[PATCH net 00/11] Netfilter fixes for net

@@ -5,7 +5,8 @@ TEST_PROGS := nft_trans_stress.sh nft_fib.sh nft_nat.sh bridge_brouter.sh \
 	conntrack_icmp_related.sh nft_flowtable.sh ipvs.sh \
 	nft_concat_range.sh nft_conntrack_helper.sh \
 	nft_queue.sh nft_meta.sh nf_nat_edemux.sh \
-	ipip-conntrack-mtu.sh conntrack_tcp_unreplied.sh
+	ipip-conntrack-mtu.sh conntrack_tcp_unreplied.sh \
+	conntrack_vrf.sh
 
 LDLIBS = -lmnl
 TEST_GEN_FILES =  nf-queue

@@ -0,0 +1,219 @@
+#!/bin/sh
+
+# This script demonstrates interaction of conntrack and vrf.
+# The vrf driver calls the netfilter hooks again, with oif/iif
+# pointing at the VRF device.
+#
+# For ingress, this means first iteration has iifname of lower/real
+# device.  In this script, thats veth0.
+# Second iteration is iifname set to vrf device, tvrf in this script.
+#
+# For egress, this is reversed: first iteration has the vrf device,
+# second iteration is done with the lower/real/veth0 device.
+#
+# test_ct_zone_in demonstrates unexpected change of nftables
+# behavior # caused by commit 09e856d54bda5f28 "vrf: Reset skb conntrack
+# connection on VRF rcv"
+#
+# It was possible to assign conntrack zone to a packet (or mark it for
+# `notracking`) in the prerouting chain before conntrack, based on real iif.
+#
+# After the change, the zone assignment is lost and the zone is assigned based
+# on the VRF master interface (in case such a rule exists).
+# assignment is lost. Instead, assignment based on the `iif` matching
+# Thus it is impossible to distinguish packets based on the original
+# interface.
+#
+# test_masquerade_vrf and test_masquerade_veth0 demonstrate the problem
+# that was supposed to be fixed by the commit mentioned above to make sure
+# that any fix to test case 1 won't break masquerade again.
+
+ksft_skip=4
+
+IP0=172.30.30.1
+IP1=172.30.30.2
+PFXL=30
+ret=0
+
+sfx=$(mktemp -u "XXXXXXXX")
+ns0="ns0-$sfx"
+ns1="ns1-$sfx"
+
+cleanup()
+{
+	ip netns pids $ns0 | xargs kill 2>/dev/null
+	ip netns pids $ns1 | xargs kill 2>/dev/null
+
+	ip netns del $ns0 $ns1
+}
+
+nft --version > /dev/null 2>&1
+if [ $? -ne 0 ];then
+	echo "SKIP: Could not run test without nft tool"
+	exit $ksft_skip
+fi
+
+ip -Version > /dev/null 2>&1
+if [ $? -ne 0 ];then
+	echo "SKIP: Could not run test without ip tool"
+	exit $ksft_skip
+fi
+
+ip netns add "$ns0"
+if [ $? -ne 0 ];then
+	echo "SKIP: Could not create net namespace $ns0"
+	exit $ksft_skip
+fi
+ip netns add "$ns1"
+
+trap cleanup EXIT
+
+ip netns exec $ns0 sysctl -q -w net.ipv4.conf.default.rp_filter=0
+ip netns exec $ns0 sysctl -q -w net.ipv4.conf.all.rp_filter=0
+ip netns exec $ns0 sysctl -q -w net.ipv4.conf.all.rp_filter=0
+
+ip link add veth0 netns "$ns0" type veth peer name veth0 netns "$ns1" > /dev/null 2>&1
+if [ $? -ne 0 ];then
+	echo "SKIP: Could not add veth device"
+	exit $ksft_skip
+fi
+
+ip -net $ns0 li add tvrf type vrf table 9876
+if [ $? -ne 0 ];then
+	echo "SKIP: Could not add vrf device"
+	exit $ksft_skip
+fi
+
+ip -net $ns0 li set lo up
+
+ip -net $ns0 li set veth0 master tvrf
+ip -net $ns0 li set tvrf up
+ip -net $ns0 li set veth0 up
+ip -net $ns1 li set veth0 up
+
+ip -net $ns0 addr add $IP0/$PFXL dev veth0
+ip -net $ns1 addr add $IP1/$PFXL dev veth0
+
+ip netns exec $ns1 iperf3 -s > /dev/null 2>&1&
+if [ $? -ne 0 ];then
+	echo "SKIP: Could not start iperf3"
+	exit $ksft_skip
+fi
+
+# test vrf ingress handling.
+# The incoming connection should be placed in conntrack zone 1,
+# as decided by the first iteration of the ruleset.
+test_ct_zone_in()
+{
+ip netns exec $ns0 nft -f - <<EOF
+table testct {
+	chain rawpre {
+		type filter hook prerouting priority raw;
+
+		iif { veth0, tvrf } counter meta nftrace set 1
+		iif veth0 counter ct zone set 1 counter return
+		iif tvrf counter ct zone set 2 counter return
+		ip protocol icmp counter
+		notrack counter
+	}
+
+	chain rawout {
+		type filter hook output priority raw;
+
+		oif veth0 counter ct zone set 1 counter return
+		oif tvrf counter ct zone set 2 counter return
+		notrack counter
+	}
+}
+EOF
+	ip netns exec $ns1 ping -W 1 -c 1 -I veth0 $IP0 > /dev/null
+
+	# should be in zone 1, not zone 2
+	count=$(ip netns exec $ns0 conntrack -L -s $IP1 -d $IP0 -p icmp --zone 1 2>/dev/null | wc -l)
+	if [ $count -eq 1 ]; then
+		echo "PASS: entry found in conntrack zone 1"
+	else
+		echo "FAIL: entry not found in conntrack zone 1"
+		count=$(ip netns exec $ns0 conntrack -L -s $IP1 -d $IP0 -p icmp --zone 2 2> /dev/null | wc -l)
+		if [ $count -eq 1 ]; then
+			echo "FAIL: entry found in zone 2 instead"
+		else
+			echo "FAIL: entry not in zone 1 or 2, dumping table"
+			ip netns exec $ns0 conntrack -L
+			ip netns exec $ns0 nft list ruleset
+		fi
+	fi
+}
+
+# add masq rule that gets evaluated w. outif set to vrf device.
+# This tests the first iteration of the packet through conntrack,
+# oifname is the vrf device.
+test_masquerade_vrf()
+{
+	ip netns exec $ns0 conntrack -F 2>/dev/null
+
+ip netns exec $ns0 nft -f - <<EOF
+flush ruleset
+table ip nat {
+	chain postrouting {
+		type nat hook postrouting priority 0;
+		# NB: masquerade should always be combined with 'oif(name) bla',
+		# lack of this is intentional here, we want to exercise double-snat.
+		ip saddr 172.30.30.0/30 counter masquerade random
+	}
+}
+EOF
+	ip netns exec $ns0 ip vrf exec tvrf iperf3 -t 1 -c $IP1 >/dev/null
+	if [ $? -ne 0 ]; then
+		echo "FAIL: iperf3 connect failure with masquerade + sport rewrite on vrf device"
+		ret=1
+		return
+	fi
+
+	# must also check that nat table was evaluated on second (lower device) iteration.
+	ip netns exec $ns0 nft list table ip nat |grep -q 'counter packets 2'
+	if [ $? -eq 0 ]; then
+		echo "PASS: iperf3 connect with masquerade + sport rewrite on vrf device"
+	else
+		echo "FAIL: vrf masq rule has unexpected counter value"
+		ret=1
+	fi
+}
+
+# add masq rule that gets evaluated w. outif set to veth device.
+# This tests the 2nd iteration of the packet through conntrack,
+# oifname is the lower device (veth0 in this case).
+test_masquerade_veth()
+{
+	ip netns exec $ns0 conntrack -F 2>/dev/null
+ip netns exec $ns0 nft -f - <<EOF
+flush ruleset
+table ip nat {
+	chain postrouting {
+		type nat hook postrouting priority 0;
+		meta oif veth0 ip saddr 172.30.30.0/30 counter masquerade random
+	}
+}
+EOF
+	ip netns exec $ns0 ip vrf exec tvrf iperf3 -t 1 -c $IP1 > /dev/null
+	if [ $? -ne 0 ]; then
+		echo "FAIL: iperf3 connect failure with masquerade + sport rewrite on veth device"
+		ret=1
+		return
+	fi
+
+	# must also check that nat table was evaluated on second (lower device) iteration.
+	ip netns exec $ns0 nft list table ip nat |grep -q 'counter packets 2'
+	if [ $? -eq 0 ]; then
+		echo "PASS: iperf3 connect with masquerade + sport rewrite on veth device"
+	else
+		echo "FAIL: vrf masq rule has unexpected counter value"
+		ret=1
+	fi
+}
+
+test_ct_zone_in
+test_masquerade_vrf
+test_masquerade_veth
+
+exit $ret

@@ -16,6 +16,10 @@ timeout=4
 
 cleanup()
 {
+	ip netns pids ${ns1} | xargs kill 2>/dev/null
+	ip netns pids ${ns2} | xargs kill 2>/dev/null
+	ip netns pids ${nsrouter} | xargs kill 2>/dev/null
+
 	ip netns del ${ns1}
 	ip netns del ${ns2}
 	ip netns del ${nsrouter}

@@ -332,6 +336,55 @@ EOF
 	echo "PASS: tcp via loopback and re-queueing"
 }
 
+test_icmp_vrf() {
+	ip -net $ns1 link add tvrf type vrf table 9876
+	if [ $? -ne 0 ];then
+		echo "SKIP: Could not add vrf device"
+		return
+	fi
+
+	ip -net $ns1 li set eth0 master tvrf
+	ip -net $ns1 li set tvrf up
+
+	ip -net $ns1 route add 10.0.2.0/24 via 10.0.1.1 dev eth0 table 9876
+ip netns exec ${ns1} nft -f /dev/stdin <<EOF
+flush ruleset
+table inet filter {
+	chain output {
+		type filter hook output priority 0; policy accept;
+		meta oifname "tvrf" icmp type echo-request counter queue num 1
+		meta oifname "eth0" icmp type echo-request counter queue num 1
+	}
+	chain post {
+		type filter hook postrouting priority 0; policy accept;
+		meta oifname "tvrf" icmp type echo-request counter queue num 1
+		meta oifname "eth0" icmp type echo-request counter queue num 1
+	}
+}
+EOF
+	ip netns exec ${ns1} ./nf-queue -q 1 -t $timeout &
+	local nfqpid=$!
+
+	sleep 1
+	ip netns exec ${ns1} ip vrf exec tvrf ping -c 1 10.0.2.99 > /dev/null
+
+	for n in output post; do
+		for d in tvrf eth0; do
+			ip netns exec ${ns1} nft list chain inet filter $n | grep -q "oifname \"$d\" icmp type echo-request counter packets 1"
+			if [ $? -ne 0 ] ; then
+				echo "FAIL: chain $n: icmp packet counter mismatch for device $d" 1>&2
+				ip netns exec ${ns1} nft list ruleset
+				ret=1
+				return
+			fi
+		done
+	done
+
+	wait $nfqpid
+	[ $? -eq 0 ] && echo "PASS: icmp+nfqueue via vrf"
+	wait 2>/dev/null
+}
+
 ip netns exec ${nsrouter} sysctl net.ipv6.conf.all.forwarding=1 > /dev/null
 ip netns exec ${nsrouter} sysctl net.ipv4.conf.veth0.forwarding=1 > /dev/null
 ip netns exec ${nsrouter} sysctl net.ipv4.conf.veth1.forwarding=1 > /dev/null

@@ -372,5 +425,6 @@ test_queue 20
 test_tcp_forward
 test_tcp_localhost
 test_tcp_localhost_requeue
+test_icmp_vrf
 
 exit $ret

@@ -759,14 +759,16 @@ test_port_shadow()
 	local result=""
 	local logmsg=""
 
+	# make shadow entry, from client (ns2), going to (ns1), port 41404, sport 1405.
+	echo "fake-entry" | ip netns exec "$ns2" nc -w 1 -p 1405 -u "$daddrc" 41404 > /dev/null
+
 	echo ROUTER | ip netns exec "$ns0" nc -w 5 -u -l -p 1405 >/dev/null 2>&1 &
 	nc_r=$!
 
 	echo CLIENT | ip netns exec "$ns2" nc -w 5 -u -l -p 1405 >/dev/null 2>&1 &
 	nc_c=$!
 
-	# make shadow entry, from client (ns2), going to (ns1), port 41404, sport 1405.
-	echo "fake-entry" | ip netns exec "$ns2" nc -w 1 -p 1405 -u "$daddrc" 41404 > /dev/null
+	sleep 0.3
 
 	# ns1 tries to connect to ns0:1405.  With default settings this should connect
 	# to client, it matches the conntrack entry created above.

@@ -1011,7 +1011,7 @@ ctnetlink_alloc_filter(const struct nlattr * const cda[], u8 family)
 						   CTA_TUPLE_REPLY,
 						   filter->family,
 						   &filter->zone,
-						   filter->orig_flags);
+						   filter->reply_flags);
 		if (err < 0) {
 			err = -EINVAL;
 			goto err_filter;

@@ -1012,10 +1012,8 @@ ctnetlink_alloc_filter(const struct nlattr * const cda[], u8 family)
 						   filter->family,
 						   &filter->zone,
 						   filter->reply_flags);
-		if (err < 0) {
-			err = -EINVAL;
+		if (err < 0)
 			goto err_filter;
-		}
 	}
 
 	return filter;

@@ -22,7 +22,6 @@
 #include <linux/icmpv6.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
-#include <linux/ip.h>
 #include <net/sctp/checksum.h>
 
 static bool nft_payload_rebuild_vlan_hdr(const struct sk_buff *skb, int mac_off,

@@ -85,9 +85,9 @@ static ssize_t idletimer_tg_show(struct device *dev,
 	mutex_unlock(&list_mutex);
 
 	if (time_after(expires, jiffies) || ktimespec.tv_sec > 0)
-		return snprintf(buf, PAGE_SIZE, "%ld\n", time_diff);
+		return sysfs_emit(buf, "%ld\n", time_diff);
 
-	return snprintf(buf, PAGE_SIZE, "0\n");
+	return sysfs_emit(buf, "0\n");
 }
 
 static void idletimer_tg_work(struct work_struct *work)

@@ -818,11 +818,10 @@ table $family raw {
 	chain prerouting {
 		type filter hook prerouting priority -300; policy accept;
 		meta iif veth0 udp dport 1405 notrack
-		udp dport 1405 notrack
 	}
 	chain output {
 		type filter hook output priority -300; policy accept;
-		udp sport 1405 notrack
+		meta oif veth0 udp sport 1405 notrack
 	}
 }
 EOF

@@ -37,8 +37,7 @@ conn_reuse_mode - INTEGER
 
 	0: disable any special handling on port reuse. The new
 	connection will be delivered to the same real server that was
-	servicing the previous connection. This will effectively
-	disable expire_nodest_conn.
+	servicing the previous connection.
 
 	bit 1: enable rescheduling of new connections when it is safe.
 	That is, whenever expire_nodest_conn and for TCP sockets, when

@@ -1919,7 +1919,6 @@ ip_vs_in_hook(void *priv, struct sk_buff *skb, const struct nf_hook_state *state
 	struct ip_vs_proto_data *pd;
 	struct ip_vs_conn *cp;
 	int ret, pkts;
-	int conn_reuse_mode;
 	struct sock *sk;
 	int af = state->pf;
 

@@ -1997,15 +1996,16 @@ ip_vs_in_hook(void *priv, struct sk_buff *skb, const struct nf_hook_state *state
 	cp = INDIRECT_CALL_1(pp->conn_in_get, ip_vs_conn_in_get_proto,
 			     ipvs, af, skb, &iph);
 
-	conn_reuse_mode = sysctl_conn_reuse_mode(ipvs);
-	if (conn_reuse_mode && !iph.fragoffs && is_new_conn(skb, &iph) && cp) {
+	if (!iph.fragoffs && is_new_conn(skb, &iph) && cp) {
+		int conn_reuse_mode = sysctl_conn_reuse_mode(ipvs);
 		bool old_ct = false, resched = false;
 
 		if (unlikely(sysctl_expire_nodest_conn(ipvs)) && cp->dest &&
 		    unlikely(!atomic_read(&cp->dest->weight))) {
 			resched = true;
 			old_ct = ip_vs_conn_uses_old_conntrack(cp, skb);
-		} else if (is_new_conn_expected(cp, conn_reuse_mode)) {
+		} else if (conn_reuse_mode &&
+			   is_new_conn_expected(cp, conn_reuse_mode)) {
 			old_ct = ip_vs_conn_uses_old_conntrack(cp, skb);
 			if (!atomic_read(&cp->n_control)) {
 				resched = true;

@@ -760,20 +760,20 @@ test_port_shadow()
 	local logmsg=""
 
 	# make shadow entry, from client (ns2), going to (ns1), port 41404, sport 1405.
-	echo "fake-entry" | ip netns exec "$ns2" nc -w 1 -p 1405 -u "$daddrc" 41404 > /dev/null
+	echo "fake-entry" | ip netns exec "$ns2" timeout 1 socat -u STDIN UDP:"$daddrc":41404,sourceport=1405
 
-	echo ROUTER | ip netns exec "$ns0" nc -w 5 -u -l -p 1405 >/dev/null 2>&1 &
-	nc_r=$!
+	echo ROUTER | ip netns exec "$ns0" timeout 5 socat -u STDIN UDP4-LISTEN:1405 &
+	sc_r=$!
 
-	echo CLIENT | ip netns exec "$ns2" nc -w 5 -u -l -p 1405 >/dev/null 2>&1 &
-	nc_c=$!
+	echo CLIENT | ip netns exec "$ns2" timeout 5 socat -u STDIN UDP4-LISTEN:1405,reuseport &
+	sc_c=$!
 
 	sleep 0.3
 
 	# ns1 tries to connect to ns0:1405.  With default settings this should connect
 	# to client, it matches the conntrack entry created above.
 
-	result=$(echo "" | ip netns exec "$ns1" nc -w 1 -p 41404 -u "$daddrs" 1405)
+	result=$(echo "data" | ip netns exec "$ns1" timeout 1 socat - UDP:"$daddrs":1405,sourceport=41404)
 
 	if [ "$result" = "$expect" ] ;then
 		echo "PASS: portshadow test $test: got reply from ${expect}${logmsg}"

@@ -782,7 +782,7 @@ test_port_shadow()
 		ret=1
 	fi
 
-	kill $nc_r $nc_c 2>/dev/null
+	kill $sc_r $sc_c 2>/dev/null
 
 	# flush udp entries for next test round, if any
 	ip netns exec "$ns0" conntrack -F >/dev/null 2>&1

@@ -852,6 +852,18 @@ test_port_shadowing()
 {
 	local family="ip"
 
+	conntrack -h >/dev/null 2>&1
+	if [ $? -ne 0 ];then
+		echo "SKIP: Could not run nat port shadowing test without conntrack tool"
+		return
+	fi
+
+	socat -h > /dev/null 2>&1
+	if [ $? -ne 0 ];then
+		echo "SKIP: Could not run nat port shadowing test without socat tool"
+		return
+	fi
+
 	ip netns exec "$ns0" sysctl net.ipv4.conf.veth0.forwarding=1 > /dev/null
 	ip netns exec "$ns0" sysctl net.ipv4.conf.veth1.forwarding=1 > /dev/null
 

@@ -65,11 +65,11 @@ static void nf_flow_rule_lwt_match(struct nf_flow_match *match,
 		       sizeof(struct in6_addr));
 		if (memcmp(&key->enc_ipv6.src, &in6addr_any,
 			   sizeof(struct in6_addr)))
-			memset(&key->enc_ipv6.src, 0xff,
+			memset(&mask->enc_ipv6.src, 0xff,
 			       sizeof(struct in6_addr));
 		if (memcmp(&key->enc_ipv6.dst, &in6addr_any,
 			   sizeof(struct in6_addr)))
-			memset(&key->enc_ipv6.dst, 0xff,
+			memset(&mask->enc_ipv6.dst, 0xff,
 			       sizeof(struct in6_addr));
 		enc_keys |= BIT(FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS);
 		key->enc_control.addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS;