Merge branch 'selftests-forwarding-Add-new-test-cases'

Ido Schimmel says:

====================
selftests: forwarding: Add new test cases

This patchset mainly adds new forwarding test cases and performs small
changes in existing infrastructure.

Patches #1-#3 add new test cases for multicast RPF check, PCP and VLAN
matching using flower and tc VLAN modify action.

The rest of the patches are from Petr who says:

In patches #4 and #5, devlink_lib.sh is fixed to first not cause double
inclusion of lib.sh, and then to deduce the device name in a simpler way.

In patch #6, helpers for dealing with shared buffer configuration are
added to devlink_lib.sh.

In patch #7, MC-awareness test is fixed to configure shared buffers
explicitly.

In patch #8, several helpers are extracted from the MC-awareness test
and put into a new mlxsw-specific library, qos_lib.sh.

In patch #9, a new test is added which checks configuration of
strictly-prioritized streams.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

tools/testing/selftests/drivers/net/mlxsw/qos_ets_strict.sh

+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+#
+# A test for strict prioritization of traffic in the switch. Run two streams of
+# traffic, each through a different ingress port, one tagged with PCP of 1, the
+# other with PCP of 2. Both streams converge at one egress port, where they are
+# assigned TC of, respectively, 1 and 2, with strict priority configured between
+# them. In H3, we expect to see (almost) exclusively the high-priority traffic.
+#
+# Please see qos_mc_aware.sh for an explanation of why we use mausezahn and
+# counters instead of just running iperf3.
+#
+# +---------------------------+                 +-----------------------------+
+# | H1                        |                 |                          H2 |
+# |         $h1.111 +         |                 |         + $h2.222           |
+# |   192.0.2.33/28 |         |                 |         | 192.0.2.65/28     |
+# |   e-qos-map 0:1 |         |                 |         | e-qos-map 0:2     |
+# |                 |         |                 |         |                   |
+# |             $h1 +         |                 |         + $h2               |
+# +-----------------|---------+                 +---------|-------------------+
+#                   |                                     |
+# +-----------------|-------------------------------------|-------------------+
+# |           $swp1 +                                     + $swp2             |
+# |          >1Gbps |                                     | >1Gbps            |
+# | +---------------|-----------+              +----------|----------------+  |
+# | |     $swp1.111 +           |              |          + $swp2.222      |  |
+# | |                     BR111 |       SW     | BR222                     |  |
+# | |     $swp3.111 +           |              |          + $swp3.222      |  |
+# | +---------------|-----------+              +----------|----------------+  |
+# |                 \_____________________________________/                   |
+# |                                    |                                      |
+# |                                    + $swp3                                |
+# |                                    | 1Gbps bottleneck                     |
+# |                                    | ETS: (up n->tc n for n in 0..7)      |
+# |                                    |      strict priority                 |
+# +------------------------------------|--------------------------------------+
+#                                      |
+#                 +--------------------|--------------------+
+#                 |                    + $h3             H3 |
+#                 |                   / \                   |
+#                 |                  /   \                  |
+#                 |         $h3.111 +     + $h3.222         |
+#                 |  192.0.2.34/28          192.0.2.66/28   |
+#                 +-----------------------------------------+
+
+ALL_TESTS="
+	ping_ipv4
+	test_ets_strict
+"
+
+lib_dir=$(dirname $0)/../../../net/forwarding
+
+NUM_NETIFS=6
+source $lib_dir/lib.sh
+source $lib_dir/devlink_lib.sh
+source qos_lib.sh
+
+h1_create()
+{
+	simple_if_init $h1
+	mtu_set $h1 10000
+
+	vlan_create $h1 111 v$h1 192.0.2.33/28
+	ip link set dev $h1.111 type vlan egress-qos-map 0:1
+}
+
+h1_destroy()
+{
+	vlan_destroy $h1 111
+
+	mtu_restore $h1
+	simple_if_fini $h1
+}
+
+h2_create()
+{
+	simple_if_init $h2
+	mtu_set $h2 10000
+
+	vlan_create $h2 222 v$h2 192.0.2.65/28
+	ip link set dev $h2.222 type vlan egress-qos-map 0:2
+}
+
+h2_destroy()
+{
+	vlan_destroy $h2 222
+
+	mtu_restore $h2
+	simple_if_fini $h2
+}
+
+h3_create()
+{
+	simple_if_init $h3
+	mtu_set $h3 10000
+
+	vlan_create $h3 111 v$h3 192.0.2.34/28
+	vlan_create $h3 222 v$h3 192.0.2.66/28
+}
+
+h3_destroy()
+{
+	vlan_destroy $h3 222
+	vlan_destroy $h3 111
+
+	mtu_restore $h3
+	simple_if_fini $h3
+}
+
+switch_create()
+{
+	ip link set dev $swp1 up
+	mtu_set $swp1 10000
+
+	ip link set dev $swp2 up
+	mtu_set $swp2 10000
+
+	# prio n -> TC n, strict scheduling
+	lldptool -T -i $swp3 -V ETS-CFG up2tc=0:0,1:1,2:2,3:3,4:4,5:5,6:6,7:7
+	lldptool -T -i $swp3 -V ETS-CFG tsa=$(
+			)"0:strict,"$(
+			)"1:strict,"$(
+			)"2:strict,"$(
+			)"3:strict,"$(
+			)"4:strict,"$(
+			)"5:strict,"$(
+			)"6:strict,"$(
+			)"7:strict"
+	sleep 1
+
+	ip link set dev $swp3 up
+	mtu_set $swp3 10000
+	ethtool -s $swp3 speed 1000 autoneg off
+
+	vlan_create $swp1 111
+	vlan_create $swp2 222
+	vlan_create $swp3 111
+	vlan_create $swp3 222
+
+	ip link add name br111 up type bridge vlan_filtering 0
+	ip link set dev $swp1.111 master br111
+	ip link set dev $swp3.111 master br111
+
+	ip link add name br222 up type bridge vlan_filtering 0
+	ip link set dev $swp2.222 master br222
+	ip link set dev $swp3.222 master br222
+
+	# Make sure that ingress quotas are smaller than egress so that there is
+	# room for both streams of traffic to be admitted to shared buffer.
+	devlink_pool_size_thtype_set 0 dynamic 10000000
+	devlink_pool_size_thtype_set 4 dynamic 10000000
+
+	devlink_port_pool_th_set $swp1 0 6
+	devlink_tc_bind_pool_th_set $swp1 1 ingress 0 6
+
+	devlink_port_pool_th_set $swp2 0 6
+	devlink_tc_bind_pool_th_set $swp2 2 ingress 0 6
+
+	devlink_tc_bind_pool_th_set $swp3 1 egress 4 7
+	devlink_tc_bind_pool_th_set $swp3 2 egress 4 7
+	devlink_port_pool_th_set $swp3 4 7
+}
+
+switch_destroy()
+{
+	devlink_port_pool_th_restore $swp3 4
+	devlink_tc_bind_pool_th_restore $swp3 2 egress
+	devlink_tc_bind_pool_th_restore $swp3 1 egress
+
+	devlink_tc_bind_pool_th_restore $swp2 2 ingress
+	devlink_port_pool_th_restore $swp2 0
+
+	devlink_tc_bind_pool_th_restore $swp1 1 ingress
+	devlink_port_pool_th_restore $swp1 0
+
+	devlink_pool_size_thtype_restore 4
+	devlink_pool_size_thtype_restore 0
+
+	ip link del dev br222
+	ip link del dev br111
+
+	vlan_destroy $swp3 222
+	vlan_destroy $swp3 111
+	vlan_destroy $swp2 222
+	vlan_destroy $swp1 111
+
+	ethtool -s $swp3 autoneg on
+	mtu_restore $swp3
+	ip link set dev $swp3 down
+	lldptool -T -i $swp3 -V ETS-CFG up2tc=0:0,1:0,2:0,3:0,4:0,5:0,6:0,7:0
+
+	mtu_restore $swp2
+	ip link set dev $swp2 down
+
+	mtu_restore $swp1
+	ip link set dev $swp1 down
+}
+
+setup_prepare()
+{
+	h1=${NETIFS[p1]}
+	swp1=${NETIFS[p2]}
+
+	swp2=${NETIFS[p3]}
+	h2=${NETIFS[p4]}
+
+	swp3=${NETIFS[p5]}
+	h3=${NETIFS[p6]}
+
+	h3mac=$(mac_get $h3)
+
+	vrf_prepare
+
+	h1_create
+	h2_create
+	h3_create
+	switch_create
+}
+
+cleanup()
+{
+	pre_cleanup
+
+	switch_destroy
+	h3_destroy
+	h2_destroy
+	h1_destroy
+
+	vrf_cleanup
+}
+
+ping_ipv4()
+{
+	ping_test $h1 192.0.2.34 " from H1"
+	ping_test $h2 192.0.2.66 " from H2"
+}
+
+rel()
+{
+	local old=$1; shift
+	local new=$1; shift
+
+	bc <<< "
+	    scale=2
+	    ret = 100 * $new / $old
+	    if (ret > 0) { ret } else { 0 }
+	"
+}
+
+test_ets_strict()
+{
+	RET=0
+
+	# Run high-prio traffic on its own.
+	start_traffic $h2.222 192.0.2.65 192.0.2.66 $h3mac
+	local -a rate_2
+	rate_2=($(measure_rate $swp2 $h3 rx_octets_prio_2 "prio 2"))
+	check_err $? "Could not get high enough prio-2 ingress rate"
+	local rate_2_in=${rate_2[0]}
+	local rate_2_eg=${rate_2[1]}
+	stop_traffic # $h2.222
+
+	# Start low-prio stream.
+	start_traffic $h1.111 192.0.2.33 192.0.2.34 $h3mac
+
+	local -a rate_1
+	rate_1=($(measure_rate $swp1 $h3 rx_octets_prio_1 "prio 1"))
+	check_err $? "Could not get high enough prio-1 ingress rate"
+	local rate_1_in=${rate_1[0]}
+	local rate_1_eg=${rate_1[1]}
+
+	# High-prio and low-prio on their own should have about the same
+	# throughput.
+	local rel21=$(rel $rate_1_eg $rate_2_eg)
+	check_err $(bc <<< "$rel21 < 95")
+	check_err $(bc <<< "$rel21 > 105")
+
+	# Start the high-prio stream--now both streams run.
+	start_traffic $h2.222 192.0.2.65 192.0.2.66 $h3mac
+	rate_3=($(measure_rate $swp2 $h3 rx_octets_prio_2 "prio 2 w/ 1"))
+	check_err $? "Could not get high enough prio-2 ingress rate with prio-1"
+	local rate_3_in=${rate_3[0]}
+	local rate_3_eg=${rate_3[1]}
+	stop_traffic # $h2.222
+
+	stop_traffic # $h1.111
+
+	# High-prio should have about the same throughput whether or not
+	# low-prio is in the system.
+	local rel32=$(rel $rate_2_eg $rate_3_eg)
+	check_err $(bc <<< "$rel32 < 95")
+
+	log_test "strict priority"
+	echo "Ingress to switch:"
+	echo "  p1 in rate            $(humanize $rate_1_in)"
+	echo "  p2 in rate            $(humanize $rate_2_in)"
+	echo "  p2 in rate w/ p1      $(humanize $rate_3_in)"
+	echo "Egress from switch:"
+	echo "  p1 eg rate            $(humanize $rate_1_eg)"
+	echo "  p2 eg rate            $(humanize $rate_2_eg) ($rel21% of p1)"
+	echo "  p2 eg rate w/ p1      $(humanize $rate_3_eg) ($rel32% of p2)"
+}
+
+trap cleanup EXIT
+
+setup_prepare
+setup_wait
+
+tests_run
+
+exit $EXIT_STATUS

tools/testing/selftests/drivers/net/mlxsw/qos_lib.sh

+# SPDX-License-Identifier: GPL-2.0
+
+humanize()
+{
+	local speed=$1; shift
+
+	for unit in bps Kbps Mbps Gbps; do
+		if (($(echo "$speed < 1024" | bc))); then
+			break
+		fi
+
+		speed=$(echo "scale=1; $speed / 1024" | bc)
+	done
+
+	echo "$speed${unit}"
+}
+
+rate()
+{
+	local t0=$1; shift
+	local t1=$1; shift
+	local interval=$1; shift
+
+	echo $((8 * (t1 - t0) / interval))
+}
+
+start_traffic()
+{
+	local h_in=$1; shift    # Where the traffic egresses the host
+	local sip=$1; shift
+	local dip=$1; shift
+	local dmac=$1; shift
+
+	$MZ $h_in -p 8000 -A $sip -B $dip -c 0 \
+		-a own -b $dmac -t udp -q &
+	sleep 1
+}
+
+stop_traffic()
+{
+	# Suppress noise from killing mausezahn.
+	{ kill %% && wait %%; } 2>/dev/null
+}
+
+check_rate()
+{
+	local rate=$1; shift
+	local min=$1; shift
+	local what=$1; shift
+
+	if ((rate > min)); then
+		return 0
+	fi
+
+	echo "$what $(humanize $ir) < $(humanize $min)" > /dev/stderr
+	return 1
+}
+
+measure_rate()
+{
+	local sw_in=$1; shift   # Where the traffic ingresses the switch
+	local host_in=$1; shift # Where it ingresses another host
+	local counter=$1; shift # Counter to use for measurement
+	local what=$1; shift
+
+	local interval=10
+	local i
+	local ret=0
+
+	# Dips in performance might cause momentary ingress rate to drop below
+	# 1Gbps. That wouldn't saturate egress and MC would thus get through,
+	# seemingly winning bandwidth on account of UC. Demand at least 2Gbps
+	# average ingress rate to somewhat mitigate this.
+	local min_ingress=2147483648
+
+	for i in {5..0}; do
+		local t0=$(ethtool_stats_get $host_in $counter)
+		local u0=$(ethtool_stats_get $sw_in $counter)
+		sleep $interval
+		local t1=$(ethtool_stats_get $host_in $counter)
+		local u1=$(ethtool_stats_get $sw_in $counter)
+
+		local ir=$(rate $u0 $u1 $interval)
+		local er=$(rate $t0 $t1 $interval)
+
+		if check_rate $ir $min_ingress "$what ingress rate"; then
+			break
+		fi
+
+		# Fail the test if we can't get the throughput.
+		if ((i == 0)); then
+			ret=1
+		fi
+	done
+
+	echo $ir $er
+	return $ret
+}

tools/testing/selftests/drivers/net/mlxsw/qos_mc_aware.sh

@@ -67,6 +67,8 @@ lib_dir=$(dirname $0)/../../../net/forwarding
 
 NUM_NETIFS=6
 source $lib_dir/lib.sh
+source $lib_dir/devlink_lib.sh
+source qos_lib.sh
 
 h1_create()
 {
@@ -140,10 +142,28 @@ switch_create()
 	ip link set dev br111 up
 	ip link set dev $swp2.111 master br111
 	ip link set dev $swp3.111 master br111
+
+	# Make sure that ingress quotas are smaller than egress so that there is
+	# room for both streams of traffic to be admitted to shared buffer.
+	devlink_port_pool_th_set $swp1 0 5
+	devlink_tc_bind_pool_th_set $swp1 0 ingress 0 5
+
+	devlink_port_pool_th_set $swp2 0 5
+	devlink_tc_bind_pool_th_set $swp2 1 ingress 0 5
+
+	devlink_port_pool_th_set $swp3 4 12
 }
 
 switch_destroy()
 {
+	devlink_port_pool_th_restore $swp3 4
+
+	devlink_tc_bind_pool_th_restore $swp2 1 ingress
+	devlink_port_pool_th_restore $swp2 0
+
+	devlink_tc_bind_pool_th_restore $swp1 0 ingress
+	devlink_port_pool_th_restore $swp1 0
+
 	ip link del dev br111
 	ip link del dev br1
 
@@ -201,107 +221,28 @@ ping_ipv4()
 	ping_test $h2 192.0.2.130
 }
 
-humanize()
-{
-	local speed=$1; shift
-
-	for unit in bps Kbps Mbps Gbps; do
-		if (($(echo "$speed < 1024" | bc))); then
-			break
-		fi
-
-		speed=$(echo "scale=1; $speed / 1024" | bc)
-	done
-
-	echo "$speed${unit}"
-}
-
-rate()
-{
-	local t0=$1; shift
-	local t1=$1; shift
-	local interval=$1; shift
-
-	echo $((8 * (t1 - t0) / interval))
-}
-
-check_rate()
-{
-	local rate=$1; shift
-	local min=$1; shift
-	local what=$1; shift
-
-	if ((rate > min)); then
-		return 0
-	fi
-
-	echo "$what $(humanize $ir) < $(humanize $min_ingress)" > /dev/stderr
-	return 1
-}
-
-measure_uc_rate()
-{
-	local what=$1; shift
-
-	local interval=10
-	local i
-	local ret=0
-
-	# Dips in performance might cause momentary ingress rate to drop below
-	# 1Gbps. That wouldn't saturate egress and MC would thus get through,
-	# seemingly winning bandwidth on account of UC. Demand at least 2Gbps
-	# average ingress rate to somewhat mitigate this.
-	local min_ingress=2147483648
-
-	$MZ $h2.111 -p 8000 -A 192.0.2.129 -B 192.0.2.130 -c 0 \
-		-a own -b $h3mac -t udp -q &
-	sleep 1
-
-	for i in {5..0}; do
-		local t0=$(ethtool_stats_get $h3 rx_octets_prio_1)
-		local u0=$(ethtool_stats_get $swp2 rx_octets_prio_1)
-		sleep $interval
-		local t1=$(ethtool_stats_get $h3 rx_octets_prio_1)
-		local u1=$(ethtool_stats_get $swp2 rx_octets_prio_1)
-
-		local ir=$(rate $u0 $u1 $interval)
-		local er=$(rate $t0 $t1 $interval)
-
-		if check_rate $ir $min_ingress "$what ingress rate"; then
-			break
-		fi
-
-		# Fail the test if we can't get the throughput.
-		if ((i == 0)); then
-			ret=1
-		fi
-	done
-
-	# Suppress noise from killing mausezahn.
-	{ kill %% && wait; } 2>/dev/null
-
-	echo $ir $er
-	exit $ret
-}
-
 test_mc_aware()
 {
 	RET=0
 
 	local -a uc_rate
-	uc_rate=($(measure_uc_rate "UC-only"))
+	start_traffic $h2.111 192.0.2.129 192.0.2.130 $h3mac
+	uc_rate=($(measure_rate $swp2 $h3 rx_octets_prio_1 "UC-only"))
 	check_err $? "Could not get high enough UC-only ingress rate"
+	stop_traffic
 	local ucth1=${uc_rate[1]}
 
-	$MZ $h1 -p 8000 -c 0 -a own -b bc -t udp -q &
+	start_traffic $h1 own bc bc
 
 	local d0=$(date +%s)
 	local t0=$(ethtool_stats_get $h3 rx_octets_prio_0)
 	local u0=$(ethtool_stats_get $swp1 rx_octets_prio_0)
 
 	local -a uc_rate_2
-	uc_rate_2=($(measure_uc_rate "UC+MC"))
+	start_traffic $h2.111 192.0.2.129 192.0.2.130 $h3mac
+	uc_rate_2=($(measure_rate $swp2 $h3 rx_octets_prio_1 "UC+MC"))
 	check_err $? "Could not get high enough UC+MC ingress rate"
+	stop_traffic
 	local ucth2=${uc_rate_2[1]}
 
 	local d1=$(date +%s)
@@ -319,8 +260,7 @@ test_mc_aware()
 	local mc_ir=$(rate $u0 $u1 $interval)
 	local mc_er=$(rate $t0 $t1 $interval)
 
-	# Suppress noise from killing mausezahn.
-	{ kill %% && wait; } 2>/dev/null
+	stop_traffic
 
 	log_test "UC performace under MC overload"
 
@@ -344,8 +284,7 @@ test_uc_aware()
 {
 	RET=0
 
-	$MZ $h2.111 -p 8000 -A 192.0.2.129 -B 192.0.2.130 -c 0 \
-		-a own -b $h3mac -t udp -q &
+	start_traffic $h2.111 192.0.2.129 192.0.2.130 $h3mac
 
 	local d0=$(date +%s)
 	local t0=$(ethtool_stats_get $h3 rx_octets_prio_1)
@@ -375,8 +314,7 @@ test_uc_aware()
 	((attempts == passes))
 	check_err $?
 
-	# Suppress noise from killing mausezahn.
-	{ kill %% && wait; } 2>/dev/null
+	stop_traffic
 
 	log_test "MC performace under UC overload"
 	echo "    ingress UC throughput $(humanize ${uc_ir})"

tools/testing/selftests/drivers/net/mlxsw/spectrum-2/tc_flower.sh

@@ -12,6 +12,7 @@ ALL_TESTS="single_mask_test identical_filters_test two_masks_test \
 	delta_two_masks_one_key_test delta_simple_rehash_test \
 	bloom_simple_test bloom_complex_test bloom_delta_test"
 NUM_NETIFS=2
+source $lib_dir/lib.sh
 source $lib_dir/tc_common.sh
 source $lib_dir/devlink_lib.sh
 

tools/testing/selftests/drivers/net/mlxsw/spectrum/devlink_resources.sh

 #!/bin/bash
 # SPDX-License-Identifier: GPL-2.0
 
+lib_dir=$(dirname $0)/../../../../net/forwarding
+
 NUM_NETIFS=1
+source $lib_dir/lib.sh
 source devlink_lib_spectrum.sh
 
 setup_prepare()

tools/testing/selftests/drivers/net/mlxsw/spectrum/resource_scale.sh

 #!/bin/bash
 # SPDX-License-Identifier: GPL-2.0
 
+lib_dir=$(dirname $0)/../../../../net/forwarding
+
 NUM_NETIFS=6
-source ../../../../net/forwarding/tc_common.sh
+source $lib_dir/lib.sh
+source $lib_dir/tc_common.sh
 source devlink_lib_spectrum.sh
 
 current_test=""

tools/testing/selftests/net/forwarding/devlink_lib.sh

 #!/bin/bash
 # SPDX-License-Identifier: GPL-2.0
 
-##############################################################################
-# Source library
-
-relative_path="${BASH_SOURCE%/*}"
-if [[ "$relative_path" == "${BASH_SOURCE}" ]]; then
-	relative_path="."
-fi
-
-source "$relative_path/lib.sh"
-
 ##############################################################################
 # Defines
 
-DEVLINK_DEV=$(devlink port show | grep "${NETIFS[p1]}" | \
-	      grep -v "${NETIFS[p1]}[0-9]" | cut -d" " -f1 | \
-	      rev | cut -d"/" -f2- | rev)
+DEVLINK_DEV=$(devlink port show "${NETIFS[p1]}" -j \
+		     | jq -r '.port | keys[]' | cut -d/ -f-2)
 if [ -z "$DEVLINK_DEV" ]; then
 	echo "SKIP: ${NETIFS[p1]} has no devlink device registered for it"
 	exit 1
@@ -106,3 +95,98 @@ devlink_reload()
 			grep -c "size_new")
 	check_err $still_pending "Failed reload - There are still unset sizes"
 }
+
+declare -A DEVLINK_ORIG
+
+devlink_port_pool_threshold()
+{
+	local port=$1; shift
+	local pool=$1; shift
+
+	devlink sb port pool show $port pool $pool -j \
+		| jq '.port_pool."'"$port"'"[].threshold'
+}
+
+devlink_port_pool_th_set()
+{
+	local port=$1; shift
+	local pool=$1; shift
+	local th=$1; shift
+	local key="port_pool($port,$pool).threshold"
+
+	DEVLINK_ORIG[$key]=$(devlink_port_pool_threshold $port $pool)
+	devlink sb port pool set $port pool $pool th $th
+}
+
+devlink_port_pool_th_restore()
+{
+	local port=$1; shift
+	local pool=$1; shift
+	local key="port_pool($port,$pool).threshold"
+
+	devlink sb port pool set $port pool $pool th ${DEVLINK_ORIG[$key]}
+}
+
+devlink_pool_size_thtype()
+{
+	local pool=$1; shift
+
+	devlink sb pool show "$DEVLINK_DEV" pool $pool -j \
+	    | jq -r '.pool[][] | (.size, .thtype)'
+}
+
+devlink_pool_size_thtype_set()
+{
+	local pool=$1; shift
+	local thtype=$1; shift
+	local size=$1; shift
+	local key="pool($pool).size_thtype"
+
+	DEVLINK_ORIG[$key]=$(devlink_pool_size_thtype $pool)
+	devlink sb pool set "$DEVLINK_DEV" pool $pool size $size thtype $thtype
+}
+
+devlink_pool_size_thtype_restore()
+{
+	local pool=$1; shift
+	local key="pool($pool).size_thtype"
+	local -a orig=(${DEVLINK_ORIG[$key]})
+
+	devlink sb pool set "$DEVLINK_DEV" pool $pool \
+		size ${orig[0]} thtype ${orig[1]}
+}
+
+devlink_tc_bind_pool_th()
+{
+	local port=$1; shift
+	local tc=$1; shift
+	local dir=$1; shift
+
+	devlink sb tc bind show $port tc $tc type $dir -j \
+	    | jq -r '.tc_bind[][] | (.pool, .threshold)'
+}
+
+devlink_tc_bind_pool_th_set()
+{
+	local port=$1; shift
+	local tc=$1; shift
+	local dir=$1; shift
+	local pool=$1; shift
+	local th=$1; shift
+	local key="tc_bind($port,$dir,$tc).pool_th"
+
+	DEVLINK_ORIG[$key]=$(devlink_tc_bind_pool_th $port $tc $dir)
+	devlink sb tc bind set $port tc $tc type $dir pool $pool th $th
+}
+
+devlink_tc_bind_pool_th_restore()
+{
+	local port=$1; shift
+	local tc=$1; shift
+	local dir=$1; shift
+	local key="tc_bind($port,$dir,$tc).pool_th"
+	local -a orig=(${DEVLINK_ORIG[$key]})
+
+	devlink sb tc bind set $port tc $tc type $dir \
+		pool ${orig[0]} th ${orig[1]}
+}

tools/testing/selftests/net/forwarding/router_multicast.sh

@@ -28,7 +28,7 @@
 # +------------------+       +------------------+
 #
 
-ALL_TESTS="mcast_v4 mcast_v6"
+ALL_TESTS="mcast_v4 mcast_v6 rpf_v4 rpf_v6"
 NUM_NETIFS=6
 source lib.sh
 source tc_common.sh
@@ -46,10 +46,14 @@ h1_create()
 
 	ip route add 2001:db8:2::/64 vrf v$h1 nexthop via 2001:db8:1::1
 	ip route add 2001:db8:3::/64 vrf v$h1 nexthop via 2001:db8:1::1
+
+	tc qdisc add dev $h1 ingress
 }
 
 h1_destroy()
 {
+	tc qdisc del dev $h1 ingress
+
 	ip route del 2001:db8:3::/64 vrf v$h1
 	ip route del 2001:db8:2::/64 vrf v$h1
 
@@ -124,10 +128,14 @@ router_create()
 	ip address add 2001:db8:1::1/64 dev $rp1
 	ip address add 2001:db8:2::1/64 dev $rp2
 	ip address add 2001:db8:3::1/64 dev $rp3
+
+	tc qdisc add dev $rp3 ingress
 }
 
 router_destroy()
 {
+	tc qdisc del dev $rp3 ingress
+
 	ip address del 2001:db8:3::1/64 dev $rp3
 	ip address del 2001:db8:2::1/64 dev $rp2
 	ip address del 2001:db8:1::1/64 dev $rp1
@@ -301,6 +309,103 @@ mcast_v6()
 	log_test "mcast IPv6"
 }
 
+rpf_v4()
+{
+	# Add a multicast route from first router port to the other two. Send
+	# matching packets and test that both hosts receive them. Then, send
+	# the same packets via the third router port and test that they do not
+	# reach any host due to RPF check. A filter with 'skip_hw' is added to
+	# test that devices capable of multicast routing offload trap those
+	# packets. The filter is essentialy a NOP in other scenarios.
+
+	RET=0
+
+	tc filter add dev $h1 ingress protocol ip pref 1 handle 1 flower \
+		dst_ip 225.1.2.3 ip_proto udp dst_port 12345 action drop
+	tc filter add dev $h2 ingress protocol ip pref 1 handle 1 flower \
+		dst_ip 225.1.2.3 ip_proto udp dst_port 12345 action drop
+	tc filter add dev $h3 ingress protocol ip pref 1 handle 1 flower \
+		dst_ip 225.1.2.3 ip_proto udp dst_port 12345 action drop
+	tc filter add dev $rp3 ingress protocol ip pref 1 handle 1 flower \
+		skip_hw dst_ip 225.1.2.3 ip_proto udp dst_port 12345 action pass
+
+	create_mcast_sg $rp1 198.51.100.2 225.1.2.3 $rp2 $rp3
+
+	$MZ $h1 -c 5 -p 128 -t udp "ttl=10,sp=54321,dp=12345" \
+		-a 00:11:22:33:44:55 -b 01:00:5e:01:02:03 \
+		-A 198.51.100.2 -B 225.1.2.3 -q
+
+	tc_check_packets "dev $h2 ingress" 1 5
+	check_err $? "Multicast not received on first host"
+	tc_check_packets "dev $h3 ingress" 1 5
+	check_err $? "Multicast not received on second host"
+
+	$MZ $h3 -c 5 -p 128 -t udp "ttl=10,sp=54321,dp=12345" \
+		-a 00:11:22:33:44:55 -b 01:00:5e:01:02:03 \
+		-A 198.51.100.2 -B 225.1.2.3 -q
+
+	tc_check_packets "dev $h1 ingress" 1 0
+	check_err $? "Multicast received on first host when should not"
+	tc_check_packets "dev $h2 ingress" 1 5
+	check_err $? "Multicast received on second host when should not"
+	tc_check_packets "dev $rp3 ingress" 1 5
+	check_err $? "Packets not trapped due to RPF check"
+
+	delete_mcast_sg $rp1 198.51.100.2 225.1.2.3 $rp2 $rp3
+
+	tc filter del dev $rp3 ingress protocol ip pref 1 handle 1 flower
+	tc filter del dev $h3 ingress protocol ip pref 1 handle 1 flower
+	tc filter del dev $h2 ingress protocol ip pref 1 handle 1 flower
+	tc filter del dev $h1 ingress protocol ip pref 1 handle 1 flower
+
+	log_test "RPF IPv4"
+}
+
+rpf_v6()
+{
+	RET=0
+
+	tc filter add dev $h1 ingress protocol ipv6 pref 1 handle 1 flower \
+		dst_ip ff0e::3 ip_proto udp dst_port 12345 action drop
+	tc filter add dev $h2 ingress protocol ipv6 pref 1 handle 1 flower \
+		dst_ip ff0e::3 ip_proto udp dst_port 12345 action drop
+	tc filter add dev $h3 ingress protocol ipv6 pref 1 handle 1 flower \
+		dst_ip ff0e::3 ip_proto udp dst_port 12345 action drop
+	tc filter add dev $rp3 ingress protocol ipv6 pref 1 handle 1 flower \
+		skip_hw dst_ip ff0e::3 ip_proto udp dst_port 12345 action pass
+
+	create_mcast_sg $rp1 2001:db8:1::2 ff0e::3 $rp2 $rp3
+
+	$MZ $h1 -6 -c 5 -p 128 -t udp "ttl=10,sp=54321,dp=12345" \
+		-a 00:11:22:33:44:55 -b 33:33:00:00:00:03 \
+		-A 2001:db8:1::2 -B ff0e::3 -q
+
+	tc_check_packets "dev $h2 ingress" 1 5
+	check_err $? "Multicast not received on first host"
+	tc_check_packets "dev $h3 ingress" 1 5
+	check_err $? "Multicast not received on second host"
+
+	$MZ $h3 -6 -c 5 -p 128 -t udp "ttl=10,sp=54321,dp=12345" \
+		-a 00:11:22:33:44:55 -b 33:33:00:00:00:03 \
+		-A 2001:db8:1::2 -B ff0e::3 -q
+
+	tc_check_packets "dev $h1 ingress" 1 0
+	check_err $? "Multicast received on first host when should not"
+	tc_check_packets "dev $h2 ingress" 1 5
+	check_err $? "Multicast received on second host when should not"
+	tc_check_packets "dev $rp3 ingress" 1 5
+	check_err $? "Packets not trapped due to RPF check"
+
+	delete_mcast_sg $rp1 2001:db8:1::2 ff0e::3 $rp2 $rp3
+
+	tc filter del dev $rp3 ingress protocol ipv6 pref 1 handle 1 flower
+	tc filter del dev $h3 ingress protocol ipv6 pref 1 handle 1 flower
+	tc filter del dev $h2 ingress protocol ipv6 pref 1 handle 1 flower
+	tc filter del dev $h1 ingress protocol ipv6 pref 1 handle 1 flower
+
+	log_test "RPF IPv6"
+}
+
 trap cleanup EXIT
 
 setup_prepare

tools/testing/selftests/net/forwarding/tc_flower.sh

@@ -2,7 +2,7 @@
 # SPDX-License-Identifier: GPL-2.0
 
 ALL_TESTS="match_dst_mac_test match_src_mac_test match_dst_ip_test \
-	match_src_ip_test match_ip_flags_test"
+	match_src_ip_test match_ip_flags_test match_pcp_test match_vlan_test"
 NUM_NETIFS=2
 source tc_common.sh
 source lib.sh
@@ -219,6 +219,63 @@ match_ip_flags_test()
 	log_test "ip_flags match ($tcflags)"
 }
 
+match_pcp_test()
+{
+	RET=0
+
+	vlan_create $h2 85 v$h2 192.0.2.11/24
+
+	tc filter add dev $h2 ingress protocol 802.1q pref 1 handle 101 \
+		flower vlan_prio 6 $tcflags dst_mac $h2mac action drop
+	tc filter add dev $h2 ingress protocol 802.1q pref 2 handle 102 \
+		flower vlan_prio 7 $tcflags dst_mac $h2mac action drop
+
+	$MZ $h1 -c 1 -p 64 -a $h1mac -b $h2mac -B 192.0.2.11 -Q 7:85 -t ip -q
+	$MZ $h1 -c 1 -p 64 -a $h1mac -b $h2mac -B 192.0.2.11 -Q 0:85 -t ip -q
+
+	tc_check_packets "dev $h2 ingress" 101 0
+	check_err $? "Matched on specified PCP when should not"
+
+	tc_check_packets "dev $h2 ingress" 102 1
+	check_err $? "Did not match on specified PCP"
+
+	tc filter del dev $h2 ingress protocol 802.1q pref 2 handle 102 flower
+	tc filter del dev $h2 ingress protocol 802.1q pref 1 handle 101 flower
+
+	vlan_destroy $h2 85
+
+	log_test "PCP match ($tcflags)"
+}
+
+match_vlan_test()
+{
+	RET=0
+
+	vlan_create $h2 85 v$h2 192.0.2.11/24
+	vlan_create $h2 75 v$h2 192.0.2.10/24
+
+	tc filter add dev $h2 ingress protocol 802.1q pref 1 handle 101 \
+		flower vlan_id 75 $tcflags action drop
+	tc filter add dev $h2 ingress protocol 802.1q pref 2 handle 102 \
+		flower vlan_id 85 $tcflags action drop
+
+	$MZ $h1 -c 1 -p 64 -a $h1mac -b $h2mac -B 192.0.2.11 -Q 0:85 -t ip -q
+
+	tc_check_packets "dev $h2 ingress" 101 0
+	check_err $? "Matched on specified VLAN when should not"
+
+	tc_check_packets "dev $h2 ingress" 102 1
+	check_err $? "Did not match on specified VLAN"
+
+	tc filter del dev $h2 ingress protocol 802.1q pref 2 handle 102 flower
+	tc filter del dev $h2 ingress protocol 802.1q pref 1 handle 101 flower
+
+	vlan_destroy $h2 75
+	vlan_destroy $h2 85
+
+	log_test "VLAN match ($tcflags)"
+}
+
 setup_prepare()
 {
 	h1=${NETIFS[p1]}

tools/testing/selftests/net/forwarding/tc_vlan_modify.sh

+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+ALL_TESTS="
+	vlan_modify_ingress
+	vlan_modify_egress
+"
+
+NUM_NETIFS=4
+CHECK_TC="yes"
+source lib.sh
+
+h1_create()
+{
+	simple_if_init $h1 192.0.2.1/28 2001:db8:1::1/64
+	vlan_create $h1 85 v$h1 192.0.2.17/28 2001:db8:2::1/64
+}
+
+h1_destroy()
+{
+	vlan_destroy $h1 85
+	simple_if_fini $h1 192.0.2.1/28 2001:db8:1::1/64
+}
+
+h2_create()
+{
+	simple_if_init $h2 192.0.2.2/28 2001:db8:1::2/64
+	vlan_create $h2 65 v$h2 192.0.2.18/28 2001:db8:2::2/64
+}
+
+h2_destroy()
+{
+	vlan_destroy $h2 65
+	simple_if_fini $h2 192.0.2.2/28 2001:db8:1::2/64
+}
+
+switch_create()
+{
+	ip link add dev br0 type bridge vlan_filtering 1 mcast_snooping 0
+
+	ip link set dev $swp1 master br0
+	ip link set dev $swp2 master br0
+
+	ip link set dev br0 up
+	ip link set dev $swp1 up
+	ip link set dev $swp2 up
+
+	bridge vlan add dev $swp1 vid 85
+	bridge vlan add dev $swp2 vid 65
+
+	bridge vlan add dev $swp2 vid 85
+	bridge vlan add dev $swp1 vid 65
+
+	tc qdisc add dev $swp1 clsact
+	tc qdisc add dev $swp2 clsact
+}
+
+switch_destroy()
+{
+	tc qdisc del dev $swp2 clsact
+	tc qdisc del dev $swp1 clsact
+
+	bridge vlan del vid 65 dev $swp1
+	bridge vlan del vid 85 dev $swp2
+
+	bridge vlan del vid 65 dev $swp2
+	bridge vlan del vid 85 dev $swp1
+
+	ip link set dev $swp2 down
+	ip link set dev $swp1 down
+
+	ip link del dev br0
+}
+
+setup_prepare()
+{
+	h1=${NETIFS[p1]}
+	swp1=${NETIFS[p2]}
+
+	swp2=${NETIFS[p3]}
+	h2=${NETIFS[p4]}
+
+	vrf_prepare
+
+	h1_create
+	h2_create
+
+	switch_create
+}
+
+cleanup()
+{
+	pre_cleanup
+
+	switch_destroy
+
+	h2_destroy
+	h1_destroy
+
+	vrf_cleanup
+}
+
+vlan_modify_ingress()
+{
+	RET=0
+
+	ping_do $h1.85 192.0.2.18
+	check_fail $? "ping between two different vlans passed when should not"
+
+	ping6_do $h1.85 2001:db8:2::2
+	check_fail $? "ping6 between two different vlans passed when should not"
+
+	tc filter add dev $swp1 ingress protocol all pref 1 handle 1 \
+		flower action vlan modify id 65
+	tc filter add dev $swp2 ingress protocol all pref 1 handle 1 \
+		flower action vlan modify id 85
+
+	ping_do $h1.85 192.0.2.18
+	check_err $? "ping between two different vlans failed when should not"
+
+	ping6_do $h1.85 2001:db8:2::2
+	check_err $? "ping6 between two different vlans failed when should not"
+
+	log_test "VLAN modify at ingress"
+
+	tc filter del dev $swp2 ingress protocol all pref 1 handle 1 flower
+	tc filter del dev $swp1 ingress protocol all pref 1 handle 1 flower
+}
+
+vlan_modify_egress()
+{
+	RET=0
+
+	ping_do $h1.85 192.0.2.18
+	check_fail $? "ping between two different vlans passed when should not"
+
+	ping6_do $h1.85 2001:db8:2::2
+	check_fail $? "ping6 between two different vlans passed when should not"
+
+	tc filter add dev $swp1 egress protocol all pref 1 handle 1 \
+		flower action vlan modify id 85
+	tc filter add dev $swp2 egress protocol all pref 1 handle 1 \
+		flower action vlan modify id 65
+
+	ping_do $h1.85 192.0.2.18
+	check_err $? "ping between two different vlans failed when should not"
+
+	ping6_do $h1.85 2001:db8:2::2
+	check_err $? "ping6 between two different vlans failed when should not"
+
+	log_test "VLAN modify at egress"
+
+	tc filter del dev $swp2 egress protocol all pref 1 handle 1 flower
+	tc filter del dev $swp1 egress protocol all pref 1 handle 1 flower
+}
+
+trap cleanup EXIT
+
+setup_prepare
+setup_wait
+
+tests_run
+
+exit $EXIT_STATUS