selftests/bpf: expand bpf tunnel test to ipv6

The test only uses ipv4 so far, expand to ipv6.
This is mostly a boilerplate near copy of the ipv4 path.
Signed-off-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>

tools/testing/selftests/bpf/config

@@ -23,3 +23,5 @@ CONFIG_LWTUNNEL=y
 CONFIG_BPF_STREAM_PARSER=y
 CONFIG_XDP_SOCKETS=y
 CONFIG_FTRACE_SYSCALLS=y
+CONFIG_IPV6_TUNNEL=y
+CONFIG_IPV6_GRE=y

tools/testing/selftests/bpf/progs/test_tc_tunnel.c

@@ -7,6 +7,7 @@
 #include <linux/if_ether.h>
 #include <linux/in.h>
 #include <linux/ip.h>
+#include <linux/ipv6.h>
 #include <linux/tcp.h>
 #include <linux/pkt_cls.h>
 #include <linux/types.h>
@@ -31,15 +32,11 @@ static __always_inline void set_ipv4_csum(struct iphdr *iph)
 	iph->check = ~((csum & 0xffff) + (csum >> 16));
 }
 
-SEC("encap")
-int encap_f(struct __sk_buff *skb)
+static int encap_ipv4(struct __sk_buff *skb)
 {
 	struct iphdr iph_outer, iph_inner;
 	struct tcphdr tcph;
 
-	if (skb->protocol != __bpf_constant_htons(ETH_P_IP))
-		return TC_ACT_OK;
-
 	if (bpf_skb_load_bytes(skb, ETH_HLEN, &iph_inner,
 			       sizeof(iph_inner)) < 0)
 		return TC_ACT_OK;
@@ -80,35 +77,118 @@ int encap_f(struct __sk_buff *skb)
 	return TC_ACT_OK;
 }
 
-SEC("decap")
-int decap_f(struct __sk_buff *skb)
+static int encap_ipv6(struct __sk_buff *skb)
 {
-	struct iphdr iph_outer, iph_inner;
+	struct ipv6hdr iph_outer, iph_inner;
+	struct tcphdr tcph;
 
-	if (skb->protocol != __bpf_constant_htons(ETH_P_IP))
+	if (bpf_skb_load_bytes(skb, ETH_HLEN, &iph_inner,
+			       sizeof(iph_inner)) < 0)
 		return TC_ACT_OK;
 
-	if (bpf_skb_load_bytes(skb, ETH_HLEN, &iph_outer,
-			       sizeof(iph_outer)) < 0)
+	/* filter only packets we want */
+	if (bpf_skb_load_bytes(skb, ETH_HLEN + sizeof(iph_inner),
+			       &tcph, sizeof(tcph)) < 0)
 		return TC_ACT_OK;
 
-	if (iph_outer.ihl != 5 || iph_outer.protocol != IPPROTO_IPIP)
+	if (tcph.dest != __bpf_constant_htons(cfg_port))
+		return TC_ACT_OK;
+
+	/* add room between mac and network header */
+	if (bpf_skb_adjust_room(skb, sizeof(iph_outer), BPF_ADJ_ROOM_NET, 0))
+		return TC_ACT_SHOT;
+
+	/* prepare new outer network header */
+	iph_outer = iph_inner;
+	iph_outer.nexthdr = IPPROTO_IPV6;
+	iph_outer.payload_len = bpf_htons(sizeof(iph_outer) +
+					  bpf_ntohs(iph_outer.payload_len));
+
+	/* store new outer network header */
+	if (bpf_skb_store_bytes(skb, ETH_HLEN, &iph_outer, sizeof(iph_outer),
+				BPF_F_INVALIDATE_HASH) < 0)
+		return TC_ACT_SHOT;
+
+	/* bpf_skb_adjust_room has moved header to start of room: restore */
+	if (bpf_skb_store_bytes(skb, ETH_HLEN + sizeof(iph_outer),
+				&iph_inner, sizeof(iph_inner),
+				BPF_F_INVALIDATE_HASH) < 0)
+		return TC_ACT_SHOT;
+
+	return TC_ACT_OK;
+}
+
+SEC("encap")
+int encap_f(struct __sk_buff *skb)
+{
+	switch (skb->protocol) {
+	case __bpf_constant_htons(ETH_P_IP):
+		return encap_ipv4(skb);
+	case __bpf_constant_htons(ETH_P_IPV6):
+		return encap_ipv6(skb);
+	default:
+		/* does not match, ignore */
 		return TC_ACT_OK;
+	}
+}
 
-	if (bpf_skb_load_bytes(skb, ETH_HLEN + sizeof(iph_outer),
-			       &iph_inner, sizeof(iph_inner)) < 0)
+static int decap_internal(struct __sk_buff *skb, int off, int len)
+{
+	char buf[sizeof(struct ipv6hdr)];
+
+	if (bpf_skb_load_bytes(skb, off + len, &buf, len) < 0)
 		return TC_ACT_OK;
 
-	if (bpf_skb_adjust_room(skb, -(int)sizeof(iph_outer),
-				BPF_ADJ_ROOM_NET, 0))
+	if (bpf_skb_adjust_room(skb, -len, BPF_ADJ_ROOM_NET, 0))
 		return TC_ACT_SHOT;
 
 	/* bpf_skb_adjust_room has moved outer over inner header: restore */
-	if (bpf_skb_store_bytes(skb, ETH_HLEN, &iph_inner, sizeof(iph_inner),
-				BPF_F_INVALIDATE_HASH) < 0)
+	if (bpf_skb_store_bytes(skb, off, buf, len, BPF_F_INVALIDATE_HASH) < 0)
 		return TC_ACT_SHOT;
 
 	return TC_ACT_OK;
 }
 
+static int decap_ipv4(struct __sk_buff *skb)
+{
+	struct iphdr iph_outer;
+
+	if (bpf_skb_load_bytes(skb, ETH_HLEN, &iph_outer,
+			       sizeof(iph_outer)) < 0)
+		return TC_ACT_OK;
+
+	if (iph_outer.ihl != 5 || iph_outer.protocol != IPPROTO_IPIP)
+		return TC_ACT_OK;
+
+	return decap_internal(skb, ETH_HLEN, sizeof(iph_outer));
+}
+
+static int decap_ipv6(struct __sk_buff *skb)
+{
+	struct ipv6hdr iph_outer;
+
+	if (bpf_skb_load_bytes(skb, ETH_HLEN, &iph_outer,
+			       sizeof(iph_outer)) < 0)
+		return TC_ACT_OK;
+
+	if (iph_outer.nexthdr != IPPROTO_IPV6)
+		return TC_ACT_OK;
+
+	return decap_internal(skb, ETH_HLEN, sizeof(iph_outer));
+}
+
+SEC("decap")
+int decap_f(struct __sk_buff *skb)
+{
+	switch (skb->protocol) {
+	case __bpf_constant_htons(ETH_P_IP):
+		return decap_ipv4(skb);
+	case __bpf_constant_htons(ETH_P_IPV6):
+		return decap_ipv6(skb);
+	default:
+		/* does not match, ignore */
+		return TC_ACT_OK;
+	}
+}
+
 char __license[] SEC("license") = "GPL";

tools/testing/selftests/bpf/test_tc_tunnel.sh

@@ -12,6 +12,9 @@ readonly ns2="${ns_prefix}2"
 
 readonly ns1_v4=192.168.1.1
 readonly ns2_v4=192.168.1.2
+readonly ns1_v6=fd::1
+readonly ns2_v6=fd::2
+
 
 setup() {
 	ip netns add "${ns1}"
@@ -25,6 +28,8 @@ setup() {
 
 	ip -netns "${ns1}" -4 addr add "${ns1_v4}/24" dev veth1
 	ip -netns "${ns2}" -4 addr add "${ns2_v4}/24" dev veth2
+	ip -netns "${ns1}" -6 addr add "${ns1_v6}/64" dev veth1 nodad
+	ip -netns "${ns2}" -6 addr add "${ns2_v6}/64" dev veth2 nodad
 
 	sleep 1
 }
@@ -35,16 +40,56 @@ cleanup() {
 }
 
 server_listen() {
-	ip netns exec "${ns2}" nc -l -p "${port}" &
+	ip netns exec "${ns2}" nc "${netcat_opt}" -l -p "${port}" &
 	sleep 0.2
 }
 
 client_connect() {
-	ip netns exec "${ns1}" nc -z -w 1 "${ns2_v4}" "${port}"
+	ip netns exec "${ns1}" nc "${netcat_opt}" -z -w 1 "${addr2}" "${port}"
 	echo $?
 }
 
 set -e
+
+# no arguments: automated test, run all
+if [[ "$#" -eq "0" ]]; then
+	echo "ipip"
+	$0 ipv4
+
+	echo "ip6ip6"
+	$0 ipv6
+
+	echo "OK. All tests passed"
+	exit 0
+fi
+
+if [[ "$#" -ne "1" ]]; then
+	echo "Usage: $0"
+	echo "   or: $0 <ipv4|ipv6>"
+	exit 1
+fi
+
+case "$1" in
+"ipv4")
+	readonly tuntype=ipip
+	readonly addr1="${ns1_v4}"
+	readonly addr2="${ns2_v4}"
+	readonly netcat_opt=-4
+	;;
+"ipv6")
+	readonly tuntype=ip6tnl
+	readonly addr1="${ns1_v6}"
+	readonly addr2="${ns2_v6}"
+	readonly netcat_opt=-6
+	;;
+*)
+	echo "unknown arg: $1"
+	exit 1
+	;;
+esac
+
+echo "encap ${addr1} to ${addr2}, type ${tuntype}"
+
 trap cleanup EXIT
 
 setup
@@ -66,8 +111,8 @@ server_listen
 # serverside, insert decap module
 # server is still running
 # client can connect again
-ip netns exec "${ns2}" ip link add dev testtun0 type ipip \
-	remote "${ns1_v4}" local "${ns2_v4}"
+ip netns exec "${ns2}" ip link add dev testtun0 type "${tuntype}" \
+	remote "${addr1}" local "${addr2}"
 ip netns exec "${ns2}" ip link set dev testtun0 up
 echo "test bpf encap with tunnel device decap"
 client_connect