[PATCH net-next 0/8] mptcp: add mixed v4/v6 support for the in-kernel PM

[PATCH net-next 0/8] mptcp: add mixed v4/v6 support for the in-kernel PM

[Matthieu Baerts]

Before these patches, the in-kernel Path-Manager would not allow, for
the same MPTCP connection, having a mix of subflows in v4 and v6.

MPTCP's RFC 8684 doesn't forbid that and it is even recommended to do so
as the path in v4 and v6 are likely different. Some networks are also
v4 or v6 only, we cannot assume they all have both v4 and v6 support.

Patch 1 then removes this artificial constraint in the in-kernel PM
currently enforcing there are no mixed subflows in place, either in
address announcement or in subflow creation areas.

Patch 2 makes sure the sk_ipv6only attribute is also propagated to
subflows, just in case a new PM wouldn't respect it.

Some selftests have also been added for the in-kernel PM (patch 3).

Patches 4 to 8 are just some cleanups and small improvements in the
printed messages in the userspace PM. It is not linked to the rest but
identified when working on a related patch modifying this selftest,
already in -net:

  commit 4656d72c1efa ("selftests: mptcp: userspace: validate v4-v6 subflows mix")

---
Matthieu Baerts (6):
      mptcp: propagate sk_ipv6only to subflows
      mptcp: userspace pm: use a single point of exit
      selftests: mptcp: userspace: print titles
      selftests: mptcp: userspace: refactor asserts
      selftests: mptcp: userspace: print error details if any
      selftests: mptcp: userspace: avoid read errors

Paolo Abeni (2):
      mptcp: let the in-kernel PM use mixed IPv4 and IPv6 addresses
      selftests: mptcp: add test-cases for mixed v4/v6 subflows

 net/mptcp/pm_netlink.c                            |  58 ++++----
 net/mptcp/pm_userspace.c                          |   5 +-
 net/mptcp/sockopt.c                               |   1 +
 tools/testing/selftests/net/mptcp/mptcp_join.sh   |  53 ++++++--
 tools/testing/selftests/net/mptcp/userspace_pm.sh | 153 +++++++++++++---------
 5 files changed, 171 insertions(+), 99 deletions(-)
---
base-commit: 4373a023e0388fc19e27d37f61401bce6ff4c9d7
change-id: 20230123-upstream-net-next-pm-v4-v6-b186481a4b00

Best regards,
-- 
Matthieu Baerts <matthieu.baerts@tessares.net>

[PATCH net-next 1/8] mptcp: let the in-kernel PM use mixed IPv4 and IPv6 addresses

[Matthieu Baerts]

From: Paolo Abeni <pabeni@redhat.com>

Currently the in-kernel PM arbitrary enforces that created subflow's
family must match the main MPTCP socket while the RFC allows mixing
IPv4 and IPv6 subflows.

This patch changes the in-kernel PM logic to create subflows matching
the currently selected source (or destination) address. IPv4 sockets
can pick only IPv4 addresses (and v4 mapped in v6), while IPv6 sockets
not restricted to V6ONLY can pick either IPv4 and IPv6 addresses as
long as the source and destination matches.

A helper, previously introduced is used to ease family matching checks,
taking care of IPv4 vs IPv4-mapped-IPv6 vs IPv6 only addresses.

Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/269
Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
---
 net/mptcp/pm_netlink.c | 58 +++++++++++++++++++++++++++-----------------------
 1 file changed, 31 insertions(+), 27 deletions(-)

diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index b5505b8167f9..db07cc5b4fcb 100644
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -152,7 +152,6 @@ static struct mptcp_pm_addr_entry *
 select_local_address(const struct pm_nl_pernet *pernet,
 		     const struct mptcp_sock *msk)
 {
-	const struct sock *sk = (const struct sock *)msk;
 	struct mptcp_pm_addr_entry *entry, *ret = NULL;
 
 	msk_owned_by_me(msk);
@@ -165,16 +164,6 @@ select_local_address(const struct pm_nl_pernet *pernet,
 		if (!test_bit(entry->addr.id, msk->pm.id_avail_bitmap))
 			continue;
 
-		if (entry->addr.family != sk->sk_family) {
-#if IS_ENABLED(CONFIG_MPTCP_IPV6)
-			if ((entry->addr.family == AF_INET &&
-			     !ipv6_addr_v4mapped(&sk->sk_v6_daddr)) ||
-			    (sk->sk_family == AF_INET &&
-			     !ipv6_addr_v4mapped(&entry->addr.addr6)))
-#endif
-				continue;
-		}
-
 		ret = entry;
 		break;
 	}
@@ -423,7 +412,9 @@ static bool lookup_address_in_vec(const struct mptcp_addr_info *addrs, unsigned
 /* Fill all the remote addresses into the array addrs[],
  * and return the array size.
  */
-static unsigned int fill_remote_addresses_vec(struct mptcp_sock *msk, bool fullmesh,
+static unsigned int fill_remote_addresses_vec(struct mptcp_sock *msk,
+					      struct mptcp_addr_info *local,
+					      bool fullmesh,
 					      struct mptcp_addr_info *addrs)
 {
 	bool deny_id0 = READ_ONCE(msk->pm.remote_deny_join_id0);
@@ -443,6 +434,9 @@ static unsigned int fill_remote_addresses_vec(struct mptcp_sock *msk, bool fullm
 		if (deny_id0)
 			return 0;
 
+		if (!mptcp_pm_addr_families_match(sk, local, &remote))
+			return 0;
+
 		msk->pm.subflows++;
 		addrs[i++] = remote;
 	} else {
@@ -453,6 +447,9 @@ static unsigned int fill_remote_addresses_vec(struct mptcp_sock *msk, bool fullm
 			if (deny_id0 && !addrs[i].id)
 				continue;
 
+			if (!mptcp_pm_addr_families_match(sk, local, &addrs[i]))
+				continue;
+
 			if (!lookup_address_in_vec(addrs, i, &addrs[i]) &&
 			    msk->pm.subflows < subflows_max) {
 				msk->pm.subflows++;
@@ -603,9 +600,11 @@ static void mptcp_pm_create_subflow_or_signal_addr(struct mptcp_sock *msk)
 		fullmesh = !!(local->flags & MPTCP_PM_ADDR_FLAG_FULLMESH);
 
 		msk->pm.local_addr_used++;
-		nr = fill_remote_addresses_vec(msk, fullmesh, addrs);
-		if (nr)
-			__clear_bit(local->addr.id, msk->pm.id_avail_bitmap);
+		__clear_bit(local->addr.id, msk->pm.id_avail_bitmap);
+		nr = fill_remote_addresses_vec(msk, &local->addr, fullmesh, addrs);
+		if (nr == 0)
+			continue;
+
 		spin_unlock_bh(&msk->pm.lock);
 		for (i = 0; i < nr; i++)
 			__mptcp_subflow_connect(sk, &local->addr, &addrs[i]);
@@ -628,11 +627,11 @@ static void mptcp_pm_nl_subflow_established(struct mptcp_sock *msk)
  * and return the array size.
  */
 static unsigned int fill_local_addresses_vec(struct mptcp_sock *msk,
+					     struct mptcp_addr_info *remote,
 					     struct mptcp_addr_info *addrs)
 {
 	struct sock *sk = (struct sock *)msk;
 	struct mptcp_pm_addr_entry *entry;
-	struct mptcp_addr_info local;
 	struct pm_nl_pernet *pernet;
 	unsigned int subflows_max;
 	int i = 0;
@@ -645,15 +644,8 @@ static unsigned int fill_local_addresses_vec(struct mptcp_sock *msk,
 		if (!(entry->flags & MPTCP_PM_ADDR_FLAG_FULLMESH))
 			continue;
 
-		if (entry->addr.family != sk->sk_family) {
-#if IS_ENABLED(CONFIG_MPTCP_IPV6)
-			if ((entry->addr.family == AF_INET &&
-			     !ipv6_addr_v4mapped(&sk->sk_v6_daddr)) ||
-			    (sk->sk_family == AF_INET &&
-			     !ipv6_addr_v4mapped(&entry->addr.addr6)))
-#endif
-				continue;
-		}
+		if (!mptcp_pm_addr_families_match(sk, &entry->addr, remote))
+			continue;
 
 		if (msk->pm.subflows < subflows_max) {
 			msk->pm.subflows++;
@@ -666,8 +658,18 @@ static unsigned int fill_local_addresses_vec(struct mptcp_sock *msk,
 	 * 'IPADDRANY' local address
 	 */
 	if (!i) {
+		struct mptcp_addr_info local;
+
 		memset(&local, 0, sizeof(local));
-		local.family = msk->pm.remote.family;
+		local.family =
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+			       remote->family == AF_INET6 &&
+			       ipv6_addr_v4mapped(&remote->addr6) ? AF_INET :
+#endif
+			       remote->family;
+
+		if (!mptcp_pm_addr_families_match(sk, &local, remote))
+			return 0;
 
 		msk->pm.subflows++;
 		addrs[i++] = local;
@@ -706,7 +708,9 @@ static void mptcp_pm_nl_add_addr_received(struct mptcp_sock *msk)
 	/* connect to the specified remote address, using whatever
 	 * local address the routing configuration will pick.
 	 */
-	nr = fill_local_addresses_vec(msk, addrs);
+	nr = fill_local_addresses_vec(msk, &remote, addrs);
+	if (nr == 0)
+		return;
 
 	msk->pm.add_addr_accepted++;
 	if (msk->pm.add_addr_accepted >= add_addr_accept_max ||

-- 
2.38.1

[PATCH net-next 2/8] mptcp: propagate sk_ipv6only to subflows

[Matthieu Baerts]

Usually, attributes are propagated to subflows as well.

Here, if subflows are created by other ways than the MPTCP path-manager,
it is important to make sure they are in v6 if it is asked by the
userspace.

Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
---
 net/mptcp/sockopt.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c
index 582ed93bcc8a..9986681aaf40 100644
--- a/net/mptcp/sockopt.c
+++ b/net/mptcp/sockopt.c
@@ -1255,6 +1255,7 @@ static void sync_socket_options(struct mptcp_sock *msk, struct sock *ssk)
 	ssk->sk_priority = sk->sk_priority;
 	ssk->sk_bound_dev_if = sk->sk_bound_dev_if;
 	ssk->sk_incoming_cpu = sk->sk_incoming_cpu;
+	ssk->sk_ipv6only = sk->sk_ipv6only;
 	__ip_sock_set_tos(ssk, inet_sk(sk)->tos);
 
 	if (sk->sk_userlocks & tx_rx_locks) {

-- 
2.38.1

[PATCH net-next 3/8] selftests: mptcp: add test-cases for mixed v4/v6 subflows

[Matthieu Baerts]

From: Paolo Abeni <pabeni@redhat.com>

Note that we can't guess the listener family anymore based on the client
target address: always use IPv6.

The fullmesh flag with endpoints from different families is also
validated here.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
---
 tools/testing/selftests/net/mptcp/mptcp_join.sh | 53 ++++++++++++++++++++-----
 1 file changed, 44 insertions(+), 9 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh
index d11d3d566608..387abdcec011 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_join.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh
@@ -774,24 +774,17 @@ do_transfer()
 		addr_nr_ns2=${addr_nr_ns2:9}
 	fi
 
-	local local_addr
-	if is_v6 "${connect_addr}"; then
-		local_addr="::"
-	else
-		local_addr="0.0.0.0"
-	fi
-
 	extra_srv_args="$extra_args $extra_srv_args"
 	if [ "$test_link_fail" -gt 1 ];then
 		timeout ${timeout_test} \
 			ip netns exec ${listener_ns} \
 				./mptcp_connect -t ${timeout_poll} -l -p $port -s ${srv_proto} \
-					$extra_srv_args ${local_addr} < "$sinfail" > "$sout" &
+					$extra_srv_args "::" < "$sinfail" > "$sout" &
 	else
 		timeout ${timeout_test} \
 			ip netns exec ${listener_ns} \
 				./mptcp_connect -t ${timeout_poll} -l -p $port -s ${srv_proto} \
-					$extra_srv_args ${local_addr} < "$sin" > "$sout" &
+					$extra_srv_args "::" < "$sin" > "$sout" &
 	fi
 	local spid=$!
 
@@ -2448,6 +2441,47 @@ v4mapped_tests()
 	fi
 }
 
+mixed_tests()
+{
+	if reset "IPv4 sockets do not use IPv6 addresses"; then
+		pm_nl_set_limits $ns1 0 1
+		pm_nl_set_limits $ns2 1 1
+		pm_nl_add_endpoint $ns1 dead:beef:2::1 flags signal
+		run_tests $ns1 $ns2 10.0.1.1 0 0 0 slow
+		chk_join_nr 0 0 0
+	fi
+
+	# Need an IPv6 mptcp socket to allow subflows of both families
+	if reset "simult IPv4 and IPv6 subflows"; then
+		pm_nl_set_limits $ns1 0 1
+		pm_nl_set_limits $ns2 1 1
+		pm_nl_add_endpoint $ns1 10.0.1.1 flags signal
+		run_tests $ns1 $ns2 dead:beef:2::1 0 0 0 slow
+		chk_join_nr 1 1 1
+	fi
+
+	# cross families subflows will not be created even in fullmesh mode
+	if reset "simult IPv4 and IPv6 subflows, fullmesh 1x1"; then
+		pm_nl_set_limits $ns1 0 4
+		pm_nl_set_limits $ns2 1 4
+		pm_nl_add_endpoint $ns2 dead:beef:2::2 flags subflow,fullmesh
+		pm_nl_add_endpoint $ns1 10.0.1.1 flags signal
+		run_tests $ns1 $ns2 dead:beef:2::1 0 0 0 slow
+		chk_join_nr 1 1 1
+	fi
+
+	# fullmesh still tries to create all the possibly subflows with
+	# matching family
+	if reset "simult IPv4 and IPv6 subflows, fullmesh 2x2"; then
+		pm_nl_set_limits $ns1 0 4
+		pm_nl_set_limits $ns2 2 4
+		pm_nl_add_endpoint $ns1 10.0.2.1 flags signal
+		pm_nl_add_endpoint $ns1 dead:beef:2::1 flags signal
+		run_tests $ns1 $ns2 dead:beef:1::1 0 0 fullmesh_1 slow
+		chk_join_nr 4 4 4
+	fi
+}
+
 backup_tests()
 {
 	# single subflow, backup
@@ -3120,6 +3154,7 @@ all_tests_sorted=(
 	a@add_tests
 	6@ipv6_tests
 	4@v4mapped_tests
+	M@mixed_tests
 	b@backup_tests
 	p@add_addr_ports_tests
 	k@syncookies_tests

-- 
2.38.1

[PATCH net-next 4/8] mptcp: userspace pm: use a single point of exit

[Matthieu Baerts]

Like in all other functions in this file, a single point of exit is used
when extra operations are needed: unlock, decrement refcount, etc.

There is no functional change for the moment but it is better to do the
same here to make sure all cleanups are done in case of intermediate
errors.

Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
---
 net/mptcp/pm_userspace.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
index ea6ad9da7493..a02d3cbf2a1b 100644
--- a/net/mptcp/pm_userspace.c
+++ b/net/mptcp/pm_userspace.c
@@ -59,8 +59,8 @@ int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk,
 		 */
 		e = sock_kmalloc(sk, sizeof(*e), GFP_ATOMIC);
 		if (!e) {
-			spin_unlock_bh(&msk->pm.lock);
-			return -ENOMEM;
+			ret = -ENOMEM;
+			goto append_err;
 		}
 
 		*e = *entry;
@@ -74,6 +74,7 @@ int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk,
 		ret = entry->addr.id;
 	}
 
+append_err:
 	spin_unlock_bh(&msk->pm.lock);
 	return ret;
 }

-- 
2.38.1

[PATCH net-next 5/8] selftests: mptcp: userspace: print titles

[Matthieu Baerts]

This script is running a few tests after having setup the environment.

Printing titles helps understand what is being tested.

Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
---
 tools/testing/selftests/net/mptcp/userspace_pm.sh | 24 ++++++++++++++++++++++-
 1 file changed, 23 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/mptcp/userspace_pm.sh b/tools/testing/selftests/net/mptcp/userspace_pm.sh
index ab2d581f28a1..7b06d9d0aa46 100755
--- a/tools/testing/selftests/net/mptcp/userspace_pm.sh
+++ b/tools/testing/selftests/net/mptcp/userspace_pm.sh
@@ -43,6 +43,11 @@ rndh=$(printf %x "$sec")-$(mktemp -u XXXXXX)
 ns1="ns1-$rndh"
 ns2="ns2-$rndh"
 
+print_title()
+{
+	stdbuf -o0 -e0 printf "INFO: %s\n" "${1}"
+}
+
 kill_wait()
 {
 	kill $1 > /dev/null 2>&1
@@ -51,7 +56,7 @@ kill_wait()
 
 cleanup()
 {
-	echo "cleanup"
+	print_title "Cleanup"
 
 	rm -rf $file $client_evts $server_evts
 
@@ -78,6 +83,8 @@ cleanup()
 	for netns in "$ns1" "$ns2" ;do
 		ip netns del "$netns"
 	done
+
+	stdbuf -o0 -e0 printf "Done\n"
 }
 
 trap cleanup EXIT
@@ -108,6 +115,7 @@ ip -net "$ns2" addr add dead:beef:1::2/64 dev ns2eth1 nodad
 ip -net "$ns2" addr add dead:beef:2::2/64 dev ns2eth1 nodad
 ip -net "$ns2" link set ns2eth1 up
 
+print_title "Init"
 stdbuf -o0 -e0 printf "Created network namespaces ns1, ns2         \t\t\t[OK]\n"
 
 make_file()
@@ -255,6 +263,8 @@ verify_announce_event()
 
 test_announce()
 {
+	print_title "Announce tests"
+
 	# Capture events on the network namespace running the server
 	:>"$server_evts"
 
@@ -359,6 +369,8 @@ verify_remove_event()
 
 test_remove()
 {
+	print_title "Remove tests"
+
 	# Capture events on the network namespace running the server
 	:>"$server_evts"
 
@@ -521,6 +533,8 @@ verify_subflow_events()
 
 test_subflows()
 {
+	print_title "Subflows v4 or v6 only tests"
+
 	# Capture events on the network namespace running the server
 	:>"$server_evts"
 
@@ -754,6 +768,8 @@ test_subflows()
 
 test_subflows_v4_v6_mix()
 {
+	print_title "Subflows v4 and v6 mix tests"
+
 	# Attempt to add a listener at 10.0.2.1:<subflow-port>
 	ip netns exec "$ns1" ./pm_nl_ctl listen 10.0.2.1\
 	   $app6_port > /dev/null 2>&1 &
@@ -800,6 +816,8 @@ test_subflows_v4_v6_mix()
 
 test_prio()
 {
+	print_title "Prio tests"
+
 	local count
 
 	# Send MP_PRIO signal from client to server machine
@@ -876,6 +894,8 @@ verify_listener_events()
 
 test_listener()
 {
+	print_title "Listener tests"
+
 	# Capture events on the network namespace running the client
 	:>$client_evts
 
@@ -902,8 +922,10 @@ test_listener()
 	verify_listener_events $client_evts $LISTENER_CLOSED $AF_INET 10.0.2.2 $client4_port
 }
 
+print_title "Make connections"
 make_connection
 make_connection "v6"
+
 test_announce
 test_remove
 test_subflows

-- 
2.38.1

[PATCH net-next 6/8] selftests: mptcp: userspace: refactor asserts

[Matthieu Baerts]

Instead of having a long list of conditions to check, it is possible to
give a list of variable names to compare with their 'e_XXX' version.

This will ease the introduction of the following commit which will print
which condition has failed (if any).

Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
---
 tools/testing/selftests/net/mptcp/userspace_pm.sh | 72 +++++++++++------------
 1 file changed, 36 insertions(+), 36 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/userspace_pm.sh b/tools/testing/selftests/net/mptcp/userspace_pm.sh
index 7b06d9d0aa46..2f2a85a212b0 100755
--- a/tools/testing/selftests/net/mptcp/userspace_pm.sh
+++ b/tools/testing/selftests/net/mptcp/userspace_pm.sh
@@ -225,6 +225,36 @@ make_connection()
 	fi
 }
 
+# $1: var name
+check_expected_one()
+{
+	local var="${1}"
+	local exp="e_${var}"
+
+	[ "${!var}" = "${!exp}" ]
+}
+
+# $@: all var names to check
+check_expected()
+{
+	local ret=0
+	local var
+
+	for var in "${@}"
+	do
+		check_expected_one "${var}" || ret=1
+	done
+
+	if [ ${ret} -eq 0 ]
+	then
+		stdbuf -o0 -e0 printf "[OK]\n"
+		return 0
+	fi
+
+	stdbuf -o0 -e0 printf "[FAIL]\n"
+	exit 1
+}
+
 verify_announce_event()
 {
 	local evt=$1
@@ -250,15 +280,8 @@ verify_announce_event()
 	fi
 	dport=$(sed --unbuffered -n 's/.*\(dport:\)\([[:digit:]]*\).*$/\2/p;q' "$evt")
 	id=$(sed --unbuffered -n 's/.*\(rem_id:\)\([[:digit:]]*\).*$/\2/p;q' "$evt")
-	if [ "$type" = "$e_type" ] && [ "$token" = "$e_token" ] &&
-		   [ "$addr" = "$e_addr" ] && [ "$dport" = "$e_dport" ] &&
-		   [ "$id" = "$e_id" ]
-	then
-		stdbuf -o0 -e0 printf "[OK]\n"
-		return 0
-	fi
-	stdbuf -o0 -e0 printf "[FAIL]\n"
-	exit 1
+
+	check_expected "type" "token" "addr" "dport" "id"
 }
 
 test_announce()
@@ -357,14 +380,8 @@ verify_remove_event()
 	type=$(sed --unbuffered -n 's/.*\(type:\)\([[:digit:]]*\).*$/\2/p;q' "$evt")
 	token=$(sed --unbuffered -n 's/.*\(token:\)\([[:digit:]]*\).*$/\2/p;q' "$evt")
 	id=$(sed --unbuffered -n 's/.*\(rem_id:\)\([[:digit:]]*\).*$/\2/p;q' "$evt")
-	if [ "$type" = "$e_type" ] && [ "$token" = "$e_token" ] &&
-		   [ "$id" = "$e_id" ]
-	then
-		stdbuf -o0 -e0 printf "[OK]\n"
-		return 0
-	fi
-	stdbuf -o0 -e0 printf "[FAIL]\n"
-	exit 1
+
+	check_expected "type" "token" "id"
 }
 
 test_remove()
@@ -519,16 +536,7 @@ verify_subflow_events()
 		daddr=$(sed --unbuffered -n 's/.*\(daddr4:\)\([0-9.]*\).*$/\2/p;q' "$evt")
 	fi
 
-	if [ "$type" = "$e_type" ] && [ "$token" = "$e_token" ] &&
-		   [ "$daddr" = "$e_daddr" ] && [ "$e_dport" = "$dport" ] &&
-		   [ "$family" = "$e_family" ] && [ "$saddr" = "$e_saddr" ] &&
-		   [ "$e_locid" = "$locid" ] && [ "$e_remid" = "$remid" ]
-	then
-		stdbuf -o0 -e0 printf "[OK]\n"
-		return 0
-	fi
-	stdbuf -o0 -e0 printf "[FAIL]\n"
-	exit 1
+	check_expected "type" "token" "daddr" "dport" "family" "saddr" "locid" "remid"
 }
 
 test_subflows()
@@ -881,15 +889,7 @@ verify_listener_events()
 			sed --unbuffered -n 's/.*\(saddr4:\)\([0-9.]*\).*$/\2/p;q')
 	fi
 
-	if [ $type ] && [ $type = $e_type ] &&
-	   [ $family ] && [ $family = $e_family ] &&
-	   [ $saddr ] && [ $saddr = $e_saddr ] &&
-	   [ $sport ] && [ $sport = $e_sport ]; then
-		stdbuf -o0 -e0 printf "[OK]\n"
-		return 0
-	fi
-	stdbuf -o0 -e0 printf "[FAIL]\n"
-	exit 1
+	check_expected "type" "family" "saddr" "sport"
 }
 
 test_listener()

-- 
2.38.1

[PATCH net-next 7/8] selftests: mptcp: userspace: print error details if any

[Matthieu Baerts]

Before, only '[FAIL]' was printed in case of error during the validation
phase.

Now, in case of failure, the variable name, its value and expected one
are displayed to help understand what was wrong.

Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
---
 tools/testing/selftests/net/mptcp/userspace_pm.sh | 33 +++++++++++++++++------
 1 file changed, 25 insertions(+), 8 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/userspace_pm.sh b/tools/testing/selftests/net/mptcp/userspace_pm.sh
index 2f2a85a212b0..259382ad552c 100755
--- a/tools/testing/selftests/net/mptcp/userspace_pm.sh
+++ b/tools/testing/selftests/net/mptcp/userspace_pm.sh
@@ -201,11 +201,16 @@ make_connection()
 	server_serverside=$(grep "type:1," "$server_evts" |
 			    sed --unbuffered -n 's/.*\(server_side:\)\([[:digit:]]*\).*$/\2/p;q')
 
+	stdbuf -o0 -e0 printf "Established IP%s MPTCP Connection ns2 => ns1    \t\t" $is_v6
 	if [ "$client_token" != "" ] && [ "$server_token" != "" ] && [ "$client_serverside" = 0 ] &&
 		   [ "$server_serverside" = 1 ]
 	then
-		stdbuf -o0 -e0 printf "Established IP%s MPTCP Connection ns2 => ns1    \t\t[OK]\n" $is_v6
+		stdbuf -o0 -e0 printf "[OK]\n"
 	else
+		stdbuf -o0 -e0 printf "[FAIL]\n"
+		stdbuf -o0 -e0 printf "\tExpected tokens (c:%s - s:%s) and server (c:%d - s:%d)\n" \
+			"${client_token}" "${server_token}" \
+			"${client_serverside}" "${server_serverside}"
 		exit 1
 	fi
 
@@ -225,13 +230,26 @@ make_connection()
 	fi
 }
 
-# $1: var name
+# $1: var name ; $2: prev ret
 check_expected_one()
 {
 	local var="${1}"
 	local exp="e_${var}"
+	local prev_ret="${2}"
 
-	[ "${!var}" = "${!exp}" ]
+	if [ "${!var}" = "${!exp}" ]
+	then
+		return 0
+	fi
+
+	if [ "${prev_ret}" = "0" ]
+	then
+		stdbuf -o0 -e0 printf "[FAIL]\n"
+	fi
+
+	stdbuf -o0 -e0 printf "\tExpected value for '%s': '%s', got '%s'.\n" \
+		"${var}" "${!var}" "${!exp}"
+	return 1
 }
 
 # $@: all var names to check
@@ -242,7 +260,7 @@ check_expected()
 
 	for var in "${@}"
 	do
-		check_expected_one "${var}" || ret=1
+		check_expected_one "${var}" "${ret}" || ret=1
 	done
 
 	if [ ${ret} -eq 0 ]
@@ -251,7 +269,6 @@ check_expected()
 		return 0
 	fi
 
-	stdbuf -o0 -e0 printf "[FAIL]\n"
 	exit 1
 }
 
@@ -303,7 +320,7 @@ test_announce()
 	then
 		stdbuf -o0 -e0 printf "[OK]\n"
 	else
-		stdbuf -o0 -e0 printf "[FAIL]\n"
+		stdbuf -o0 -e0 printf "[FAIL]\n\ttype defined: %s\n" "${type}"
 		exit 1
 	fi
 
@@ -837,7 +854,7 @@ test_prio()
 	count=$(ip netns exec "$ns2" nstat -as | grep MPTcpExtMPPrioTx | awk '{print $2}')
 	[ -z "$count" ] && count=0
 	if [ $count != 1 ]; then
-		stdbuf -o0 -e0 printf "[FAIL]\n"
+		stdbuf -o0 -e0 printf "[FAIL]\n\tCount != 1: %d\n" "${count}"
 		exit 1
 	else
 		stdbuf -o0 -e0 printf "[OK]\n"
@@ -848,7 +865,7 @@ test_prio()
 	count=$(ip netns exec "$ns1" nstat -as | grep MPTcpExtMPPrioRx | awk '{print $2}')
 	[ -z "$count" ] && count=0
 	if [ $count != 1 ]; then
-		stdbuf -o0 -e0 printf "[FAIL]\n"
+		stdbuf -o0 -e0 printf "[FAIL]\n\tCount != 1: %d\n" "${count}"
 		exit 1
 	else
 		stdbuf -o0 -e0 printf "[OK]\n"

-- 
2.38.1

[PATCH net-next 8/8] selftests: mptcp: userspace: avoid read errors

[Matthieu Baerts]

During the cleanup phase, the server pids were killed with a SIGTERM
directly, not using a SIGUSR1 first to quit safely. As a result, this
test was often ending with two error messages:

  read: Connection reset by peer

While at it, use a for-loop to terminate all the PIDs the same way.

Also the different files are now removed after having killed the PIDs
using them. It makes more sense to do that in this order.

Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
---
 tools/testing/selftests/net/mptcp/userspace_pm.sh | 32 +++++++++--------------
 1 file changed, 12 insertions(+), 20 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/userspace_pm.sh b/tools/testing/selftests/net/mptcp/userspace_pm.sh
index 259382ad552c..66c5be25c13d 100755
--- a/tools/testing/selftests/net/mptcp/userspace_pm.sh
+++ b/tools/testing/selftests/net/mptcp/userspace_pm.sh
@@ -50,6 +50,9 @@ print_title()
 
 kill_wait()
 {
+	[ $1 -eq 0 ] && return 0
+
+	kill -SIGUSR1 $1 > /dev/null 2>&1
 	kill $1 > /dev/null 2>&1
 	wait $1 2>/dev/null
 }
@@ -58,32 +61,21 @@ cleanup()
 {
 	print_title "Cleanup"
 
-	rm -rf $file $client_evts $server_evts
-
 	# Terminate the MPTCP connection and related processes
-	if [ $client4_pid -ne 0 ]; then
-		kill -SIGUSR1 $client4_pid > /dev/null 2>&1
-	fi
-	if [ $server4_pid -ne 0 ]; then
-		kill_wait $server4_pid
-	fi
-	if [ $client6_pid -ne 0 ]; then
-		kill -SIGUSR1 $client6_pid > /dev/null 2>&1
-	fi
-	if [ $server6_pid -ne 0 ]; then
-		kill_wait $server6_pid
-	fi
-	if [ $server_evts_pid -ne 0 ]; then
-		kill_wait $server_evts_pid
-	fi
-	if [ $client_evts_pid -ne 0 ]; then
-		kill_wait $client_evts_pid
-	fi
+	local pid
+	for pid in $client4_pid $server4_pid $client6_pid $server6_pid\
+		   $server_evts_pid $client_evts_pid
+	do
+		kill_wait $pid
+	done
+
 	local netns
 	for netns in "$ns1" "$ns2" ;do
 		ip netns del "$netns"
 	done
 
+	rm -rf $file $client_evts $server_evts
+
 	stdbuf -o0 -e0 printf "Done\n"
 }
 

-- 
2.38.1

Re: [PATCH net-next 0/8] mptcp: add mixed v4/v6 support for the in-kernel PM

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (master)
by Paolo Abeni <pabeni@redhat.com>:

On Wed, 25 Jan 2023 11:47:20 +0100 you wrote:
> Before these patches, the in-kernel Path-Manager would not allow, for
> the same MPTCP connection, having a mix of subflows in v4 and v6.
> 
> MPTCP's RFC 8684 doesn't forbid that and it is even recommended to do so
> as the path in v4 and v6 are likely different. Some networks are also
> v4 or v6 only, we cannot assume they all have both v4 and v6 support.
> 
> [...]

Here is the summary with links:
  - [net-next,1/8] mptcp: let the in-kernel PM use mixed IPv4 and IPv6 addresses
    https://git.kernel.org/netdev/net-next/c/b9d69db87fb7
  - [net-next,2/8] mptcp: propagate sk_ipv6only to subflows
    https://git.kernel.org/netdev/net-next/c/7e9740e0e84e
  - [net-next,3/8] selftests: mptcp: add test-cases for mixed v4/v6 subflows
    https://git.kernel.org/netdev/net-next/c/ad3493746ebe
  - [net-next,4/8] mptcp: userspace pm: use a single point of exit
    https://git.kernel.org/netdev/net-next/c/40c71f763f87
  - [net-next,5/8] selftests: mptcp: userspace: print titles
    https://git.kernel.org/netdev/net-next/c/f790ae03db33
  - [net-next,6/8] selftests: mptcp: userspace: refactor asserts
    https://git.kernel.org/netdev/net-next/c/1c0b0ee2640b
  - [net-next,7/8] selftests: mptcp: userspace: print error details if any
    https://git.kernel.org/netdev/net-next/c/10d4273411be
  - [net-next,8/8] selftests: mptcp: userspace: avoid read errors
    https://git.kernel.org/netdev/net-next/c/8dbdf24f4e9e

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html