Merge branch 'pmtu-esp'

Vadim Fedorenko ays:

====================
Fix PMTU for ESP-in-UDP encapsulation

Bug 213669 uncovered regression in PMTU discovery for UDP-encapsulated
routes and some incorrect usage in udp tunnel fields. This series fixes
problems and also adds such case for selftests

v3:
 - update checking logic to account SCTP use case
v2:
 - remove refactor code that was in first patch
 - move checking logic to __udp{4,6}_lib_err_encap
 - add more tests, especially routed configuration
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

net/ipv4/udp.c

@@ -645,10 +645,12 @@ static struct sock *__udp4_lib_err_encap(struct net *net,
 					 const struct iphdr *iph,
 					 struct udphdr *uh,
 					 struct udp_table *udptable,
+					 struct sock *sk,
 					 struct sk_buff *skb, u32 info)
 {
+	int (*lookup)(struct sock *sk, struct sk_buff *skb);
 	int network_offset, transport_offset;
-	struct sock *sk;
+	struct udp_sock *up;
 
 	network_offset = skb_network_offset(skb);
 	transport_offset = skb_transport_offset(skb);
@@ -659,18 +661,28 @@ static struct sock *__udp4_lib_err_encap(struct net *net,
 	/* Transport header needs to point to the UDP header */
 	skb_set_transport_header(skb, iph->ihl << 2);
 
+	if (sk) {
+		up = udp_sk(sk);
+
+		lookup = READ_ONCE(up->encap_err_lookup);
+		if (lookup && lookup(sk, skb))
+			sk = NULL;
+
+		goto out;
+	}
+
 	sk = __udp4_lib_lookup(net, iph->daddr, uh->source,
 			       iph->saddr, uh->dest, skb->dev->ifindex, 0,
 			       udptable, NULL);
 	if (sk) {
-		int (*lookup)(struct sock *sk, struct sk_buff *skb);
-		struct udp_sock *up = udp_sk(sk);
+		up = udp_sk(sk);
 
 		lookup = READ_ONCE(up->encap_err_lookup);
 		if (!lookup || lookup(sk, skb))
 			sk = NULL;
 	}
 
+out:
 	if (!sk)
 		sk = ERR_PTR(__udp4_lib_err_encap_no_sk(skb, info));
 
@@ -707,15 +719,16 @@ int __udp4_lib_err(struct sk_buff *skb, u32 info, struct udp_table *udptable)
 	sk = __udp4_lib_lookup(net, iph->daddr, uh->dest,
 			       iph->saddr, uh->source, skb->dev->ifindex,
 			       inet_sdif(skb), udptable, NULL);
+
 	if (!sk || udp_sk(sk)->encap_type) {
 		/* No socket for error: try tunnels before discarding */
-		sk = ERR_PTR(-ENOENT);
 		if (static_branch_unlikely(&udp_encap_needed_key)) {
-			sk = __udp4_lib_err_encap(net, iph, uh, udptable, skb,
+			sk = __udp4_lib_err_encap(net, iph, uh, udptable, sk, skb,
 						  info);
 			if (!sk)
 				return 0;
-		}
+		} else
+			sk = ERR_PTR(-ENOENT);
 
 		if (IS_ERR(sk)) {
 			__ICMP_INC_STATS(net, ICMP_MIB_INERRORS);

net/ipv6/udp.c

@@ -502,12 +502,14 @@ static struct sock *__udp6_lib_err_encap(struct net *net,
 					 const struct ipv6hdr *hdr, int offset,
 					 struct udphdr *uh,
 					 struct udp_table *udptable,
+					 struct sock *sk,
 					 struct sk_buff *skb,
 					 struct inet6_skb_parm *opt,
 					 u8 type, u8 code, __be32 info)
 {
+	int (*lookup)(struct sock *sk, struct sk_buff *skb);
 	int network_offset, transport_offset;
-	struct sock *sk;
+	struct udp_sock *up;
 
 	network_offset = skb_network_offset(skb);
 	transport_offset = skb_transport_offset(skb);
@@ -518,18 +520,28 @@ static struct sock *__udp6_lib_err_encap(struct net *net,
 	/* Transport header needs to point to the UDP header */
 	skb_set_transport_header(skb, offset);
 
+	if (sk) {
+		up = udp_sk(sk);
+
+		lookup = READ_ONCE(up->encap_err_lookup);
+		if (lookup && lookup(sk, skb))
+			sk = NULL;
+
+		goto out;
+	}
+
 	sk = __udp6_lib_lookup(net, &hdr->daddr, uh->source,
 			       &hdr->saddr, uh->dest,
 			       inet6_iif(skb), 0, udptable, skb);
 	if (sk) {
-		int (*lookup)(struct sock *sk, struct sk_buff *skb);
-		struct udp_sock *up = udp_sk(sk);
+		up = udp_sk(sk);
 
 		lookup = READ_ONCE(up->encap_err_lookup);
 		if (!lookup || lookup(sk, skb))
 			sk = NULL;
 	}
 
+out:
 	if (!sk) {
 		sk = ERR_PTR(__udp6_lib_err_encap_no_sk(skb, opt, type, code,
 							offset, info));
@@ -558,16 +570,17 @@ int __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 
 	sk = __udp6_lib_lookup(net, daddr, uh->dest, saddr, uh->source,
 			       inet6_iif(skb), inet6_sdif(skb), udptable, NULL);
+
 	if (!sk || udp_sk(sk)->encap_type) {
 		/* No socket for error: try tunnels before discarding */
-		sk = ERR_PTR(-ENOENT);
 		if (static_branch_unlikely(&udpv6_encap_needed_key)) {
 			sk = __udp6_lib_err_encap(net, hdr, offset, uh,
-						  udptable, skb,
+						  udptable, sk, skb,
 						  opt, type, code, info);
 			if (!sk)
 				return 0;
-		}
+		} else
+			sk = ERR_PTR(-ENOENT);
 
 		if (IS_ERR(sk)) {
 			__ICMP6_INC_STATS(net, __in6_dev_get(skb->dev),

tools/testing/selftests/net/nettest.c

@@ -11,9 +11,11 @@
 #include <sys/socket.h>
 #include <sys/wait.h>
 #include <linux/tcp.h>
+#include <linux/udp.h>
 #include <arpa/inet.h>
 #include <net/if.h>
 #include <netinet/in.h>
+#include <netinet/ip.h>
 #include <netdb.h>
 #include <fcntl.h>
 #include <libgen.h>
@@ -27,6 +29,10 @@
 #include <time.h>
 #include <errno.h>
 
+#include <linux/xfrm.h>
+#include <linux/ipsec.h>
+#include <linux/pfkeyv2.h>
+
 #ifndef IPV6_UNICAST_IF
 #define IPV6_UNICAST_IF         76
 #endif
@@ -114,6 +120,9 @@ struct sock_args {
 		struct in_addr  in;
 		struct in6_addr in6;
 	} expected_raddr;
+
+	/* ESP in UDP encap test */
+	int use_xfrm;
 };
 
 static int server_mode;
@@ -1346,6 +1355,41 @@ static int bind_socket(int sd, struct sock_args *args)
 	return 0;
 }
 
+static int config_xfrm_policy(int sd, struct sock_args *args)
+{
+	struct xfrm_userpolicy_info policy = {};
+	int type = UDP_ENCAP_ESPINUDP;
+	int xfrm_af = IP_XFRM_POLICY;
+	int level = SOL_IP;
+
+	if (args->type != SOCK_DGRAM) {
+		log_error("Invalid socket type. Only DGRAM could be used for XFRM\n");
+		return 1;
+	}
+
+	policy.action = XFRM_POLICY_ALLOW;
+	policy.sel.family = args->version;
+	if (args->version == AF_INET6) {
+		xfrm_af = IPV6_XFRM_POLICY;
+		level = SOL_IPV6;
+	}
+
+	policy.dir = XFRM_POLICY_OUT;
+	if (setsockopt(sd, level, xfrm_af, &policy, sizeof(policy)) < 0)
+		return 1;
+
+	policy.dir = XFRM_POLICY_IN;
+	if (setsockopt(sd, level, xfrm_af, &policy, sizeof(policy)) < 0)
+		return 1;
+
+	if (setsockopt(sd, IPPROTO_UDP, UDP_ENCAP, &type, sizeof(type)) < 0) {
+		log_err_errno("Failed to set xfrm encap");
+		return 1;
+	}
+
+	return 0;
+}
+
 static int lsock_init(struct sock_args *args)
 {
 	long flags;
@@ -1389,6 +1433,11 @@ static int lsock_init(struct sock_args *args)
 	if (fcntl(sd, F_SETFD, FD_CLOEXEC) < 0)
 		log_err_errno("Failed to set close-on-exec flag");
 
+	if (args->use_xfrm && config_xfrm_policy(sd, args)) {
+		log_err_errno("Failed to set xfrm policy");
+		goto err;
+	}
+
 out:
 	return sd;
 
@@ -1772,7 +1821,7 @@ static int ipc_parent(int cpid, int fd, struct sock_args *args)
 	return client_status;
 }
 
-#define GETOPT_STR  "sr:l:c:p:t:g:P:DRn:M:X:m:d:I:BN:O:SCi6L:0:1:2:3:Fbq"
+#define GETOPT_STR  "sr:l:c:p:t:g:P:DRn:M:X:m:d:I:BN:O:SCi6xL:0:1:2:3:Fbq"
 
 static void print_usage(char *prog)
 {
@@ -1795,6 +1844,7 @@ static void print_usage(char *prog)
 	"    -D|R          datagram (D) / raw (R) socket (default stream)\n"
 	"    -l addr       local address to bind to in server mode\n"
 	"    -c addr       local address to bind to in client mode\n"
+	"    -x            configure XFRM policy on socket\n"
 	"\n"
 	"    -d dev        bind socket to given device name\n"
 	"    -I dev        bind socket to given device name - server mode\n"
@@ -1966,6 +2016,9 @@ int main(int argc, char *argv[])
 		case 'q':
 			quiet = 1;
 			break;
+		case 'x':
+			args.use_xfrm = 1;
+			break;
 		default:
 			print_usage(argv[0]);
 			return 1;