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

Pablo Neira Ayuso says:

====================
Netfilter fixes for net

The following patchset contains Netfilter fixes for net:

1) Fix a crash when stateful expression with its own gc callback
   is used in a set definition.

2) Skip IPv6 packets from any link-local address in IPv6 fib expression.
   Add a selftest for this scenario, from Florian Westphal.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

4 files changed, 283 insertions, 47 deletions

diff --git a/net/ipv6/netfilter/nft_fib_ipv6.c b/net/ipv6/netfilter/nft_fib_ipv6.c
index e204163c7036..92f3235fa287 100644
--- a/net/ipv6/netfilter/nft_fib_ipv6.c
+++ b/net/ipv6/netfilter/nft_fib_ipv6.c
@@ -135,6 +135,17 @@ void nft_fib6_eval_type(const struct nft_expr *expr, struct nft_regs *regs,
 }
 EXPORT_SYMBOL_GPL(nft_fib6_eval_type);
 
+static bool nft_fib_v6_skip_icmpv6(const struct sk_buff *skb, u8 next, const struct ipv6hdr *iph)
+{
+	if (likely(next != IPPROTO_ICMPV6))
+		return false;
+
+	if (ipv6_addr_type(&iph->saddr) != IPV6_ADDR_ANY)
+		return false;
+
+	return ipv6_addr_type(&iph->daddr) & IPV6_ADDR_LINKLOCAL;
+}
+
 void nft_fib6_eval(const struct nft_expr *expr, struct nft_regs *regs,
 		   const struct nft_pktinfo *pkt)
 {
@@ -163,10 +174,13 @@ void nft_fib6_eval(const struct nft_expr *expr, struct nft_regs *regs,
 
 	lookup_flags = nft_fib6_flowi_init(&fl6, priv, pkt, oif, iph);
 
-	if (nft_hook(pkt) == NF_INET_PRE_ROUTING &&
-	    nft_fib_is_loopback(pkt->skb, nft_in(pkt))) {
-		nft_fib_store_result(dest, priv, nft_in(pkt));
-		return;
+	if (nft_hook(pkt) == NF_INET_PRE_ROUTING ||
+	    nft_hook(pkt) == NF_INET_INGRESS) {
+		if (nft_fib_is_loopback(pkt->skb, nft_in(pkt)) ||
+		    nft_fib_v6_skip_icmpv6(pkt->skb, pkt->tprot, iph)) {
+			nft_fib_store_result(dest, priv, nft_in(pkt));
+			return;
+		}
 	}
 
 	*dest = 0;
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 72bc759179ef..bf4d6ec9fc55 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -4364,13 +4364,45 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info,
 	err = nf_tables_set_alloc_name(&ctx, set, name);
 	kfree(name);
 	if (err < 0)
-		goto err_set_alloc_name;
+		goto err_set_name;
+
+	udata = NULL;
+	if (udlen) {
+		udata = set->data + size;
+		nla_memcpy(udata, nla[NFTA_SET_USERDATA], udlen);
+	}
+
+	INIT_LIST_HEAD(&set->bindings);
+	INIT_LIST_HEAD(&set->catchall_list);
+	set->table = table;
+	write_pnet(&set->net, net);
+	set->ops = ops;
+	set->ktype = ktype;
+	set->klen = desc.klen;
+	set->dtype = dtype;
+	set->objtype = objtype;
+	set->dlen = desc.dlen;
+	set->flags = flags;
+	set->size = desc.size;
+	set->policy = policy;
+	set->udlen = udlen;
+	set->udata = udata;
+	set->timeout = timeout;
+	set->gc_int = gc_int;
+
+	set->field_count = desc.field_count;
+	for (i = 0; i < desc.field_count; i++)
+		set->field_len[i] = desc.field_len[i];
+
+	err = ops->init(set, &desc, nla);
+	if (err < 0)
+		goto err_set_init;
 
 	if (nla[NFTA_SET_EXPR]) {
 		expr = nft_set_elem_expr_alloc(&ctx, set, nla[NFTA_SET_EXPR]);
 		if (IS_ERR(expr)) {
 			err = PTR_ERR(expr);
-			goto err_set_alloc_name;
+			goto err_set_expr_alloc;
 		}
 		set->exprs[0] = expr;
 		set->num_exprs++;
@@ -4381,75 +4413,44 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info,
 
 		if (!(flags & NFT_SET_EXPR)) {
 			err = -EINVAL;
-			goto err_set_alloc_name;
+			goto err_set_expr_alloc;
 		}
 		i = 0;
 		nla_for_each_nested(tmp, nla[NFTA_SET_EXPRESSIONS], left) {
 			if (i == NFT_SET_EXPR_MAX) {
 				err = -E2BIG;
-				goto err_set_init;
+				goto err_set_expr_alloc;
 			}
 			if (nla_type(tmp) != NFTA_LIST_ELEM) {
 				err = -EINVAL;
-				goto err_set_init;
+				goto err_set_expr_alloc;
 			}
 			expr = nft_set_elem_expr_alloc(&ctx, set, tmp);
 			if (IS_ERR(expr)) {
 				err = PTR_ERR(expr);
-				goto err_set_init;
+				goto err_set_expr_alloc;
 			}
 			set->exprs[i++] = expr;
 			set->num_exprs++;
 		}
 	}
 
-	udata = NULL;
-	if (udlen) {
-		udata = set->data + size;
-		nla_memcpy(udata, nla[NFTA_SET_USERDATA], udlen);
-	}
-
-	INIT_LIST_HEAD(&set->bindings);
-	INIT_LIST_HEAD(&set->catchall_list);
-	set->table = table;
-	write_pnet(&set->net, net);
-	set->ops   = ops;
-	set->ktype = ktype;
-	set->klen  = desc.klen;
-	set->dtype = dtype;
-	set->objtype = objtype;
-	set->dlen  = desc.dlen;
-	set->flags = flags;
-	set->size  = desc.size;
-	set->policy = policy;
-	set->udlen  = udlen;
-	set->udata  = udata;
-	set->timeout = timeout;
-	set->gc_int = gc_int;
 	set->handle = nf_tables_alloc_handle(table);
 
-	set->field_count = desc.field_count;
-	for (i = 0; i < desc.field_count; i++)
-		set->field_len[i] = desc.field_len[i];
-
-	err = ops->init(set, &desc, nla);
-	if (err < 0)
-		goto err_set_init;
-
 	err = nft_trans_set_add(&ctx, NFT_MSG_NEWSET, set);
 	if (err < 0)
-		goto err_set_trans;
+		goto err_set_expr_alloc;
 
 	list_add_tail_rcu(&set->list, &table->sets);
 	table->use++;
 	return 0;
 
-err_set_trans:
-	ops->destroy(set);
-err_set_init:
+err_set_expr_alloc:
 	for (i = 0; i < set->num_exprs; i++)
 		nft_expr_destroy(&ctx, set->exprs[i]);
-err_set_alloc_name:
+
+	ops->destroy(set);
+err_set_init:
 	kfree(set->name);
 err_set_name:
 	kvfree(set);
diff --git a/tools/testing/selftests/netfilter/Makefile b/tools/testing/selftests/netfilter/Makefile
index 3171069a6b46..cd6430b39982 100644
--- a/tools/testing/selftests/netfilter/Makefile
+++ b/tools/testing/selftests/netfilter/Makefile
@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0
 # Makefile for netfilter selftests
 
-TEST_PROGS := nft_trans_stress.sh nft_nat.sh bridge_brouter.sh \
+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 \
diff --git a/tools/testing/selftests/netfilter/nft_fib.sh b/tools/testing/selftests/netfilter/nft_fib.sh
new file mode 100755
index 000000000000..6caf6ac8c285
--- /dev/null
+++ b/tools/testing/selftests/netfilter/nft_fib.sh
@@ -0,0 +1,221 @@
+#!/bin/bash
+#
+# This tests the fib expression.
+#
+# Kselftest framework requirement - SKIP code is 4.
+ksft_skip=4
+ret=0
+
+sfx=$(mktemp -u "XXXXXXXX")
+ns1="ns1-$sfx"
+ns2="ns2-$sfx"
+nsrouter="nsrouter-$sfx"
+timeout=4
+
+log_netns=$(sysctl -n net.netfilter.nf_log_all_netns)
+
+cleanup()
+{
+	ip netns del ${ns1}
+	ip netns del ${ns2}
+	ip netns del ${nsrouter}
+
+	[ $log_netns -eq 0 ] && sysctl -q net.netfilter.nf_log_all_netns=$log_netns
+}
+
+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 ${nsrouter}
+if [ $? -ne 0 ];then
+	echo "SKIP: Could not create net namespace"
+	exit $ksft_skip
+fi
+
+trap cleanup EXIT
+
+dmesg | grep -q ' nft_rpfilter: '
+if [ $? -eq 0 ]; then
+	dmesg -c | grep ' nft_rpfilter: '
+	echo "WARN: a previous test run has failed" 1>&2
+fi
+
+sysctl -q net.netfilter.nf_log_all_netns=1
+ip netns add ${ns1}
+ip netns add ${ns2}
+
+load_ruleset() {
+	local netns=$1
+
+ip netns exec ${netns} nft -f /dev/stdin <<EOF
+table inet filter {
+	chain prerouting {
+		type filter hook prerouting priority 0; policy accept;
+	        fib saddr . iif oif missing counter log prefix "$netns nft_rpfilter: " drop
+	}
+}
+EOF
+}
+
+load_ruleset_count() {
+	local netns=$1
+
+ip netns exec ${netns} nft -f /dev/stdin <<EOF
+table inet filter {
+	chain prerouting {
+		type filter hook prerouting priority 0; policy accept;
+		ip daddr 1.1.1.1 fib saddr . iif oif missing counter drop
+		ip6 daddr 1c3::c01d fib saddr . iif oif missing counter drop
+	}
+}
+EOF
+}
+
+check_drops() {
+	dmesg | grep -q ' nft_rpfilter: '
+	if [ $? -eq 0 ]; then
+		dmesg | grep ' nft_rpfilter: '
+		echo "FAIL: rpfilter did drop packets"
+		return 1
+	fi
+
+	return 0
+}
+
+check_fib_counter() {
+	local want=$1
+	local ns=$2
+	local address=$3
+
+	line=$(ip netns exec ${ns} nft list table inet filter | grep 'fib saddr . iif' | grep $address | grep "packets $want" )
+	ret=$?
+
+	if [ $ret -ne 0 ];then
+		echo "Netns $ns fib counter doesn't match expected packet count of $want for $address" 1>&2
+		ip netns exec ${ns} nft list table inet filter
+		return 1
+	fi
+
+	if [ $want -gt 0 ]; then
+		echo "PASS: fib expression did drop packets for $address"
+	fi
+
+	return 0
+}
+
+load_ruleset ${nsrouter}
+load_ruleset ${ns1}
+load_ruleset ${ns2}
+
+ip link add veth0 netns ${nsrouter} type veth peer name eth0 netns ${ns1} > /dev/null 2>&1
+if [ $? -ne 0 ];then
+    echo "SKIP: No virtual ethernet pair device support in kernel"
+    exit $ksft_skip
+fi
+ip link add veth1 netns ${nsrouter} type veth peer name eth0 netns ${ns2}
+
+ip -net ${nsrouter} link set lo up
+ip -net ${nsrouter} link set veth0 up
+ip -net ${nsrouter} addr add 10.0.1.1/24 dev veth0
+ip -net ${nsrouter} addr add dead:1::1/64 dev veth0
+
+ip -net ${nsrouter} link set veth1 up
+ip -net ${nsrouter} addr add 10.0.2.1/24 dev veth1
+ip -net ${nsrouter} addr add dead:2::1/64 dev veth1
+
+ip -net ${ns1} link set lo up
+ip -net ${ns1} link set eth0 up
+
+ip -net ${ns2} link set lo up
+ip -net ${ns2} link set eth0 up
+
+ip -net ${ns1} addr add 10.0.1.99/24 dev eth0
+ip -net ${ns1} addr add dead:1::99/64 dev eth0
+ip -net ${ns1} route add default via 10.0.1.1
+ip -net ${ns1} route add default via dead:1::1
+
+ip -net ${ns2} addr add 10.0.2.99/24 dev eth0
+ip -net ${ns2} addr add dead:2::99/64 dev eth0
+ip -net ${ns2} route add default via 10.0.2.1
+ip -net ${ns2} route add default via dead:2::1
+
+test_ping() {
+  local daddr4=$1
+  local daddr6=$2
+
+  ip netns exec ${ns1} ping -c 1 -q $daddr4 > /dev/null
+  ret=$?
+  if [ $ret -ne 0 ];then
+	check_drops
+	echo "FAIL: ${ns1} cannot reach $daddr4, ret $ret" 1>&2
+	return 1
+  fi
+
+  ip netns exec ${ns1} ping -c 3 -q $daddr6 > /dev/null
+  ret=$?
+  if [ $ret -ne 0 ];then
+	check_drops
+	echo "FAIL: ${ns1} cannot reach $daddr6, ret $ret" 1>&2
+	return 1
+  fi
+
+  return 0
+}
+
+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
+
+sleep 3
+
+test_ping 10.0.2.1 dead:2::1 || exit 1
+check_drops || exit 1
+
+test_ping 10.0.2.99 dead:2::99 || exit 1
+check_drops || exit 1
+
+echo "PASS: fib expression did not cause unwanted packet drops"
+
+ip netns exec ${nsrouter} nft flush table inet filter
+
+ip -net ${ns1} route del default
+ip -net ${ns1} -6 route del default
+
+ip -net ${ns1} addr del 10.0.1.99/24 dev eth0
+ip -net ${ns1} addr del dead:1::99/64 dev eth0
+
+ip -net ${ns1} addr add 10.0.2.99/24 dev eth0
+ip -net ${ns1} addr add dead:2::99/64 dev eth0
+
+ip -net ${ns1} route add default via 10.0.2.1
+ip -net ${ns1} -6 route add default via dead:2::1
+
+ip -net ${nsrouter} addr add dead:2::1/64 dev veth0
+
+# switch to ruleset that doesn't log, this time
+# its expected that this does drop the packets.
+load_ruleset_count ${nsrouter}
+
+# ns1 has a default route, but nsrouter does not.
+# must not check return value, ping to 1.1.1.1 will
+# fail.
+check_fib_counter 0 ${nsrouter} 1.1.1.1 || exit 1
+check_fib_counter 0 ${nsrouter} 1c3::c01d || exit 1
+
+ip netns exec ${ns1} ping -c 1 -W 1 -q 1.1.1.1 > /dev/null
+check_fib_counter 1 ${nsrouter} 1.1.1.1 || exit 1
+
+sleep 2
+ip netns exec ${ns1} ping -c 3 -q 1c3::c01d > /dev/null
+check_fib_counter 3 ${nsrouter} 1c3::c01d || exit 1
+
+exit 0