,
 		       const struct sk_buff *skb,
 		       unsigned int dataoff,
 		       enum ip_conntrack_info ctinfo,
-		       u_int8_t pf,
 		       unsigned int *timeout)
 {
 	/* Do not immediately delete the connection after the first
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 01130392b7c0..28e675150853 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -1419,7 +1419,7 @@ repeat:
 	/* Decide what timeout policy we want to apply to this flow. */
 	timeouts = nf_ct_timeout_lookup(net, ct, l4proto);
 
-	ret = l4proto->packet(ct, skb, dataoff, ctinfo, pf, timeouts);
+	ret = l4proto->packet(ct, skb, dataoff, ctinfo, timeouts);
 	if (ret <= 0) {
 		/* Invalid: inverse of the return code tells
 		 * the netfilter core what to do */
diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c
index 49e0abcdc6f4..2a446f4a554c 100644
--- a/net/netfilter/nf_conntrack_proto_dccp.c
+++ b/net/netfilter/nf_conntrack_proto_dccp.c
@@ -467,7 +467,6 @@ static unsigned int *dccp_get_timeouts(struct net *net)
 
 static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb,
 		       unsigned int dataoff, enum ip_conntrack_info ctinfo,
-		       u_int8_t pf,
 		       unsigned int *timeouts)
 {
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
diff --git a/net/netfilter/nf_conntrack_proto_generic.c b/net/netfilter/nf_conntrack_proto_generic.c
index 9cd40700842e..1f86ddf6649a 100644
--- a/net/netfilter/nf_conntrack_proto_generic.c
+++ b/net/netfilter/nf_conntrack_proto_generic.c
@@ -60,7 +60,6 @@ static int generic_packet(struct nf_conn *ct,
 			  const struct sk_buff *skb,
 			  unsigned int dataoff,
 			  enum ip_conntrack_info ctinfo,
-			  u_int8_t pf,
 			  unsigned int *timeout)
 {
 	nf_ct_refresh_acct(ct, ctinfo, skb, *timeout);
diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c
index 09a90484c27d..a2503005d80b 100644
--- a/net/netfilter/nf_conntrack_proto_gre.c
+++ b/net/netfilter/nf_conntrack_proto_gre.c
@@ -244,7 +244,6 @@ static int gre_packet(struct nf_conn *ct,
 		      const struct sk_buff *skb,
 		      unsigned int dataoff,
 		      enum ip_conntrack_info ctinfo,
-		      u_int8_t pf,
 		      unsigned int *timeouts)
 {
 	/* If we've seen traffic both ways, this is a GRE connection.
diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c
index aa630c561361..80faf04ddf15 100644
--- a/net/netfilter/nf_conntrack_proto_sctp.c
+++ b/net/netfilter/nf_conntrack_proto_sctp.c
@@ -306,7 +306,6 @@ static int sctp_packet(struct nf_conn *ct,
 		       const struct sk_buff *skb,
 		       unsigned int dataoff,
 		       enum ip_conntrack_info ctinfo,
-		       u_int8_t pf,
 		       unsigned int *timeouts)
 {
 	enum sctp_conntrack new_state, old_state;
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index dced574f6006..8f283294d70f 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -493,8 +493,7 @@ static bool tcp_in_window(const struct nf_conn *ct,
 			  unsigned int index,
 			  const struct sk_buff *skb,
 			  unsigned int dataoff,
-			  const struct tcphdr *tcph,
-			  u_int8_t pf)
+			  const struct tcphdr *tcph)
 {
 	struct net *net = nf_ct_net(ct);
 	struct nf_tcp_net *tn = tcp_pernet(net);
@@ -801,7 +800,6 @@ static int tcp_packet(struct nf_conn *ct,
 		      const struct sk_buff *skb,
 		      unsigned int dataoff,
 		      enum ip_conntrack_info ctinfo,
-		      u_int8_t pf,
 		      unsigned int *timeouts)
 {
 	struct net *net = nf_ct_net(ct);
@@ -1013,7 +1011,7 @@ static int tcp_packet(struct nf_conn *ct,
 	}
 
 	if (!tcp_in_window(ct, &ct->proto.tcp, dir, index,
-			   skb, dataoff, th, pf)) {
+			   skb, dataoff, th)) {
 		spin_unlock_bh(&ct->lock);
 		return -NF_ACCEPT;
 	}
diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c
index fc20cf430251..3a5f727103af 100644
--- a/net/netfilter/nf_conntrack_proto_udp.c
+++ b/net/netfilter/nf_conntrack_proto_udp.c
@@ -73,7 +73,6 @@ static int udp_packet(struct nf_conn *ct,
 		      const struct sk_buff *skb,
 		      unsigned int dataoff,
 		      enum ip_conntrack_info ctinfo,
-		      u_int8_t pf,
 		      unsigned int *timeouts)
 {
 	/* If we've seen traffic both ways, this is some kind of UDP
-- 
cgit v1.2.3-59-g8ed1b


From 28efb0046512e8a13ed9f9bdf0d68d10bbfbe9cf Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 12 Oct 2017 09:38:30 +0200
Subject: netfilter: conntrack: make l3proto trackers const

previous patches removed all writes to them.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/ipv4/nf_conntrack_ipv4.h | 2 +-
 include/net/netfilter/ipv6/nf_conntrack_ipv6.h | 2 +-
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 2 +-
 net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/net/netfilter')

diff --git a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h
index 919e4e8af327..5534ecca7a5d 100644
--- a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h
+++ b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h
@@ -10,7 +10,7 @@
 #define _NF_CONNTRACK_IPV4_H
 
 
-extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4;
+const extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4;
 
 extern struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp4;
 extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp4;
diff --git a/include/net/netfilter/ipv6/nf_conntrack_ipv6.h b/include/net/netfilter/ipv6/nf_conntrack_ipv6.h
index eaea968f8657..30dc57980866 100644
--- a/include/net/netfilter/ipv6/nf_conntrack_ipv6.h
+++ b/include/net/netfilter/ipv6/nf_conntrack_ipv6.h
@@ -1,7 +1,7 @@
 #ifndef _NF_CONNTRACK_IPV6_H
 #define _NF_CONNTRACK_IPV6_H
 
-extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6;
+extern const struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6;
 
 extern struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp6;
 extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp6;
diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
index fe374da4bc13..89af9d88ca21 100644
--- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
+++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
@@ -344,7 +344,7 @@ static void ipv4_hooks_unregister(struct net *net)
 	mutex_unlock(&register_ipv4_hooks);
 }
 
-struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 __read_mostly = {
+const struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 = {
 	.l3proto	 = PF_INET,
 	.pkt_to_tuple	 = ipv4_pkt_to_tuple,
 	.invert_tuple	 = ipv4_invert_tuple,
diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
index fe01dc953c56..3b80a38f62b8 100644
--- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
@@ -339,7 +339,7 @@ static void ipv6_hooks_unregister(struct net *net)
 	mutex_unlock(&register_ipv6_hooks);
 }
 
-struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6 __read_mostly = {
+const struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6 = {
 	.l3proto		= PF_INET6,
 	.pkt_to_tuple		= ipv6_pkt_to_tuple,
 	.invert_tuple		= ipv6_invert_tuple,
-- 
cgit v1.2.3-59-g8ed1b


From 5caaed151a68ae36aca2981cc245f5960a0a7603 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 2 Nov 2017 19:41:09 +0100
Subject: netfilter: conntrack: don't cache nlattr_tuple_size result in
 nla_size

We currently call ->nlattr_tuple_size() once at register time and
cache result in l4proto->nla_size.

nla_size is the only member that is written to, avoiding this would
allow to make l4proto trackers const.

We can use ->nlattr_tuple_size() at run time, and cache result in
the individual trackers instead.

This is an intermediate step, next patch removes nlattr_size()
callback and computes size at compile time, then removes nla_size.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_conntrack_l4proto.h   |  4 ++--
 net/ipv4/netfilter/nf_conntrack_proto_icmp.c   |  9 +++++++--
 net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c |  9 +++++++--
 net/netfilter/nf_conntrack_core.c              |  9 +++++++--
 net/netfilter/nf_conntrack_netlink.c           | 10 +++++++---
 net/netfilter/nf_conntrack_proto.c             |  2 --
 net/netfilter/nf_conntrack_proto_tcp.c         |  9 +++++++--
 7 files changed, 37 insertions(+), 15 deletions(-)

(limited to 'include/net/netfilter')

diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h
index e06518874144..46e786ffcf2f 100644
--- a/include/net/netfilter/nf_conntrack_l4proto.h
+++ b/include/net/netfilter/nf_conntrack_l4proto.h
@@ -74,7 +74,7 @@ struct nf_conntrack_l4proto {
 	int (*tuple_to_nlattr)(struct sk_buff *skb,
 			       const struct nf_conntrack_tuple *t);
 	/* Calculate tuple nlattr size */
-	int (*nlattr_tuple_size)(void);
+	unsigned int (*nlattr_tuple_size)(void);
 	int (*nlattr_to_tuple)(struct nlattr *tb[],
 			       struct nf_conntrack_tuple *t);
 	const struct nla_policy *nla_policy;
@@ -144,7 +144,7 @@ int nf_ct_port_tuple_to_nlattr(struct sk_buff *skb,
 			       const struct nf_conntrack_tuple *tuple);
 int nf_ct_port_nlattr_to_tuple(struct nlattr *tb[],
 			       struct nf_conntrack_tuple *t);
-int nf_ct_port_nlattr_tuple_size(void);
+unsigned int nf_ct_port_nlattr_tuple_size(void);
 extern const struct nla_policy nf_ct_port_nla_policy[];
 
 #ifdef CONFIG_SYSCTL
diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
index 8969420cecc3..1849fedd9b81 100644
--- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
+++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c
@@ -258,9 +258,14 @@ static int icmp_nlattr_to_tuple(struct nlattr *tb[],
 	return 0;
 }
 
-static int icmp_nlattr_tuple_size(void)
+static unsigned int icmp_nlattr_tuple_size(void)
 {
-	return nla_policy_len(icmp_nla_policy, CTA_PROTO_MAX + 1);
+	static unsigned int size __read_mostly;
+
+	if (!size)
+		size = nla_policy_len(icmp_nla_policy, CTA_PROTO_MAX + 1);
+
+	return size;
 }
 #endif
 
diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
index dca921df28e1..3ac0d826afc4 100644
--- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
@@ -259,9 +259,14 @@ static int icmpv6_nlattr_to_tuple(struct nlattr *tb[],
 	return 0;
 }
 
-static int icmpv6_nlattr_tuple_size(void)
+static unsigned int icmpv6_nlattr_tuple_size(void)
 {
-	return nla_policy_len(icmpv6_nla_policy, CTA_PROTO_MAX + 1);
+	static unsigned int size __read_mostly;
+
+	if (!size)
+		size = nla_policy_len(icmpv6_nla_policy, CTA_PROTO_MAX + 1);
+
+	return size;
 }
 #endif
 
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 28e675150853..0e516947c16f 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -1563,9 +1563,14 @@ int nf_ct_port_nlattr_to_tuple(struct nlattr *tb[],
 }
 EXPORT_SYMBOL_GPL(nf_ct_port_nlattr_to_tuple);
 
-int nf_ct_port_nlattr_tuple_size(void)
+unsigned int nf_ct_port_nlattr_tuple_size(void)
 {
-	return nla_policy_len(nf_ct_port_nla_policy, CTA_PROTO_MAX + 1);
+	static unsigned int size __read_mostly;
+
+	if (!size)
+		size = nla_policy_len(nf_ct_port_nla_policy, CTA_PROTO_MAX + 1);
+
+	return size;
 }
 EXPORT_SYMBOL_GPL(nf_ct_port_nlattr_tuple_size);
 #endif
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index de4053d84364..6e0adfefb9ed 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -533,11 +533,11 @@ nla_put_failure:
 	return -1;
 }
 
-static inline size_t ctnetlink_proto_size(const struct nf_conn *ct)
+static size_t ctnetlink_proto_size(const struct nf_conn *ct)
 {
 	const struct nf_conntrack_l3proto *l3proto;
 	const struct nf_conntrack_l4proto *l4proto;
-	size_t len;
+	size_t len, len4 = 0;
 
 	l3proto = __nf_ct_l3proto_find(nf_ct_l3num(ct));
 	len = l3proto->nla_size;
@@ -545,8 +545,12 @@ static inline size_t ctnetlink_proto_size(const struct nf_conn *ct)
 
 	l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct));
 	len += l4proto->nla_size;
+	if (l4proto->nlattr_tuple_size) {
+		len4 = l4proto->nlattr_tuple_size();
+		len4 *= 3u; /* ORIG, REPLY, MASTER */
+	}
 
-	return len;
+	return len + len4;
 }
 
 static inline size_t ctnetlink_acct_size(const struct nf_conn *ct)
diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c
index 83f739e9dc08..3b06ff3f2dee 100644
--- a/net/netfilter/nf_conntrack_proto.c
+++ b/net/netfilter/nf_conntrack_proto.c
@@ -398,8 +398,6 @@ int nf_ct_l4proto_register_one(struct nf_conntrack_l4proto *l4proto)
 	l4proto->nla_size = 0;
 	if (l4proto->nlattr_size)
 		l4proto->nla_size += l4proto->nlattr_size();
-	if (l4proto->nlattr_tuple_size)
-		l4proto->nla_size += 3 * l4proto->nlattr_tuple_size();
 
 	rcu_assign_pointer(nf_ct_protos[l4proto->l3proto][l4proto->l4proto],
 			   l4proto);
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index 8f283294d70f..b12fc07111d0 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -1277,9 +1277,14 @@ static int tcp_nlattr_size(void)
 		+ nla_policy_len(tcp_nla_policy, CTA_PROTOINFO_TCP_MAX + 1);
 }
 
-static int tcp_nlattr_tuple_size(void)
+static unsigned int tcp_nlattr_tuple_size(void)
 {
-	return nla_policy_len(nf_ct_port_nla_policy, CTA_PROTO_MAX + 1);
+	static unsigned int size __read_mostly;
+
+	if (!size)
+		size = nla_policy_len(nf_ct_port_nla_policy, CTA_PROTO_MAX + 1);
+
+	return size;
 }
 #endif
 
-- 
cgit v1.2.3-59-g8ed1b


From ba0e4d9917b43dfa746cbbcb4477da59aae73bd6 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Mon, 9 Oct 2017 19:52:28 +0200
Subject: netfilter: nf_tables: get set elements via netlink

This patch adds a new get operation to look up for specific elements in
a set via netlink interface. You can also use it to check if an interval
already exists.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h |   5 ++
 net/netfilter/nf_tables_api.c     | 184 ++++++++++++++++++++++++++------------
 net/netfilter/nft_set_bitmap.c    |  18 ++++
 net/netfilter/nft_set_hash.c      |  39 ++++++++
 net/netfilter/nft_set_rbtree.c    |  73 +++++++++++++++
 5 files changed, 264 insertions(+), 55 deletions(-)

(limited to 'include/net/netfilter')

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index 0f5b12a4ad09..d011e56cc7a9 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -311,6 +311,7 @@ struct nft_expr;
  *	@flush: deactivate element in the next generation
  *	@remove: remove element from set
  *	@walk: iterate over all set elemeennts
+ *	@get: get set elements
  *	@privsize: function to return size of set private data
  *	@init: initialize private data of new set instance
  *	@destroy: destroy private data of set instance
@@ -350,6 +351,10 @@ struct nft_set_ops {
 	void				(*walk)(const struct nft_ctx *ctx,
 						struct nft_set *set,
 						struct nft_set_iter *iter);
+	void *				(*get)(const struct net *net,
+					       const struct nft_set *set,
+					       const struct nft_set_elem *elem,
+					       unsigned int flags);
 
 	unsigned int			(*privsize)(const struct nlattr * const nla[],
 						    const struct nft_set_desc *desc);
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 3b4a0739ee39..1d66be0d8ef7 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -3586,45 +3586,6 @@ static int nf_tables_dump_set_done(struct netlink_callback *cb)
 	return 0;
 }
 
-static int nf_tables_getsetelem(struct net *net, struct sock *nlsk,
-				struct sk_buff *skb, const struct nlmsghdr *nlh,
-				const struct nlattr * const nla[],
-				struct netlink_ext_ack *extack)
-{
-	u8 genmask = nft_genmask_cur(net);
-	const struct nft_set *set;
-	struct nft_ctx ctx;
-	int err;
-
-	err = nft_ctx_init_from_elemattr(&ctx, net, skb, nlh, nla, genmask);
-	if (err < 0)
-		return err;
-
-	set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET],
-				   genmask);
-	if (IS_ERR(set))
-		return PTR_ERR(set);
-
-	if (nlh->nlmsg_flags & NLM_F_DUMP) {
-		struct netlink_dump_control c = {
-			.dump = nf_tables_dump_set,
-			.done = nf_tables_dump_set_done,
-		};
-		struct nft_set_dump_ctx *dump_ctx;
-
-		dump_ctx = kmalloc(sizeof(*dump_ctx), GFP_KERNEL);
-		if (!dump_ctx)
-			return -ENOMEM;
-
-		dump_ctx->set = set;
-		dump_ctx->ctx = ctx;
-
-		c.data = dump_ctx;
-		return netlink_dump_start(nlsk, skb, nlh, &c);
-	}
-	return -EOPNOTSUPP;
-}
-
 static int nf_tables_fill_setelem_info(struct sk_buff *skb,
 				       const struct nft_ctx *ctx, u32 seq,
 				       u32 portid, int event, u16 flags,
@@ -3670,6 +3631,135 @@ nla_put_failure:
 	return -1;
 }
 
+static int nft_setelem_parse_flags(const struct nft_set *set,
+				   const struct nlattr *attr, u32 *flags)
+{
+	if (attr == NULL)
+		return 0;
+
+	*flags = ntohl(nla_get_be32(attr));
+	if (*flags & ~NFT_SET_ELEM_INTERVAL_END)
+		return -EINVAL;
+	if (!(set->flags & NFT_SET_INTERVAL) &&
+	    *flags & NFT_SET_ELEM_INTERVAL_END)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int nft_get_set_elem(struct nft_ctx *ctx, struct nft_set *set,
+			    const struct nlattr *attr)
+{
+	struct nlattr *nla[NFTA_SET_ELEM_MAX + 1];
+	const struct nft_set_ext *ext;
+	struct nft_data_desc desc;
+	struct nft_set_elem elem;
+	struct sk_buff *skb;
+	uint32_t flags = 0;
+	void *priv;
+	int err;
+
+	err = nla_parse_nested(nla, NFTA_SET_ELEM_MAX, attr,
+			       nft_set_elem_policy, NULL);
+	if (err < 0)
+		return err;
+
+	if (!nla[NFTA_SET_ELEM_KEY])
+		return -EINVAL;
+
+	err = nft_setelem_parse_flags(set, nla[NFTA_SET_ELEM_FLAGS], &flags);
+	if (err < 0)
+		return err;
+
+	err = nft_data_init(ctx, &elem.key.val, sizeof(elem.key), &desc,
+			    nla[NFTA_SET_ELEM_KEY]);
+	if (err < 0)
+		return err;
+
+	err = -EINVAL;
+	if (desc.type != NFT_DATA_VALUE || desc.len != set->klen)
+		return err;
+
+	priv = set->ops->get(ctx->net, set, &elem, flags);
+	if (IS_ERR(priv))
+		return PTR_ERR(priv);
+
+	elem.priv = priv;
+	ext = nft_set_elem_ext(set, &elem);
+
+	err = -ENOMEM;
+	skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
+	if (skb == NULL)
+		goto err1;
+
+	err = nf_tables_fill_setelem_info(skb, ctx, ctx->seq, ctx->portid,
+					  NFT_MSG_NEWSETELEM, 0, set, &elem);
+	if (err < 0)
+		goto err2;
+
+	err = nfnetlink_unicast(skb, ctx->net, ctx->portid, MSG_DONTWAIT);
+	/* This avoids a loop in nfnetlink. */
+	if (err < 0)
+		goto err1;
+
+	return 0;
+err2:
+	kfree_skb(skb);
+err1:
+	/* this avoids a loop in nfnetlink. */
+	return err == -EAGAIN ? -ENOBUFS : err;
+}
+
+static int nf_tables_getsetelem(struct net *net, struct sock *nlsk,
+				struct sk_buff *skb, const struct nlmsghdr *nlh,
+				const struct nlattr * const nla[],
+				struct netlink_ext_ack *extack)
+{
+	u8 genmask = nft_genmask_cur(net);
+	struct nft_set *set;
+	struct nlattr *attr;
+	struct nft_ctx ctx;
+	int rem, err = 0;
+
+	err = nft_ctx_init_from_elemattr(&ctx, net, skb, nlh, nla, genmask);
+	if (err < 0)
+		return err;
+
+	set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET],
+				   genmask);
+	if (IS_ERR(set))
+		return PTR_ERR(set);
+
+	if (nlh->nlmsg_flags & NLM_F_DUMP) {
+		struct netlink_dump_control c = {
+			.dump = nf_tables_dump_set,
+			.done = nf_tables_dump_set_done,
+		};
+		struct nft_set_dump_ctx *dump_ctx;
+
+		dump_ctx = kmalloc(sizeof(*dump_ctx), GFP_KERNEL);
+		if (!dump_ctx)
+			return -ENOMEM;
+
+		dump_ctx->set = set;
+		dump_ctx->ctx = ctx;
+
+		c.data = dump_ctx;
+		return netlink_dump_start(nlsk, skb, nlh, &c);
+	}
+
+	if (!nla[NFTA_SET_ELEM_LIST_ELEMENTS])
+		return -EINVAL;
+
+	nla_for_each_nested(attr, nla[NFTA_SET_ELEM_LIST_ELEMENTS], rem) {
+		err = nft_get_set_elem(&ctx, set, attr);
+		if (err < 0)
+			break;
+	}
+
+	return err;
+}
+
 static void nf_tables_setelem_notify(const struct nft_ctx *ctx,
 				     const struct nft_set *set,
 				     const struct nft_set_elem *elem,
@@ -3770,22 +3860,6 @@ static void nf_tables_set_elem_destroy(const struct nft_set *set, void *elem)
 	kfree(elem);
 }
 
-static int nft_setelem_parse_flags(const struct nft_set *set,
-				   const struct nlattr *attr, u32 *flags)
-{
-	if (attr == NULL)
-		return 0;
-
-	*flags = ntohl(nla_get_be32(attr));
-	if (*flags & ~NFT_SET_ELEM_INTERVAL_END)
-		return -EINVAL;
-	if (!(set->flags & NFT_SET_INTERVAL) &&
-	    *flags & NFT_SET_ELEM_INTERVAL_END)
-		return -EINVAL;
-
-	return 0;
-}
-
 static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 			    const struct nlattr *attr, u32 nlmsg_flags)
 {
diff --git a/net/netfilter/nft_set_bitmap.c b/net/netfilter/nft_set_bitmap.c
index 734989c40579..45fb2752fb63 100644
--- a/net/netfilter/nft_set_bitmap.c
+++ b/net/netfilter/nft_set_bitmap.c
@@ -106,6 +106,23 @@ nft_bitmap_elem_find(const struct nft_set *set, struct nft_bitmap_elem *this,
 	return NULL;
 }
 
+static void *nft_bitmap_get(const struct net *net, const struct nft_set *set,
+			    const struct nft_set_elem *elem, unsigned int flags)
+{
+	const struct nft_bitmap *priv = nft_set_priv(set);
+	u8 genmask = nft_genmask_cur(net);
+	struct nft_bitmap_elem *be;
+
+	list_for_each_entry_rcu(be, &priv->list, head) {
+		if (memcmp(nft_set_ext_key(&be->ext), elem->key.val.data, set->klen) ||
+		    !nft_set_elem_active(&be->ext, genmask))
+			continue;
+
+		return be;
+	}
+	return ERR_PTR(-ENOENT);
+}
+
 static int nft_bitmap_insert(const struct net *net, const struct nft_set *set,
 			     const struct nft_set_elem *elem,
 			     struct nft_set_ext **ext)
@@ -294,6 +311,7 @@ static struct nft_set_ops nft_bitmap_ops __read_mostly = {
 	.activate	= nft_bitmap_activate,
 	.lookup		= nft_bitmap_lookup,
 	.walk		= nft_bitmap_walk,
+	.get		= nft_bitmap_get,
 };
 
 static struct nft_set_type nft_bitmap_type __read_mostly = {
diff --git a/net/netfilter/nft_set_hash.c b/net/netfilter/nft_set_hash.c
index 650677f1e539..c68a7e0fcf1e 100644
--- a/net/netfilter/nft_set_hash.c
+++ b/net/netfilter/nft_set_hash.c
@@ -95,6 +95,24 @@ static bool nft_rhash_lookup(const struct net *net, const struct nft_set *set,
 	return !!he;
 }
 
+static void *nft_rhash_get(const struct net *net, const struct nft_set *set,
+			   const struct nft_set_elem *elem, unsigned int flags)
+{
+	struct nft_rhash *priv = nft_set_priv(set);
+	struct nft_rhash_elem *he;
+	struct nft_rhash_cmp_arg arg = {
+		.genmask = nft_genmask_cur(net),
+		.set	 = set,
+		.key	 = elem->key.val.data,
+	};
+
+	he = rhashtable_lookup_fast(&priv->ht, &arg, nft_rhash_params);
+	if (he != NULL)
+		return he;
+
+	return ERR_PTR(-ENOENT);
+}
+
 static bool nft_rhash_update(struct nft_set *set, const u32 *key,
 			     void *(*new)(struct nft_set *,
 					  const struct nft_expr *,
@@ -409,6 +427,24 @@ static bool nft_hash_lookup(const struct net *net, const struct nft_set *set,
 	return false;
 }
 
+static void *nft_hash_get(const struct net *net, const struct nft_set *set,
+			  const struct nft_set_elem *elem, unsigned int flags)
+{
+	struct nft_hash *priv = nft_set_priv(set);
+	u8 genmask = nft_genmask_cur(net);
+	struct nft_hash_elem *he;
+	u32 hash;
+
+	hash = jhash(elem->key.val.data, set->klen, priv->seed);
+	hash = reciprocal_scale(hash, priv->buckets);
+	hlist_for_each_entry_rcu(he, &priv->table[hash], node) {
+		if (!memcmp(nft_set_ext_key(&he->ext), elem->key.val.data, set->klen) &&
+		    nft_set_elem_active(&he->ext, genmask))
+			return he;
+	}
+	return ERR_PTR(-ENOENT);
+}
+
 /* nft_hash_select_ops() makes sure key size can be either 2 or 4 bytes . */
 static inline u32 nft_hash_key(const u32 *key, u32 klen)
 {
@@ -600,6 +636,7 @@ static struct nft_set_ops nft_rhash_ops __read_mostly = {
 	.lookup		= nft_rhash_lookup,
 	.update		= nft_rhash_update,
 	.walk		= nft_rhash_walk,
+	.get		= nft_rhash_get,
 	.features	= NFT_SET_MAP | NFT_SET_OBJECT | NFT_SET_TIMEOUT,
 };
 
@@ -617,6 +654,7 @@ static struct nft_set_ops nft_hash_ops __read_mostly = {
 	.remove		= nft_hash_remove,
 	.lookup		= nft_hash_lookup,
 	.walk		= nft_hash_walk,
+	.get		= nft_hash_get,
 	.features	= NFT_SET_MAP | NFT_SET_OBJECT,
 };
 
@@ -634,6 +672,7 @@ static struct nft_set_ops nft_hash_fast_ops __read_mostly = {
 	.remove		= nft_hash_remove,
 	.lookup		= nft_hash_lookup_fast,
 	.walk		= nft_hash_walk,
+	.get		= nft_hash_get,
 	.features	= NFT_SET_MAP | NFT_SET_OBJECT,
 };
 
diff --git a/net/netfilter/nft_set_rbtree.c b/net/netfilter/nft_set_rbtree.c
index d83a4ec5900d..e6f08bc5f359 100644
--- a/net/netfilter/nft_set_rbtree.c
+++ b/net/netfilter/nft_set_rbtree.c
@@ -113,6 +113,78 @@ static bool nft_rbtree_lookup(const struct net *net, const struct nft_set *set,
 	return ret;
 }
 
+static bool __nft_rbtree_get(const struct net *net, const struct nft_set *set,
+			     const u32 *key, struct nft_rbtree_elem **elem,
+			     unsigned int seq, unsigned int flags, u8 genmask)
+{
+	struct nft_rbtree_elem *rbe, *interval = NULL;
+	struct nft_rbtree *priv = nft_set_priv(set);
+	const struct rb_node *parent;
+	const void *this;
+	int d;
+
+	parent = rcu_dereference_raw(priv->root.rb_node);
+	while (parent != NULL) {
+		if (read_seqcount_retry(&priv->count, seq))
+			return false;
+
+		rbe = rb_entry(parent, struct nft_rbtree_elem, node);
+
+		this = nft_set_ext_key(&rbe->ext);
+		d = memcmp(this, key, set->klen);
+		if (d < 0) {
+			parent = rcu_dereference_raw(parent->rb_left);
+			interval = rbe;
+		} else if (d > 0) {
+			parent = rcu_dereference_raw(parent->rb_right);
+		} else {
+			if (!nft_set_elem_active(&rbe->ext, genmask))
+				parent = rcu_dereference_raw(parent->rb_left);
+
+			if (!nft_set_ext_exists(&rbe->ext, NFT_SET_EXT_FLAGS) ||
+			    (*nft_set_ext_flags(&rbe->ext) & NFT_SET_ELEM_INTERVAL_END) ==
+			    (flags & NFT_SET_ELEM_INTERVAL_END)) {
+				*elem = rbe;
+				return true;
+			}
+			return false;
+		}
+	}
+
+	if (set->flags & NFT_SET_INTERVAL && interval != NULL &&
+	    nft_set_elem_active(&interval->ext, genmask) &&
+	    !nft_rbtree_interval_end(interval)) {
+		*elem = interval;
+		return true;
+	}
+
+	return false;
+}
+
+static void *nft_rbtree_get(const struct net *net, const struct nft_set *set,
+			    const struct nft_set_elem *elem, unsigned int flags)
+{
+	struct nft_rbtree *priv = nft_set_priv(set);
+	unsigned int seq = read_seqcount_begin(&priv->count);
+	struct nft_rbtree_elem *rbe = ERR_PTR(-ENOENT);
+	const u32 *key = (const u32 *)&elem->key.val;
+	u8 genmask = nft_genmask_cur(net);
+	bool ret;
+
+	ret = __nft_rbtree_get(net, set, key, &rbe, seq, flags, genmask);
+	if (ret || !read_seqcount_retry(&priv->count, seq))
+		return rbe;
+
+	read_lock_bh(&priv->lock);
+	seq = read_seqcount_begin(&priv->count);
+	ret = __nft_rbtree_get(net, set, key, &rbe, seq, flags, genmask);
+	if (!ret)
+		rbe = ERR_PTR(-ENOENT);
+	read_unlock_bh(&priv->lock);
+
+	return rbe;
+}
+
 static int __nft_rbtree_insert(const struct net *net, const struct nft_set *set,
 			       struct nft_rbtree_elem *new,
 			       struct nft_set_ext **ext)
@@ -336,6 +408,7 @@ static struct nft_set_ops nft_rbtree_ops __read_mostly = {
 	.activate	= nft_rbtree_activate,
 	.lookup		= nft_rbtree_lookup,
 	.walk		= nft_rbtree_walk,
+	.get		= nft_rbtree_get,
 	.features	= NFT_SET_INTERVAL | NFT_SET_MAP | NFT_SET_OBJECT,
 };
 
-- 
cgit v1.2.3-59-g8ed1b