Merge branch 'selftests-updates-for-mlxsw-driver-test'

Jiri Pirko says:

====================
selftests: updates for mlxsw driver test

This patchset contains tweaks to the existing tests and is also adding
couple of new ones, namely tests for shared buffer and red offload.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

17 files changed, 1653 insertions, 186 deletions

diff --git a/tools/testing/selftests/drivers/net/mlxsw/blackhole_routes.sh b/tools/testing/selftests/drivers/net/mlxsw/blackhole_routes.sh
index 5ba5bef44d5b..bdffe698e1d1 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/blackhole_routes.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/blackhole_routes.sh
@@ -45,6 +45,7 @@ ALL_TESTS="
 	blackhole_ipv6
 "
 NUM_NETIFS=4
+: ${TIMEOUT:=20000} # ms
 source $lib_dir/tc_common.sh
 source $lib_dir/lib.sh
 
@@ -123,7 +124,7 @@ blackhole_ipv4()
 		skip_hw dst_ip 198.51.100.1 src_ip 192.0.2.1 ip_proto icmp \
 		action pass
 
-	ip -4 route show 198.51.100.0/30 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload ip -4 route show 198.51.100.0/30
 	check_err $? "route not marked as offloaded when should"
 
 	ping_do $h1 198.51.100.1
@@ -147,7 +148,7 @@ blackhole_ipv6()
 		skip_hw dst_ip 2001:db8:2::1 src_ip 2001:db8:1::1 \
 		ip_proto icmpv6 action pass
 
-	ip -6 route show 2001:db8:2::/120 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload ip -6 route show 2001:db8:2::/120
 	check_err $? "route not marked as offloaded when should"
 
 	ping6_do $h1 2001:db8:2::1
diff --git a/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_drops.sh b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_drops.sh
index 053e5c7b303d..616f47d86a61 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_drops.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_drops.sh
@@ -641,13 +641,9 @@ erif_disabled_test()
 	mz_pid=$!
 
 	sleep 5
-	# In order to see this trap we need a route that points to disabled RIF.
-	# When ipv6 address is flushed, there is a delay and the routes are
-	# deleted before the RIF and we cannot get state that we have route
-	# to disabled RIF.
-	# Delete IPv6 address first and then check this trap with flushing IPv4.
-	ip -6 add flush dev br0
-	ip -4 add flush dev br0
+	# Unlinking the port from the bridge will disable the RIF associated
+	# with br0 as it is no longer an upper of any mlxsw port.
+	ip link set dev $rp1 nomaster
 
 	t1_packets=$(devlink_trap_rx_packets_get $trap_name)
 	t1_bytes=$(devlink_trap_rx_bytes_get $trap_name)
@@ -659,7 +655,6 @@ erif_disabled_test()
 	log_test "Egress RIF disabled"
 
 	kill $mz_pid && wait $mz_pid &> /dev/null
-	ip link set dev $rp1 nomaster
 	__addr_add_del $rp1 add 192.0.2.2/24 2001:db8:1::2/64
 	ip link del dev br0 type bridge
 	devlink_trap_action_set $trap_name "drop"
diff --git a/tools/testing/selftests/drivers/net/mlxsw/mlxsw_lib.sh b/tools/testing/selftests/drivers/net/mlxsw/mlxsw_lib.sh
new file mode 100644
index 000000000000..cbe50f260a40
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/mlxsw/mlxsw_lib.sh
@@ -0,0 +1,13 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+##############################################################################
+# Defines
+
+if [[ ! -v MLXSW_CHIP ]]; then
+	MLXSW_CHIP=$(devlink -j dev info $DEVLINK_DEV | jq -r '.[][]["driver"]')
+	if [ -z "$MLXSW_CHIP" ]; then
+		echo "SKIP: Device $DEVLINK_DEV doesn't support devlink info command"
+		exit 1
+	fi
+fi
diff --git a/tools/testing/selftests/drivers/net/mlxsw/router_scale.sh b/tools/testing/selftests/drivers/net/mlxsw/router_scale.sh
index d231649b4f01..e93878d42596 100644
--- a/tools/testing/selftests/drivers/net/mlxsw/router_scale.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/router_scale.sh
@@ -2,16 +2,15 @@
 # SPDX-License-Identifier: GPL-2.0
 
 ROUTER_NUM_NETIFS=4
+: ${TIMEOUT:=20000} # ms
 
 router_h1_create()
 {
 	simple_if_init $h1 192.0.1.1/24
-	ip route add 193.0.0.0/8 via 192.0.1.2 dev $h1
 }
 
 router_h1_destroy()
 {
-	ip route del 193.0.0.0/8 via 192.0.1.2 dev $h1
 	simple_if_fini $h1 192.0.1.1/24
 }
 
@@ -64,13 +63,15 @@ router_setup_prepare()
 	router_create
 }
 
-router_offload_validate()
+wait_for_routes()
 {
-	local route_count=$1
-	local offloaded_count
+	local t0=$1; shift
+	local route_count=$1; shift
 
-	offloaded_count=$(ip route | grep -o 'offload' | wc -l)
-	[[ $offloaded_count -ge $route_count ]]
+	local t1=$(ip route | grep -o 'offload' | wc -l)
+	local delta=$((t1 - t0))
+	echo $delta
+	[[ $delta -ge $route_count ]]
 }
 
 router_routes_create()
@@ -90,8 +91,8 @@ router_routes_create()
 					break 3
 				fi
 
-				echo route add 193.${i}.${j}.${k}/32 via \
-				       192.0.2.1 dev $rp2  >> $ROUTE_FILE
+				echo route add 193.${i}.${j}.${k}/32 dev $rp2 \
+					>> $ROUTE_FILE
 				((count++))
 			done
 		done
@@ -111,45 +112,19 @@ router_test()
 {
 	local route_count=$1
 	local should_fail=$2
-	local count=0
+	local delta
 
 	RET=0
 
+	local t0=$(ip route | grep -o 'offload' | wc -l)
 	router_routes_create $route_count
+	delta=$(busywait "$TIMEOUT" wait_for_routes $t0 $route_count)
 
-	router_offload_validate $route_count
-	check_err_fail $should_fail $? "Offload of $route_count routes"
+	check_err_fail $should_fail $? "Offload routes: Expected $route_count, got $delta."
 	if [[ $RET -ne 0 ]] || [[ $should_fail -eq 1 ]]; then
 		return
 	fi
 
-	tc filter add dev $h2 ingress protocol ip pref 1 flower \
-		skip_sw dst_ip 193.0.0.0/8 action drop
-
-	for i in {0..255}
-	do
-		for j in {0..255}
-		do
-			for k in {0..255}
-			do
-				if [[ $count -eq $route_count ]]; then
-					break 3
-				fi
-
-				$MZ $h1 -c 1 -p 64 -a $h1mac -b $rp1mac \
-					-A 192.0.1.1 -B 193.${i}.${j}.${k} \
-					-t ip -q
-				((count++))
-			done
-		done
-	done
-
-	tc_check_packets "dev $h2 ingress" 1 $route_count
-	check_err $? "Offload mismatch"
-
-	tc filter del dev $h2 ingress protocol ip pref 1 flower \
-		skip_sw dst_ip 193.0.0.0/8 action drop
-
 	router_routes_destroy
 }
 
diff --git a/tools/testing/selftests/drivers/net/mlxsw/rtnetlink.sh b/tools/testing/selftests/drivers/net/mlxsw/rtnetlink.sh
index 5c39e5f6a480..f4031002d5e9 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/rtnetlink.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/rtnetlink.sh
@@ -32,6 +32,7 @@ ALL_TESTS="
 	devlink_reload_test
 "
 NUM_NETIFS=2
+: ${TIMEOUT:=20000} # ms
 source $lib_dir/lib.sh
 source $lib_dir/devlink_lib.sh
 
@@ -360,20 +361,24 @@ vlan_rif_refcount_test()
 	ip link add link br0 name br0.10 up type vlan id 10
 	ip -6 address add 2001:db8:1::1/64 dev br0.10
 
-	ip -6 route get fibmatch 2001:db8:1::2 dev br0.10 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		ip -6 route get fibmatch 2001:db8:1::2 dev br0.10
 	check_err $? "vlan rif was not created before adding port to vlan"
 
 	bridge vlan add vid 10 dev $swp1
-	ip -6 route get fibmatch 2001:db8:1::2 dev br0.10 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		ip -6 route get fibmatch 2001:db8:1::2 dev br0.10
 	check_err $? "vlan rif was destroyed after adding port to vlan"
 
 	bridge vlan del vid 10 dev $swp1
-	ip -6 route get fibmatch 2001:db8:1::2 dev br0.10 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		ip -6 route get fibmatch 2001:db8:1::2 dev br0.10
 	check_err $? "vlan rif was destroyed after removing port from vlan"
 
 	ip link set dev $swp1 nomaster
-	ip -6 route get fibmatch 2001:db8:1::2 dev br0.10 | grep -q offload
-	check_fail $? "vlan rif was not destroyed after unlinking port from bridge"
+	busywait "$TIMEOUT" not wait_for_offload \
+		ip -6 route get fibmatch 2001:db8:1::2 dev br0.10
+	check_err $? "vlan rif was not destroyed after unlinking port from bridge"
 
 	log_test "vlan rif refcount"
 
@@ -401,22 +406,28 @@ subport_rif_refcount_test()
 	ip -6 address add 2001:db8:1::1/64 dev bond1
 	ip -6 address add 2001:db8:2::1/64 dev bond1.10
 
-	ip -6 route get fibmatch 2001:db8:1::2 dev bond1 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		ip -6 route get fibmatch 2001:db8:1::2 dev bond1
 	check_err $? "subport rif was not created on lag device"
-	ip -6 route get fibmatch 2001:db8:2::2 dev bond1.10 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		ip -6 route get fibmatch 2001:db8:2::2 dev bond1.10
 	check_err $? "subport rif was not created on vlan device"
 
 	ip link set dev $swp1 nomaster
-	ip -6 route get fibmatch 2001:db8:1::2 dev bond1 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		ip -6 route get fibmatch 2001:db8:1::2 dev bond1
 	check_err $? "subport rif of lag device was destroyed when should not"
-	ip -6 route get fibmatch 2001:db8:2::2 dev bond1.10 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		ip -6 route get fibmatch 2001:db8:2::2 dev bond1.10
 	check_err $? "subport rif of vlan device was destroyed when should not"
 
 	ip link set dev $swp2 nomaster
-	ip -6 route get fibmatch 2001:db8:1::2 dev bond1 | grep -q offload
-	check_fail $? "subport rif of lag device was not destroyed when should"
-	ip -6 route get fibmatch 2001:db8:2::2 dev bond1.10 | grep -q offload
-	check_fail $? "subport rif of vlan device was not destroyed when should"
+	busywait "$TIMEOUT" not wait_for_offload \
+		ip -6 route get fibmatch 2001:db8:1::2 dev bond1
+	check_err $? "subport rif of lag device was not destroyed when should"
+	busywait "$TIMEOUT" not wait_for_offload \
+		ip -6 route get fibmatch 2001:db8:2::2 dev bond1.10
+	check_err $? "subport rif of vlan device was not destroyed when should"
 
 	log_test "subport rif refcount"
 
@@ -575,7 +586,8 @@ bridge_extern_learn_test()
 
 	bridge fdb add de:ad:be:ef:13:37 dev $swp1 master extern_learn
 
-	bridge fdb show brport $swp1 | grep de:ad:be:ef:13:37 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		bridge fdb show brport $swp1 de:ad:be:ef:13:37
 	check_err $? "fdb entry not marked as offloaded when should"
 
 	log_test "externally learned fdb entry"
@@ -595,9 +607,11 @@ neigh_offload_test()
 	ip -6 neigh add 2001:db8:1::2 lladdr de:ad:be:ef:13:37 nud perm \
 		dev $swp1
 
-	ip -4 neigh show dev $swp1 | grep 192.0.2.2 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		ip -4 neigh show dev $swp1 192.0.2.2
 	check_err $? "ipv4 neigh entry not marked as offloaded when should"
-	ip -6 neigh show dev $swp1 | grep 2001:db8:1::2 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		ip -6 neigh show dev $swp1 2001:db8:1::2
 	check_err $? "ipv6 neigh entry not marked as offloaded when should"
 
 	log_test "neighbour offload indication"
@@ -623,25 +637,31 @@ nexthop_offload_test()
 	ip -6 route add 2001:db8:2::/64 vrf v$swp1 \
 		nexthop via 2001:db8:1::2 dev $swp1
 
-	ip -4 route show 198.51.100.0/24 vrf v$swp1 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		ip -4 route show 198.51.100.0/24 vrf v$swp1
 	check_err $? "ipv4 nexthop not marked as offloaded when should"
-	ip -6 route show 2001:db8:2::/64 vrf v$swp1 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		ip -6 route show 2001:db8:2::/64 vrf v$swp1
 	check_err $? "ipv6 nexthop not marked as offloaded when should"
 
 	ip link set dev $swp2 down
 	sleep 1
 
-	ip -4 route show 198.51.100.0/24 vrf v$swp1 | grep -q offload
-	check_fail $? "ipv4 nexthop marked as offloaded when should not"
-	ip -6 route show 2001:db8:2::/64 vrf v$swp1 | grep -q offload
-	check_fail $? "ipv6 nexthop marked as offloaded when should not"
+	busywait "$TIMEOUT" not wait_for_offload \
+		ip -4 route show 198.51.100.0/24 vrf v$swp1
+	check_err $? "ipv4 nexthop marked as offloaded when should not"
+	busywait "$TIMEOUT" not wait_for_offload \
+		ip -6 route show 2001:db8:2::/64 vrf v$swp1
+	check_err $? "ipv6 nexthop marked as offloaded when should not"
 
 	ip link set dev $swp2 up
 	setup_wait
 
-	ip -4 route show 198.51.100.0/24 vrf v$swp1 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		ip -4 route show 198.51.100.0/24 vrf v$swp1
 	check_err $? "ipv4 nexthop not marked as offloaded after neigh add"
-	ip -6 route show 2001:db8:2::/64 vrf v$swp1 | grep -q offload
+	busywait "$TIMEOUT" wait_for_offload \
+		ip -6 route show 2001:db8:2::/64 vrf v$swp1
 	check_err $? "ipv6 nexthop not marked as offloaded after neigh add"
 
 	log_test "nexthop offload indication"
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_red_core.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_red_core.sh
new file mode 100644
index 000000000000..ebf7752f6d93
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_red_core.sh
@@ -0,0 +1,499 @@
+# SPDX-License-Identifier: GPL-2.0
+
+# This test sends a >1Gbps stream of traffic from H1, to the switch, which
+# forwards it to a 1Gbps port. This 1Gbps stream is then looped back to the
+# switch and forwarded to the port under test $swp3, which is also 1Gbps.
+#
+# This way, $swp3 should be 100% filled with traffic without any of it spilling
+# to the backlog. Any extra packets sent should almost 1:1 go to backlog. That
+# is what H2 is used for--it sends the extra traffic to create backlog.
+#
+# A RED Qdisc is installed on $swp3. The configuration is such that the minimum
+# and maximum size are 1 byte apart, so there is a very clear border under which
+# no marking or dropping takes place, and above which everything is marked or
+# dropped.
+#
+# The test uses the buffer build-up behavior to test the installed RED.
+#
+# In order to test WRED, $swp3 actually contains RED under PRIO, with two
+# different configurations. Traffic is prioritized using 802.1p and relies on
+# the implicit mlxsw configuration, where packet priority is taken 1:1 from the
+# 802.1p marking.
+#
+# +--------------------------+                     +--------------------------+
+# | H1                       |                     | H2                       |
+# |     + $h1.10             |                     |     + $h2.10             |
+# |     | 192.0.2.1/28       |                     |     | 192.0.2.2/28       |
+# |     |                    |                     |     |                    |
+# |     |         $h1.11 +   |                     |     |         $h2.11 +   |
+# |     |  192.0.2.17/28 |   |                     |     |  192.0.2.18/28 |   |
+# |     |                |   |                     |     |                |   |
+# |     \______    ______/   |                     |     \______    ______/   |
+# |            \ /           |                     |            \ /           |
+# |             + $h1        |                     |             + $h2        |
+# +-------------|------------+                     +-------------|------------+
+#               | >1Gbps                                         |
+# +-------------|------------------------------------------------|------------+
+# | SW          + $swp1                                          + $swp2      |
+# |     _______/ \___________                        ___________/ \_______    |
+# |    /                     \                      /                     \   |
+# |  +-|-----------------+   |                    +-|-----------------+   |   |
+# |  | + $swp1.10        |   |                    | + $swp2.10        |   |   |
+# |  |                   |   |        .-------------+ $swp5.10        |   |   |
+# |  |     BR1_10        |   |        |           |                   |   |   |
+# |  |                   |   |        |           |     BR2_10        |   |   |
+# |  | + $swp2.10        |   |        |           |                   |   |   |
+# |  +-|-----------------+   |        |           | + $swp3.10        |   |   |
+# |    |                     |        |           +-|-----------------+   |   |
+# |    |   +-----------------|-+      |             |   +-----------------|-+ |
+# |    |   |        $swp1.11 + |      |             |   |        $swp2.11 + | |
+# |    |   |                   |      | .-----------------+ $swp5.11        | |
+# |    |   |      BR1_11       |      | |           |   |                   | |
+# |    |   |                   |      | |           |   |      BR2_11       | |
+# |    |   |        $swp2.11 + |      | |           |   |                   | |
+# |    |   +-----------------|-+      | |           |   |        $swp3.11 + | |
+# |    |                     |        | |           |   +-----------------|-+ |
+# |    \_______   ___________/        | |           \___________   _______/   |
+# |            \ /                    \ /                       \ /           |
+# |             + $swp4                + $swp5                   + $swp3      |
+# +-------------|----------------------|-------------------------|------------+
+#               |                      |                         | 1Gbps
+#               \________1Gbps_________/                         |
+#                                   +----------------------------|------------+
+#                                   | H3                         + $h3        |
+#                                   |      _____________________/ \_______    |
+#                                   |     /                               \   |
+#                                   |     |                               |   |
+#                                   |     + $h3.10                 $h3.11 +   |
+#                                   |       192.0.2.3/28    192.0.2.19/28     |
+#                                   +-----------------------------------------+
+
+NUM_NETIFS=8
+CHECK_TC="yes"
+lib_dir=$(dirname $0)/../../../net/forwarding
+source $lib_dir/lib.sh
+source $lib_dir/devlink_lib.sh
+source qos_lib.sh
+
+ipaddr()
+{
+	local host=$1; shift
+	local vlan=$1; shift
+
+	echo 192.0.2.$((16 * (vlan - 10) + host))
+}
+
+host_create()
+{
+	local dev=$1; shift
+	local host=$1; shift
+
+	simple_if_init $dev
+	mtu_set $dev 10000
+
+	vlan_create $dev 10 v$dev $(ipaddr $host 10)/28
+	ip link set dev $dev.10 type vlan egress 0:0
+
+	vlan_create $dev 11 v$dev $(ipaddr $host 11)/28
+	ip link set dev $dev.11 type vlan egress 0:1
+}
+
+host_destroy()
+{
+	local dev=$1; shift
+
+	vlan_destroy $dev 11
+	vlan_destroy $dev 10
+	mtu_restore $dev
+	simple_if_fini $dev
+}
+
+h1_create()
+{
+	host_create $h1 1
+}
+
+h1_destroy()
+{
+	host_destroy $h1
+}
+
+h2_create()
+{
+	host_create $h2 2
+
+	# Some of the tests in this suite use multicast traffic. As this traffic
+	# enters BR2_10 resp. BR2_11, it is flooded to all other ports. Thus
+	# e.g. traffic ingressing through $swp2 is flooded to $swp3 (the
+	# intended destination) and $swp5 (which is intended as ingress for
+	# another stream of traffic).
+	#
+	# This is generally not a problem, but if the $swp5 throughput is lower
+	# than $swp2 throughput, there will be a build-up at $swp5. That may
+	# cause packets to fail to queue up at $swp3 due to shared buffer
+	# quotas, and the test to spuriously fail.
+	#
+	# Prevent this by setting the speed of $h2 to 1Gbps.
+
+	ethtool -s $h2 speed 1000 autoneg off
+}
+
+h2_destroy()
+{
+	ethtool -s $h2 autoneg on
+	host_destroy $h2
+}
+
+h3_create()
+{
+	host_create $h3 3
+	ethtool -s $h3 speed 1000 autoneg off
+}
+
+h3_destroy()
+{
+	ethtool -s $h3 autoneg on
+	host_destroy $h3
+}
+
+switch_create()
+{
+	local intf
+	local vlan
+
+	ip link add dev br1_10 type bridge
+	ip link add dev br1_11 type bridge
+
+	ip link add dev br2_10 type bridge
+	ip link add dev br2_11 type bridge
+
+	for intf in $swp1 $swp2 $swp3 $swp4 $swp5; do
+		ip link set dev $intf up
+		mtu_set $intf 10000
+	done
+
+	for intf in $swp1 $swp4; do
+		for vlan in 10 11; do
+			vlan_create $intf $vlan
+			ip link set dev $intf.$vlan master br1_$vlan
+			ip link set dev $intf.$vlan up
+		done
+	done
+
+	for intf in $swp2 $swp3 $swp5; do
+		for vlan in 10 11; do
+			vlan_create $intf $vlan
+			ip link set dev $intf.$vlan master br2_$vlan
+			ip link set dev $intf.$vlan up
+		done
+	done
+
+	ip link set dev $swp4.10 type vlan egress 0:0
+	ip link set dev $swp4.11 type vlan egress 0:1
+	for intf in $swp1 $swp2 $swp5; do
+		for vlan in 10 11; do
+			ip link set dev $intf.$vlan type vlan ingress 0:0 1:1
+		done
+	done
+
+	for intf in $swp2 $swp3 $swp4 $swp5; do
+		ethtool -s $intf speed 1000 autoneg off
+	done
+
+	ip link set dev br1_10 up
+	ip link set dev br1_11 up
+	ip link set dev br2_10 up
+	ip link set dev br2_11 up
+
+	local size=$(devlink_pool_size_thtype 0 | cut -d' ' -f 1)
+	devlink_port_pool_th_set $swp3 8 $size
+}
+
+switch_destroy()
+{
+	local intf
+	local vlan
+
+	devlink_port_pool_th_restore $swp3 8
+
+	tc qdisc del dev $swp3 root 2>/dev/null
+
+	ip link set dev br2_11 down
+	ip link set dev br2_10 down
+	ip link set dev br1_11 down
+	ip link set dev br1_10 down
+
+	for intf in $swp5 $swp4 $swp3 $swp2; do
+		ethtool -s $intf autoneg on
+	done
+
+	for intf in $swp5 $swp3 $swp2 $swp4 $swp1; do
+		for vlan in 11 10; do
+			ip link set dev $intf.$vlan down
+			ip link set dev $intf.$vlan nomaster
+			vlan_destroy $intf $vlan
+		done
+
+		mtu_restore $intf
+		ip link set dev $intf down
+	done
+
+	ip link del dev br2_11
+	ip link del dev br2_10
+	ip link del dev br1_11
+	ip link del dev br1_10
+}
+
+setup_prepare()
+{
+	h1=${NETIFS[p1]}
+	swp1=${NETIFS[p2]}
+
+	swp2=${NETIFS[p3]}
+	h2=${NETIFS[p4]}
+
+	swp3=${NETIFS[p5]}
+	h3=${NETIFS[p6]}
+
+	swp4=${NETIFS[p7]}
+	swp5=${NETIFS[p8]}
+
+	h3_mac=$(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 $h1.10 $(ipaddr 3 10) " from host 1, vlan 10"
+	ping_test $h1.11 $(ipaddr 3 11) " from host 1, vlan 11"
+	ping_test $h2.10 $(ipaddr 3 10) " from host 2, vlan 10"
+	ping_test $h2.11 $(ipaddr 3 11) " from host 2, vlan 11"
+}
+
+get_tc()
+{
+	local vlan=$1; shift
+
+	echo $((vlan - 10))
+}
+
+get_qdisc_handle()
+{
+	local vlan=$1; shift
+
+	local tc=$(get_tc $vlan)
+	local band=$((8 - tc))
+
+	# Handle is 107: for TC1, 108: for TC0.
+	echo "10$band:"
+}
+
+get_qdisc_backlog()
+{
+	local vlan=$1; shift
+
+	qdisc_stats_get $swp3 $(get_qdisc_handle $vlan) .backlog
+}
+
+get_mc_transmit_queue()
+{
+	local vlan=$1; shift
+
+	local tc=$(($(get_tc $vlan) + 8))
+	ethtool_stats_get $swp3 tc_transmit_queue_tc_$tc
+}
+
+get_nmarked()
+{
+	local vlan=$1; shift
+
+	ethtool_stats_get $swp3 ecn_marked
+}
+
+get_qdisc_npackets()
+{
+	local vlan=$1; shift
+
+	busywait_for_counter 1100 +1 \
+		qdisc_stats_get $swp3 $(get_qdisc_handle $vlan) .packets
+}
+
+# This sends traffic in an attempt to build a backlog of $size. Returns 0 on
+# success. After 10 failed attempts it bails out and returns 1. It dumps the
+# backlog size to stdout.
+build_backlog()
+{
+	local vlan=$1; shift
+	local size=$1; shift
+	local proto=$1; shift
+
+	local tc=$((vlan - 10))
+	local band=$((8 - tc))
+	local cur=-1
+	local i=0
+
+	while :; do
+		local cur=$(busywait 1100 until_counter_is $((cur + 1)) \
+					    get_qdisc_backlog $vlan)
+		local diff=$((size - cur))
+		local pkts=$(((diff + 7999) / 8000))
+
+		if ((cur >= size)); then
+			echo $cur
+			return 0
+		elif ((i++ > 10)); then
+			echo $cur
+			return 1
+		fi
+
+		$MZ $h2.$vlan -p 8000 -a own -b $h3_mac \
+		    -A $(ipaddr 2 $vlan) -B $(ipaddr 3 $vlan) \
+		    -t $proto -q -c $pkts "$@"
+	done
+}
+
+check_marking()
+{
+	local vlan=$1; shift
+	local cond=$1; shift
+
+	local npackets_0=$(get_qdisc_npackets $vlan)
+	local nmarked_0=$(get_nmarked $vlan)
+	sleep 5
+	local npackets_1=$(get_qdisc_npackets $vlan)
+	local nmarked_1=$(get_nmarked $vlan)
+
+	local nmarked_d=$((nmarked_1 - nmarked_0))
+	local npackets_d=$((npackets_1 - npackets_0))
+	local pct=$((100 * nmarked_d / npackets_d))
+
+	echo $pct
+	((pct $cond))
+}
+
+do_ecn_test()
+{
+	local vlan=$1; shift
+	local limit=$1; shift
+	local backlog
+	local pct
+
+	# Main stream.
+	start_tcp_traffic $h1.$vlan $(ipaddr 1 $vlan) $(ipaddr 3 $vlan) \
+			  $h3_mac tos=0x01
+
+	# Build the below-the-limit backlog using UDP. We could use TCP just
+	# fine, but this way we get a proof that UDP is accepted when queue
+	# length is below the limit. The main stream is using TCP, and if the
+	# limit is misconfigured, we would see this traffic being ECN marked.
+	RET=0
+	backlog=$(build_backlog $vlan $((2 * limit / 3)) udp)
+	check_err $? "Could not build the requested backlog"
+	pct=$(check_marking $vlan "== 0")
+	check_err $? "backlog $backlog / $limit Got $pct% marked packets, expected == 0."
+	log_test "TC $((vlan - 10)): ECN backlog < limit"
+
+	# Now push TCP, because non-TCP traffic would be early-dropped after the
+	# backlog crosses the limit, and we want to make sure that the backlog
+	# is above the limit.
+	RET=0
+	backlog=$(build_backlog $vlan $((3 * limit / 2)) tcp tos=0x01)
+	check_err $? "Could not build the requested backlog"
+	pct=$(check_marking $vlan ">= 95")
+	check_err $? "backlog $backlog / $limit Got $pct% marked packets, expected >= 95."
+	log_test "TC $((vlan - 10)): ECN backlog > limit"
+
+	# Up there we saw that UDP gets accepted when backlog is below the
+	# limit. Now that it is above, it should all get dropped, and backlog
+	# building should fail.
+	RET=0
+	build_backlog $vlan $((2 * limit)) udp >/dev/null
+	check_fail $? "UDP traffic went into backlog instead of being early-dropped"
+	log_test "TC $((vlan - 10)): ECN backlog > limit: UDP early-dropped"
+
+	stop_traffic
+	sleep 1
+}
+
+do_red_test()
+{
+	local vlan=$1; shift
+	local limit=$1; shift
+	local backlog
+	local pct
+
+	# Use ECN-capable TCP to verify there's no marking even though the queue
+	# is above limit.
+	start_tcp_traffic $h1.$vlan $(ipaddr 1 $vlan) $(ipaddr 3 $vlan) \
+			  $h3_mac tos=0x01
+
+	# Pushing below the queue limit should work.
+	RET=0
+	backlog=$(build_backlog $vlan $((2 * limit / 3)) tcp tos=0x01)
+	check_err $? "Could not build the requested backlog"
+	pct=$(check_marking $vlan "== 0")
+	check_err $? "backlog $backlog / $limit Got $pct% marked packets, expected == 0."
+	log_test "TC $((vlan - 10)): RED backlog < limit"
+
+	# Pushing above should not.
+	RET=0
+	backlog=$(build_backlog $vlan $((3 * limit / 2)) tcp tos=0x01)
+	check_fail $? "Traffic went into backlog instead of being early-dropped"
+	pct=$(check_marking $vlan "== 0")
+	check_err $? "backlog $backlog / $limit Got $pct% marked packets, expected == 0."
+	local diff=$((limit - backlog))
+	pct=$((100 * diff / limit))
+	((0 <= pct && pct <= 5))
+	check_err $? "backlog $backlog / $limit expected <= 5% distance"
+	log_test "TC $((vlan - 10)): RED backlog > limit"
+
+	stop_traffic
+	sleep 1
+}
+
+do_mc_backlog_test()
+{
+	local vlan=$1; shift
+	local limit=$1; shift
+	local backlog
+	local pct
+
+	RET=0
+
+	start_tcp_traffic $h1.$vlan $(ipaddr 1 $vlan) $(ipaddr 3 $vlan) bc
+	start_tcp_traffic $h2.$vlan $(ipaddr 2 $vlan) $(ipaddr 3 $vlan) bc
+
+	qbl=$(busywait 5000 until_counter_is 500000 \
+		       get_qdisc_backlog $vlan)
+	check_err $? "Could not build MC backlog"
+
+	# Verify that we actually see the backlog on BUM TC. Do a busywait as
+	# well, performance blips might cause false fail.
+	local ebl
+	ebl=$(busywait 5000 until_counter_is 500000 \
+		       get_mc_transmit_queue $vlan)
+	check_err $? "MC backlog reported by qdisc not visible in ethtool"
+
+	stop_traffic
+	stop_traffic
+
+	log_test "TC $((vlan - 10)): Qdisc reports MC backlog"
+}
diff --git a/tools/testing/selftests/drivers/net/mlxsw/sch_red_ets.sh b/tools/testing/selftests/drivers/net/mlxsw/sch_red_ets.sh
new file mode 100755
index 000000000000..af83efe9ccf1
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/mlxsw/sch_red_ets.sh
@@ -0,0 +1,83 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+ALL_TESTS="
+	ping_ipv4
+	ecn_test
+	red_test
+	mc_backlog_test
+"
+: ${QDISC:=ets}
+source sch_red_core.sh
+
+# do_ecn_test first build 2/3 of the requested backlog and expects no marking,
+# and then builds 3/2 of it and does expect marking. The values of $BACKLOG1 and
+# $BACKLOG2 are far enough not to overlap, so that we can assume that if we do
+# see (do not see) marking, it is actually due to the configuration of that one
+# TC, and not due to configuration of the other TC leaking over.
+BACKLOG1=200000
+BACKLOG2=500000
+
+install_qdisc()
+{
+	local -a args=("$@")
+
+	tc qdisc add dev $swp3 root handle 10: $QDISC \
+	   bands 8 priomap 7 6 5 4 3 2 1 0
+	tc qdisc add dev $swp3 parent 10:8 handle 108: red \
+	   limit 1000000 min $BACKLOG1 max $((BACKLOG1 + 1)) \
+	   probability 1.0 avpkt 8000 burst 38 "${args[@]}"
+	tc qdisc add dev $swp3 parent 10:7 handle 107: red \
+	   limit 1000000 min $BACKLOG2 max $((BACKLOG2 + 1)) \
+	   probability 1.0 avpkt 8000 burst 63 "${args[@]}"
+	sleep 1
+}
+
+uninstall_qdisc()
+{
+	tc qdisc del dev $swp3 parent 10:7
+	tc qdisc del dev $swp3 parent 10:8
+	tc qdisc del dev $swp3 root
+}
+
+ecn_test()
+{
+	install_qdisc ecn
+
+	do_ecn_test 10 $BACKLOG1
+	do_ecn_test 11 $BACKLOG2
+
+	uninstall_qdisc
+}
+
+red_test()
+{
+	install_qdisc
+
+	do_red_test 10 $BACKLOG1
+	do_red_test 11 $BACKLOG2
+
+	uninstall_qdisc
+}
+
+mc_backlog_test()
+{
+	install_qdisc
+
+	# Note that the backlog numbe