tools: selftests: add test for changing routes with PTMU exceptions

Adds new 2 new tests to the PTMU script: pmtu_ipv4/6_route_change.

These tests explicitly test for a recently discovered problem in the
IPv6 routing framework where PMTU exceptions were not properly released
when replacing a route via "ip route change ...".

After creating PMTU exceptions, the route from the device A to R1 will be
replaced with a new route, then device A will be deleted. If the PMTU
exceptions were properly cleaned up by the kernel, this device deletion
will succeed. Otherwise, the unregistration of the device will stall, and
messages such as the following will be logged in dmesg:

unregister_netdevice: waiting for veth_A-R1 to become free. Usage count = 4

Signed-off-by: Sean Tranchetti <[email protected]>
Reviewed-by: David Ahern <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

tools/testing/selftests/net/pmtu.sh

@@ -162,7 +162,15 @@
 # - list_flush_ipv6_exception
 #	Using the same topology as in pmtu_ipv6, create exceptions, and check
 #	they are shown when listing exception caches, gone after flushing them
-
+#
+# - pmtu_ipv4_route_change
+#	Use the same topology as in pmtu_ipv4, but issue a route replacement
+#	command and delete the corresponding device afterward. This tests for
+#	proper cleanup of the PMTU exceptions by the route replacement path.
+#	Device unregistration should complete successfully
+#
+# - pmtu_ipv6_route_change
+#	Same as above but with IPv6
 
 # Kselftest framework requirement - SKIP code is 4.
 ksft_skip=4
@@ -224,7 +232,9 @@ tests="
 	cleanup_ipv4_exception		ipv4: cleanup of cached exceptions	1
 	cleanup_ipv6_exception		ipv6: cleanup of cached exceptions	1
 	list_flush_ipv4_exception	ipv4: list and flush cached exceptions	1
-	list_flush_ipv6_exception	ipv6: list and flush cached exceptions	1"
+	list_flush_ipv6_exception	ipv6: list and flush cached exceptions	1
+	pmtu_ipv4_route_change		ipv4: PMTU exception w/route replace	1
+	pmtu_ipv6_route_change		ipv6: PMTU exception w/route replace	1"
 
 NS_A="ns-A"
 NS_B="ns-B"
@@ -1782,6 +1792,63 @@ test_list_flush_ipv6_exception() {
 	return ${fail}
 }
 
+test_pmtu_ipvX_route_change() {
+	family=${1}
+
+	setup namespaces routing || return 2
+	trace "${ns_a}"  veth_A-R1    "${ns_r1}" veth_R1-A \
+	      "${ns_r1}" veth_R1-B    "${ns_b}"  veth_B-R1 \
+	      "${ns_a}"  veth_A-R2    "${ns_r2}" veth_R2-A \
+	      "${ns_r2}" veth_R2-B    "${ns_b}"  veth_B-R2
+
+	if [ ${family} -eq 4 ]; then
+		ping=ping
+		dst1="${prefix4}.${b_r1}.1"
+		dst2="${prefix4}.${b_r2}.1"
+		gw="${prefix4}.${a_r1}.2"
+	else
+		ping=${ping6}
+		dst1="${prefix6}:${b_r1}::1"
+		dst2="${prefix6}:${b_r2}::1"
+		gw="${prefix6}:${a_r1}::2"
+	fi
+
+	# Set up initial MTU values
+	mtu "${ns_a}"  veth_A-R1 2000
+	mtu "${ns_r1}" veth_R1-A 2000
+	mtu "${ns_r1}" veth_R1-B 1400
+	mtu "${ns_b}"  veth_B-R1 1400
+
+	mtu "${ns_a}"  veth_A-R2 2000
+	mtu "${ns_r2}" veth_R2-A 2000
+	mtu "${ns_r2}" veth_R2-B 1500
+	mtu "${ns_b}"  veth_B-R2 1500
+
+	# Create route exceptions
+	run_cmd ${ns_a} ${ping} -q -M want -i 0.1 -w 1 -s 1800 ${dst1}
+	run_cmd ${ns_a} ${ping} -q -M want -i 0.1 -w 1 -s 1800 ${dst2}
+
+	# Check that exceptions have been created with the correct PMTU
+	pmtu_1="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst1})"
+	check_pmtu_value "1400" "${pmtu_1}" "exceeding MTU" || return 1
+	pmtu_2="$(route_get_dst_pmtu_from_exception "${ns_a}" ${dst2})"
+	check_pmtu_value "1500" "${pmtu_2}" "exceeding MTU" || return 1
+
+	# Replace the route from A to R1
+	run_cmd ${ns_a} ip route change default via ${gw}
+
+	# Delete the device in A
+	run_cmd ${ns_a} ip link del "veth_A-R1"
+}
+
+test_pmtu_ipv4_route_change() {
+	test_pmtu_ipvX_route_change 4
+}
+
+test_pmtu_ipv6_route_change() {
+	test_pmtu_ipvX_route_change 6
+}
+
 usage() {
 	echo
 	echo "$0 [OPTIONS] [TEST]..."