aboutsummaryrefslogtreecommitdiffstats
path: root/tests/vpp/perf
diff options
context:
space:
mode:
Diffstat (limited to 'tests/vpp/perf')
-rw-r--r--tests/vpp/perf/__init__.robot45
-rw-r--r--tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-int-aes-gcm-ndrpdrdisc.robot337
-rw-r--r--tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-int-cbc-sha1-ndrpdrdisc.robot333
-rw-r--r--tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-tnl-aes-gcm-ndrpdrdisc.robot361
-rw-r--r--tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-tnl-cbc-sha1-ndrpdrdisc.robot357
-rw-r--r--tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-int-aes-gcm-ndrpdrdisc.robot337
-rw-r--r--tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-int-cbc-sha1-ndrpdrdisc.robot333
-rw-r--r--tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-tnl-aes-gcm-ndrpdrdisc.robot361
-rw-r--r--tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-tnl-cbc-sha1-ndrpdrdisc.robot357
-rw-r--r--tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsectptlispgpe-ip4base-cbc-sha1-ndrpdrdisc.robot374
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-copwhtlistbase-ndrchk.robot260
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-copwhtlistbase-ndrpdrdisc.robot569
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-iacldstbase-ndrchk.robot332
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-iacldstbase-ndrpdrdisc.robot713
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ipolicemarkbase-ndrchk.robot226
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ipolicemarkbase-ndrpdrdisc.robot517
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ndrchk.robot198
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ndrpdrdisc.robot489
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-pdrchk.robot190
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-snat-ndrpdrdisc.robot227
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale200k-ndrchk.robot201
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale200k-ndrpdrdisc.robot457
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale20k-ndrchk.robot210
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale20k-ndrpdrdisc.robot457
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale2m-ndrchk.robot210
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale2m-ndrpdrdisc.robot457
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale2m-pdrchk.robot185
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4base-udpsrcscale15-snat-ndrpdrdisc.robot229
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale10-udpsrcscale15-snat-ndrpdrdisc.robot230
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale100-udpsrcscale15-snat-ndrpdrdisc.robot230
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale1000-udpsrcscale15-snat-ndrpdrdisc.robot230
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale2000-udpsrcscale15-snat-ndrpdrdisc.robot124
-rw-r--r--tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale4000-udpsrcscale15-snat-ndrpdrdisc.robot124
-rw-r--r--tests/vpp/perf/ip4/40ge2p1xl710-ethip4-ip4base-ndrpdrdisc.robot257
-rw-r--r--tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip4-ip4base-ndrchk.robot266
-rw-r--r--tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip4-ip4base-ndrpdrdisc.robot595
-rw-r--r--tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip4-ip4base-pdrchk.robot193
-rw-r--r--tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip6-ip4base-ndrchk.robot266
-rw-r--r--tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip6-ip4base-ndrpdrdisc.robot596
-rw-r--r--tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip6-ip4base-pdrchk.robot193
-rw-r--r--tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-ndrpdrdisc.robot487
-rw-r--r--tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-pdrchk.robot186
-rw-r--r--tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2xcbase-ndrchk.robot190
-rw-r--r--tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2xcbase-ndrpdrdisc.robot432
-rw-r--r--tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2xcbase-pdrchk.robot186
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-copwhtlistbase-ndrchk.robot256
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-copwhtlistbase-ndrpdrdisc.robot565
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-iacldstbase-ndrchk.robot210
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-iacldstbase-ndrpdrdisc.robot474
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-ndrchk.robot191
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-ndrpdrdisc.robot436
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-pdrchk.robot183
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale200k-ndrchk.robot201
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale200k-ndrpdrdisc.robot457
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale20k-ndrchk.robot201
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale20k-ndrpdrdisc.robot457
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale2m-ndrchk.robot210
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale2m-ndrpdrdisc.robot457
-rw-r--r--tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale2m-pdrchk.robot185
-rw-r--r--tests/vpp/perf/ip6/40ge2p1xl710-ethip6-ip6base-ndrpdrdisc.robot183
-rw-r--r--tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip4-ip6base-ndrchk.robot266
-rw-r--r--tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip4-ip6base-ndrpdrdisc.robot596
-rw-r--r--tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip4-ip6base-pdrchk.robot193
-rw-r--r--tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip6-ip6base-ndrchk.robot266
-rw-r--r--tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip6-ip6base-ndrpdrdisc.robot596
-rw-r--r--tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip6-ip6base-pdrchk.robot193
-rw-r--r--tests/vpp/perf/l2/10ge2p1vic1227-eth-l2bdbasemaclrn-ndrpdrdisc.robot288
-rw-r--r--tests/vpp/perf/l2/10ge2p1x520-dot1ad-l2xcbase-ndrchk.robot279
-rw-r--r--tests/vpp/perf/l2/10ge2p1x520-dot1ad-l2xcbase-ndrpdrdisc.robot606
-rw-r--r--tests/vpp/perf/l2/10ge2p1x520-dot1q-l2xcbase-ndrchk.robot266
-rw-r--r--tests/vpp/perf/l2/10ge2p1x520-dot1q-l2xcbase-ndrpdrdisc.robot584
-rw-r--r--tests/vpp/perf/l2/10ge2p1x520-eth-l2bdbasemaclrn-ndrchk.robot192
-rw-r--r--tests/vpp/perf/l2/10ge2p1x520-eth-l2bdbasemaclrn-ndrpdrdisc.robot274
-rw-r--r--tests/vpp/perf/l2/10ge2p1x520-eth-l2bdbasemaclrn-pdrchk.robot184
-rw-r--r--tests/vpp/perf/l2/10ge2p1x520-eth-l2xcbase-ndrchk.robot192
-rw-r--r--tests/vpp/perf/l2/10ge2p1x520-eth-l2xcbase-ndrpdrdisc.robot483
-rw-r--r--tests/vpp/perf/l2/10ge2p1x520-eth-l2xcbase-pdrchk.robot184
-rw-r--r--tests/vpp/perf/l2/10ge2p1x710-eth-l2bdbasemaclrn-ndrpdrdisc.robot274
-rw-r--r--tests/vpp/perf/l2/40ge2p1vic1385-eth-l2bdbasemaclrn-ndrpdrdisc.robot352
-rw-r--r--tests/vpp/perf/l2/40ge2p1xl710-eth-l2bdbasemaclrn-ndrpdrdisc.robot232
-rw-r--r--tests/vpp/perf/l2/40ge2p1xl710-eth-l2xcbase-ndrpdrdisc.robot251
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot705
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrchk.robot212
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2xcbase-eth-2vhost-1vm-ndrpdrdisc.robot684
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2xcbase-eth-2vhost-1vm-pdrchk.robot209
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot623
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrpdrdisc.robot548
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2xcbase-eth-2vhost-1vm-ndrpdrdisc.robot621
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2xcbase-eth-2vhost-1vm-pdrchk.robot208
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2xcbase-eth-4vhost-2vm-ndrpdrdisc.robot549
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4-ip4base-eth-2vhost-1vm-ndrpdrdisc.robot658
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4-ip4base-eth-2vhost-1vm-pdrchk.robot209
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4-ip4base-eth-4vhost-2vm-ndrpdrdisc.robot550
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot675
-rw-r--r--tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot658
-rw-r--r--tests/vpp/perf/vm_vhost/40ge2p1xl710-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot660
-rw-r--r--tests/vpp/perf/vm_vhost/40ge2p1xl710-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrpdrdisc.robot588
-rw-r--r--tests/vpp/perf/vm_vhost/40ge2p1xl710-eth-l2xcbase-eth-4vhost-2vm-ndrpdrdisc.robot588
-rw-r--r--tests/vpp/perf/vm_vhost/40ge2p1xl710-ethip4-ip4base-eth-4vhost-2vm-ndrpdrdisc.robot588
99 files changed, 35057 insertions, 0 deletions
diff --git a/tests/vpp/perf/__init__.robot b/tests/vpp/perf/__init__.robot
new file mode 100644
index 0000000000..0dfc31f07e
--- /dev/null
+++ b/tests/vpp/perf/__init__.robot
@@ -0,0 +1,45 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/shared/default.robot
+| Resource | resources/libraries/robot/shared/interfaces.robot
+| Library | resources.libraries.python.SetupFramework
+| Library | resources.libraries.python.CpuUtils
+| Suite Setup | Run Keywords | Setup performance global Variables
+| ... | AND | Setup Framework | ${nodes}
+| ... | AND | Setup All DUTs | ${nodes}
+| ... | AND | Get CPU Layout from all nodes | ${nodes}
+| ... | AND | Update All Interface Data On All Nodes
+| ... | ${nodes} | skip_tg=${True} | numa_node=${True}
+
+*** Keywords ***
+| Setup performance global Variables
+| | [Documentation]
+| | ... | Setup suite Variables. Variables are used across performance testing.
+| | ...
+| | ... | _NOTE:_ This KW sets following suite variables:
+| | ... | - perf_trial_duration - Duration of traffic run [s]
+| | ... | - perf_pdr_loss_acceptance - Loss acceptance treshold
+| | ... | - perf_pdr_loss_acceptance_type - Loss acceptance treshold type
+| | ... | - perf_vm_image - Guest VM disk image
+| | ... | - perf_qemu_bin - Path to QEMU binary
+| | ... | - perf_qemu_qsz - QEMU virtio queue size
+| | ...
+| | Set Global Variable | ${perf_trial_duration} | 10
+| | Set Global Variable | ${perf_pdr_loss_acceptance} | 0.5
+| | Set Global Variable | ${perf_pdr_loss_acceptance_type} | percentage
+| | Set Global Variable | ${perf_vm_image} | /var/lib/vm/csit-nested-1.6.img
+| | Set Global Variable | ${perf_qemu_bin}
+| | ... | /opt/qemu-2.5.0/bin/qemu-system-x86_64
+| | Set Global Variable | ${perf_qemu_qsz} | 1024
diff --git a/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-int-aes-gcm-ndrpdrdisc.robot b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-int-aes-gcm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..86670fecc6
--- /dev/null
+++ b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-int-aes-gcm-ndrpdrdisc.robot
@@ -0,0 +1,337 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/crypto/ipsec.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | IP4FWD | IPSEC | IPSECHW | IPSECTUN | NIC_Intel-XL710 | BASE
+| ...
+| Suite Setup | Set up IPSec performance test suite | L3 | Intel-XL710
+| ...
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *IPv4 IPsec tunnel mode performance test suite.*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 on TG-DUTn,
+| ... | Eth-IPv4-IPSec on DUT1-DUT2
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with multiple
+| ... | IPsec tunnels between them. DUTs get IPv4 traffic from TG, encrypt it
+| ... | and send to another DUT, where packets are decrypted and sent back to TG
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in number
+| ... | of packets transmitted. NDR is discovered for different
+| ... | number of IPsec tunnels using binary search algorithms with configured
+| ... | starting rate and final step that determines throughput measurement
+| ... | resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, number of flows per flow-group equals to
+| ... | number of IPSec tunnels) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | static payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC4303 and RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+| ${tg_if1_ip4}= | 192.168.10.2
+| ${dut1_if1_ip4}= | 192.168.10.1
+| ${dut1_if2_ip4}= | 172.168.1.1
+| ${dut2_if1_ip4}= | 172.168.1.2
+| ${dut2_if2_ip4}= | 192.168.20.1
+| ${tg_if2_ip4}= | 192.168.20.2
+| ${raddr_ip4}= | 20.0.0.0
+| ${laddr_ip4}= | 10.0.0.0
+| ${addr_range}= | ${32}
+| ${ipsec_overhead_gcm}= | ${54}
+| ${n_tunnels}= | ${1}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${n_tunnels}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4ipsecbasetnl-ip4base-int-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4ipsecbasetnl-ip4base-int-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4ipsecbasetnl-ip4base-int-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${framesize + ${ipsec_overhead_gcm}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4ipsecbasetnl-ip4base-int-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${framesize + ${ipsec_overhead_gcm}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4ipsecbasetnl-ip4base-int-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${imix_size} + ${ipsec_overhead_gcm}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4ipsecbasetnl-ip4base-int-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${imix_size} + ${ipsec_overhead_gcm}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4ipsecbasetnl-ip4base-int-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4ipsecbasetnl-ip4base-int-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-int-cbc-sha1-ndrpdrdisc.robot b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-int-cbc-sha1-ndrpdrdisc.robot
new file mode 100644
index 0000000000..c78d3ca67d
--- /dev/null
+++ b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-int-cbc-sha1-ndrpdrdisc.robot
@@ -0,0 +1,333 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/crypto/ipsec.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | IP4FWD | IPSEC | IPSECHW | IPSECTUN | NIC_Intel-XL710 | BASE
+| ...
+| Suite Setup | Set up IPSec performance test suite | L3 | Intel-XL710
+| ...
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *IPv4 IPsec tunnel mode performance test suite.*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 on TG-DUTn,
+| ... | Eth-IPv4-IPSec on DUT1-DUT2
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with multiple
+| ... | IPsec tunnels between them. DUTs get IPv4 traffic from TG, encrypt it
+| ... | and send to another DUT, where packets are decrypted and sent back to TG
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in number
+| ... | of packets transmitted. NDR is discovered for different
+| ... | number of IPsec tunnels using binary search algorithms with configured
+| ... | starting rate and final step that determines throughput measurement
+| ... | resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, number of flows per flow-group equals to
+| ... | number of IPSec tunnels) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | static payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC4303 and RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+| ${tg_if1_ip4}= | 192.168.10.2
+| ${dut1_if1_ip4}= | 192.168.10.1
+| ${dut1_if2_ip4}= | 172.168.1.1
+| ${dut2_if1_ip4}= | 172.168.1.2
+| ${dut2_if2_ip4}= | 192.168.20.1
+| ${tg_if2_ip4}= | 192.168.20.2
+| ${raddr_ip4}= | 20.0.0.0
+| ${laddr_ip4}= | 10.0.0.0
+| ${addr_range}= | ${32}
+| ${ipsec_overhead}= | ${58}
+| ${n_tunnels}= | ${1}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${n_tunnels}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4ipsecbasetnl-ip4base-int-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4ipsecbasetnl-ip4base-int-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4ipsecbasetnl-ip4base-int-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize + ${ipsec_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4ipsecbasetnl-ip4base-int-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize + ${ipsec_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4ipsecbasetnl-ip4base-int-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${imix_size} + ${ipsec_overhead}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4ipsecbasetnl-ip4base-int-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${imix_size} + ${ipsec_overhead}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4ipsecbasetnl-ip4base-int-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4ipsecbasetnl-ip4base-int-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-tnl-aes-gcm-ndrpdrdisc.robot b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-tnl-aes-gcm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..85ddc9f663
--- /dev/null
+++ b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-tnl-aes-gcm-ndrpdrdisc.robot
@@ -0,0 +1,361 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/crypto/ipsec.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | IP4FWD | IPSEC | IPSECHW | IPSECTUN | NIC_Intel-XL710 | BASE
+| ...
+| Suite Setup | Set up IPSec performance test suite | L3 | Intel-XL710
+| ...
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *IPv4 IPsec tunnel mode performance test suite.*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 on TG-DUTn,
+| ... | Eth-IPv4-IPSec on DUT1-DUT2
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with multiple
+| ... | IPsec tunnels between them. DUTs get IPv4 traffic from TG, encrypt it
+| ... | and send to another DUT, where packets are decrypted and sent back to TG
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in number
+| ... | of packets transmitted. NDR is discovered for different
+| ... | number of IPsec tunnels using binary search algorithms with configured
+| ... | starting rate and final step that determines throughput measurement
+| ... | resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, number of flows per flow-group equals to
+| ... | number of IPSec tunnels) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | static payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC4303 and RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+| ${tg_if1_ip4}= | 192.168.10.2
+| ${dut1_if1_ip4}= | 192.168.10.1
+| ${dut1_if2_ip4}= | 172.168.1.1
+| ${dut2_if1_ip4}= | 172.168.1.2
+| ${dut2_if2_ip4}= | 192.168.20.1
+| ${tg_if2_ip4}= | 192.168.20.2
+| ${raddr_ip4}= | 20.0.0.0
+| ${laddr_ip4}= | 10.0.0.0
+| ${addr_range}= | ${32}
+| ${ipsec_overhead_gcm}= | ${54}
+| ${n_tunnels}= | ${1}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${n_tunnels}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4ipsecbasetnl-ip4base-tnl-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4ipsecbasetnl-ip4base-tnl-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4ipsecbasetnl-ip4base-tnl-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${framesize + ${ipsec_overhead_gcm}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4ipsecbasetnl-ip4base-tnl-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${framesize + ${ipsec_overhead_gcm}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4ipsecbasetnl-ip4base-tnl-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${imix_size} + ${ipsec_overhead_gcm}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4ipsecbasetnl-ip4base-tnl-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${imix_size} + ${ipsec_overhead_gcm}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4ipsecbasetnl-ip4base-tnl-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4ipsecbasetnl-ip4base-tnl-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel AES GCM in each direction, configured\
+| | ... | with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-tnl-cbc-sha1-ndrpdrdisc.robot b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-tnl-cbc-sha1-ndrpdrdisc.robot
new file mode 100644
index 0000000000..1002f53e5d
--- /dev/null
+++ b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecbasetnl-ip4base-tnl-cbc-sha1-ndrpdrdisc.robot
@@ -0,0 +1,357 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/crypto/ipsec.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | IP4FWD | IPSEC | IPSECHW | IPSECTUN | NIC_Intel-XL710 | BASE
+| ...
+| Suite Setup | Set up IPSec performance test suite | L3 | Intel-XL710
+| ...
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *IPv4 IPsec tunnel mode performance test suite.*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 on TG-DUTn,
+| ... | Eth-IPv4-IPSec on DUT1-DUT2
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with multiple
+| ... | IPsec tunnels between them. DUTs get IPv4 traffic from TG, encrypt it
+| ... | and send to another DUT, where packets are decrypted and sent back to TG
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in number
+| ... | of packets transmitted. NDR is discovered for different
+| ... | number of IPsec tunnels using binary search algorithms with configured
+| ... | starting rate and final step that determines throughput measurement
+| ... | resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, number of flows per flow-group equals to
+| ... | number of IPSec tunnels) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | static payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC4303 and RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+| ${tg_if1_ip4}= | 192.168.10.2
+| ${dut1_if1_ip4}= | 192.168.10.1
+| ${dut1_if2_ip4}= | 172.168.1.1
+| ${dut2_if1_ip4}= | 172.168.1.2
+| ${dut2_if2_ip4}= | 192.168.20.1
+| ${tg_if2_ip4}= | 192.168.20.2
+| ${raddr_ip4}= | 20.0.0.0
+| ${laddr_ip4}= | 10.0.0.0
+| ${addr_range}= | ${32}
+| ${ipsec_overhead}= | ${58}
+| ${n_tunnels}= | ${1}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${n_tunnels}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4ipsecbasetnl-ip4base-tnl-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4ipsecbasetnl-ip4base-tnl-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4ipsecbasetnl-ip4base-tnl-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize + ${ipsec_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4ipsecbasetnl-ip4base-tnl-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize + ${ipsec_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4ipsecbasetnl-ip4base-tnl-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${imix_size} + ${ipsec_overhead}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4ipsecbasetnl-ip4base-tnl-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${imix_size} + ${ipsec_overhead}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4ipsecbasetnl-ip4base-tnl-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4ipsecbasetnl-ip4base-tnl-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1 IPsec tunnel CBC-SHA1 in each direction, configured\
+| | ... | with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-int-aes-gcm-ndrpdrdisc.robot b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-int-aes-gcm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..ac8f8603b9
--- /dev/null
+++ b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-int-aes-gcm-ndrpdrdisc.robot
@@ -0,0 +1,337 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/crypto/ipsec.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | SCALE | NDRPDRDISC
+| ... | IP4FWD | IPSEC | IPSECHW | IPSECTUN | NIC_Intel-XL710 | TNL_1000
+| ...
+| Suite Setup | Set up IPSec performance test suite | L3 | Intel-XL710
+| ...
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *IPv4 IPsec tunnel mode performance test suite.*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 on TG-DUTn,
+| ... | Eth-IPv4-IPSec on DUT1-DUT2
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with multiple
+| ... | IPsec tunnels between them. DUTs get IPv4 traffic from TG, encrypt it
+| ... | and send to another DUT, where packets are decrypted and sent back to TG
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in number
+| ... | of packets transmitted. NDR is discovered for different
+| ... | number of IPsec tunnels using binary search algorithms with configured
+| ... | starting rate and final step that determines throughput measurement
+| ... | resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, number of flows per flow-group equals to
+| ... | number of IPSec tunnels) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | static payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC4303 and RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+| ${tg_if1_ip4}= | 192.168.10.2
+| ${dut1_if1_ip4}= | 192.168.10.1
+| ${dut1_if2_ip4}= | 172.168.1.1
+| ${dut2_if1_ip4}= | 172.168.1.2
+| ${dut2_if2_ip4}= | 192.168.20.1
+| ${tg_if2_ip4}= | 192.168.20.2
+| ${raddr_ip4}= | 20.0.0.0
+| ${laddr_ip4}= | 10.0.0.0
+| ${addr_range}= | ${32}
+| ${ipsec_overhead_gcm}= | ${54}
+| ${n_tunnels}= | ${1000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${n_tunnels}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4ipsecscale1000tnl-ip4base-int-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4ipsecscale1000tnl-ip4base-int-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4ipsecscale1000tnl-ip4base-int-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${framesize + ${ipsec_overhead_gcm}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4ipsecscale1000tnl-ip4base-int-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnel AES GCM in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${framesize + ${ipsec_overhead_gcm}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4ipsecscale1000tnl-ip4base-int-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${imix_size} + ${ipsec_overhead_gcm}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4ipsecscale1000tnl-ip4base-int-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${imix_size} + ${ipsec_overhead_gcm}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4ipsecscale1000tnl-ip4base-int-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4ipsecscale1000tnl-ip4base-int-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-int-cbc-sha1-ndrpdrdisc.robot b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-int-cbc-sha1-ndrpdrdisc.robot
new file mode 100644
index 0000000000..c46da5a82d
--- /dev/null
+++ b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-int-cbc-sha1-ndrpdrdisc.robot
@@ -0,0 +1,333 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/crypto/ipsec.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | SCALE | NDRPDRDISC
+| ... | IP4FWD | IPSEC | IPSECHW | IPSECTUN | NIC_Intel-XL710 | TNL_1000
+| ...
+| Suite Setup | Set up IPSec performance test suite | L3 | Intel-XL710
+| ...
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *IPv4 IPsec tunnel mode performance test suite.*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 on TG-DUTn,
+| ... | Eth-IPv4-IPSec on DUT1-DUT2
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with multiple
+| ... | IPsec tunnels between them. DUTs get IPv4 traffic from TG, encrypt it
+| ... | and send to another DUT, where packets are decrypted and sent back to TG
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in number
+| ... | of packets transmitted. NDR is discovered for different
+| ... | number of IPsec tunnels using binary search algorithms with configured
+| ... | starting rate and final step that determines throughput measurement
+| ... | resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, number of flows per flow-group equals to
+| ... | number of IPSec tunnels) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | static payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC4303 and RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+| ${tg_if1_ip4}= | 192.168.10.2
+| ${dut1_if1_ip4}= | 192.168.10.1
+| ${dut1_if2_ip4}= | 172.168.1.1
+| ${dut2_if1_ip4}= | 172.168.1.2
+| ${dut2_if2_ip4}= | 192.168.20.1
+| ${tg_if2_ip4}= | 192.168.20.2
+| ${raddr_ip4}= | 20.0.0.0
+| ${laddr_ip4}= | 10.0.0.0
+| ${addr_range}= | ${32}
+| ${ipsec_overhead}= | ${58}
+| ${n_tunnels}= | ${1000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${n_tunnels}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4ipsecscale1000tnl-ip4base-int-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4ipsecscale1000tnl-ip4base-int-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4ipsecscale1000tnl-ip4base-int-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize + ${ipsec_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4ipsecscale1000tnl-ip4base-int-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnel CBC-SHA1 in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize + ${ipsec_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4ipsecscale1000tnl-ip4base-int-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${imix_size} + ${ipsec_overhead}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4ipsecscale1000tnl-ip4base-int-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${imix_size} + ${ipsec_overhead}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4ipsecscale1000tnl-ip4base-int-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4ipsecscale1000tnl-ip4base-int-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | And VPP IPsec Create Tunnel Interfaces
+| | ... | ${dut1} | ${dut2} | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key} | ${laddr_ip4}
+| | ... | ${raddr_ip4} | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-tnl-aes-gcm-ndrpdrdisc.robot b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-tnl-aes-gcm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..6e5a7dfbd9
--- /dev/null
+++ b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-tnl-aes-gcm-ndrpdrdisc.robot
@@ -0,0 +1,361 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/crypto/ipsec.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | SCALE | NDRPDRDISC
+| ... | IP4FWD | IPSEC | IPSECHW | IPSECTUN | NIC_Intel-XL710 | TNL_1000
+| ...
+| Suite Setup | Set up IPSec performance test suite | L3 | Intel-XL710
+| ...
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *IPv4 IPsec tunnel mode performance test suite.*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 on TG-DUTn,
+| ... | Eth-IPv4-IPSec on DUT1-DUT2
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with multiple
+| ... | IPsec tunnels between them. DUTs get IPv4 traffic from TG, encrypt it
+| ... | and send to another DUT, where packets are decrypted and sent back to TG
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in number
+| ... | of packets transmitted. NDR is discovered for different
+| ... | number of IPsec tunnels using binary search algorithms with configured
+| ... | starting rate and final step that determines throughput measurement
+| ... | resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, number of flows per flow-group equals to
+| ... | number of IPSec tunnels) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | static payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC4303 and RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+| ${tg_if1_ip4}= | 192.168.10.2
+| ${dut1_if1_ip4}= | 192.168.10.1
+| ${dut1_if2_ip4}= | 172.168.1.1
+| ${dut2_if1_ip4}= | 172.168.1.2
+| ${dut2_if2_ip4}= | 192.168.20.1
+| ${tg_if2_ip4}= | 192.168.20.2
+| ${raddr_ip4}= | 20.0.0.0
+| ${laddr_ip4}= | 10.0.0.0
+| ${addr_range}= | ${32}
+| ${ipsec_overhead_gcm}= | ${54}
+| ${n_tunnels}= | ${1000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${n_tunnels}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4ipsecscale1000tnl-ip4base-tnl-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4ipsecscale1000tnl-ip4base-tnl-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4ipsecscale1000tnl-ip4base-tnl-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${framesize + ${ipsec_overhead_gcm}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4ipsecscale1000tnl-ip4base-tnl-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnel AES GCM in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${framesize + ${ipsec_overhead_gcm}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4ipsecscale1000tnl-ip4base-tnl-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${imix_size} + ${ipsec_overhead_gcm}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4ipsecscale1000tnl-ip4base-tnl-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G}
+| | ... | ${imix_size} + ${ipsec_overhead_gcm}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4ipsecscale1000tnl-ip4base-tnl-aes-gcm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4ipsecscale1000tnl-ip4base-tnl-aes-gcm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels AES GCM in each direction,\
+| | ... | configured with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES GCM 128
+| | ${auth_alg}= | Integ Alg AES GCM 128
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-tnl-cbc-sha1-ndrpdrdisc.robot b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-tnl-cbc-sha1-ndrpdrdisc.robot
new file mode 100644
index 0000000000..cceaaf846a
--- /dev/null
+++ b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsecscale1000tnl-ip4base-tnl-cbc-sha1-ndrpdrdisc.robot
@@ -0,0 +1,357 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/crypto/ipsec.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | SCALE | NDRPDRDISC
+| ... | IP4FWD | IPSEC | IPSECHW | IPSECTUN | NIC_Intel-XL710 | TNL_1000
+| ...
+| Suite Setup | Set up IPSec performance test suite | L3 | Intel-XL710
+| ...
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *IPv4 IPsec tunnel mode performance test suite.*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 on TG-DUTn,
+| ... | Eth-IPv4-IPSec on DUT1-DUT2
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with multiple
+| ... | IPsec tunnels between them. DUTs get IPv4 traffic from TG, encrypt it
+| ... | and send to another DUT, where packets are decrypted and sent back to TG
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in number
+| ... | of packets transmitted. NDR is discovered for different
+| ... | number of IPsec tunnels using binary search algorithms with configured
+| ... | starting rate and final step that determines throughput measurement
+| ... | resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, number of flows per flow-group equals to
+| ... | number of IPSec tunnels) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | static payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC4303 and RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+| ${tg_if1_ip4}= | 192.168.10.2
+| ${dut1_if1_ip4}= | 192.168.10.1
+| ${dut1_if2_ip4}= | 172.168.1.1
+| ${dut2_if1_ip4}= | 172.168.1.2
+| ${dut2_if2_ip4}= | 192.168.20.1
+| ${tg_if2_ip4}= | 192.168.20.2
+| ${raddr_ip4}= | 20.0.0.0
+| ${laddr_ip4}= | 10.0.0.0
+| ${addr_range}= | ${32}
+| ${ipsec_overhead}= | ${58}
+| ${n_tunnels}= | ${1000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${n_tunnels}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4ipsecscale1000tnl-ip4base-tnl-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4ipsecscale1000tnl-ip4base-tnl-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4ipsecscale1000tnl-ip4base-tnl-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize + ${ipsec_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4ipsecscale1000tnl-ip4base-tnl-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnel CBC-SHA1 in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize + ${ipsec_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4ipsecscale1000tnl-ip4base-tnl-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${imix_size} + ${ipsec_overhead}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4ipsecscale1000tnl-ip4base-tnl-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${imix_size} + ${ipsec_overhead}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4ipsecscale1000tnl-ip4base-tnl-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find NDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4ipsecscale1000tnl-ip4base-tnl-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUTs run 1000 IPsec tunnels CBC-SHA1 in each direction,\
+| | ... | configured with 2 thread, 2 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps and loss tolerance of 0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPSec in 3-node circular topology
+| | Vpp Route Add | ${dut1} | ${raddr_ip4} | 8 | ${dut2_if1_ip4} | ${dut1_if2}
+| | Vpp Route Add | ${dut2} | ${laddr_ip4} | 8 | ${dut1_if2_ip4} | ${dut2_if1}
+| | And VPP IPsec Add Multiple Tunnels
+| | ... | ${dut1} | ${dut2} | ${dut1_if2} | ${dut2_if1} | ${n_tunnels}
+| | ... | ${encr_alg} | ${encr_key} | ${auth_alg} | ${auth_key}
+| | ... | ${dut1_if2_ip4} | ${dut2_if1_ip4} | ${laddr_ip4} | ${raddr_ip4}
+| | ... | ${addr_range}
+| | Then Find PDR using binary search and pps | ${framesize}
+| | ... | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsectptlispgpe-ip4base-cbc-sha1-ndrpdrdisc.robot b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsectptlispgpe-ip4base-cbc-sha1-ndrpdrdisc.robot
new file mode 100644
index 0000000000..141d3f7ce1
--- /dev/null
+++ b/tests/vpp/perf/crypto/40ge2p1xl710-ethip4ipsectptlispgpe-ip4base-cbc-sha1-ndrpdrdisc.robot
@@ -0,0 +1,374 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/crypto/ipsec.robot
+| Resource | resources/libraries/robot/overlay/lispgpe.robot
+| ...
+| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC | IP4FWD
+| ... | LISPGPE | IPSEC | IPSECHW | IPSECTRAN | ENCAP | IP4UNRLAY | IP4OVRLAY
+| ... | NIC_Intel-XL710
+| ...
+| Suite Setup | Set up IPSec performance test suite | L3 | Intel-XL710
+| ...
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *IPv4 IPsec transport mode performance test suite.*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology\
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 on TG-DUTn,\
+| ... | Eth-IPv4-IPSec-LISPGPE-IPv4 on DUT1-DUT2
+| ... | *[Cfg] DUT configuration:* Each DUT is configured with LISP and IPsec\
+| ... | in each direction. IPsec is in transport mode. DUTs get IPv4 traffic\
+| ... | from TG, encrypt it and send to another DUT, where packets are\
+| ... | decrypted and sent back to TG.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop\
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in number\
+| ... | of packets transmitted. NDR is discovered for different\
+| ... | number of IPsec tunnels using binary search algorithms with configured\
+| ... | starting rate and final step that determines throughput measurement\
+| ... | resolution. Test packets are generated by TG on\
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups\
+| ... | (flow-group per direction, number of flows per flow-group equals to\
+| ... | number of IPSec tunnels) with all packets\
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and\
+| ... | static payload. MAC addresses are matching MAC addresses of the TG\
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address)\
+| ... | field is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC6830, RFC4303 and\
+| ... | RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+| ${dut2_spi}= | ${1000}
+| ${dut1_spi}= | ${1001}
+| ${ESP_PROTO}= | ${50}
+| ${ipsec_overhead}= | ${54}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4ipsectptlispgpe-ip4base-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] Each DUT is configured with LISP and IPsec in each direction.\
+| | ... | IPsec is in transport mode. DUTs are configured with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP GPE IPv4 over IPsec in 3-node circular topology
+| | ... | ${encr_alg} | ${auth_alg}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4ipsectptlispgpe-ip4base-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] Each DUT is configured with LISP and IPsec in each direction.\
+| | ... | IPsec is in transport mode. DUTs are configured with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP GPE IPv4 over IPsec in 3-node circular topology
+| | ... | ${encr_alg} | ${auth_alg}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4ipsectptlispgpe-ip4base-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] Each DUT is configured with LISP and IPsec in each direction.\
+| | ... | IPsec is in transport mode. DUTs are configured with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize + ${ipsec_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP GPE IPv4 over IPsec in 3-node circular topology
+| | ... | ${encr_alg} | ${auth_alg}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4ipsectptlispgpe-ip4base-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] Each DUT is configured with LISP and IPsec in each direction.\
+| | ... | IPsec is in transport mode. DUTs are configured with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize + ${ipsec_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP GPE IPv4 over IPsec in 3-node circular topology
+| | ... | ${encr_alg} | ${auth_alg}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4ipsectptlispgpe-ip4base-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] Each DUT is configured with LISP and IPsec in each direction.\
+| | ... | IPsec is in transport mode. DUTs are configured with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${imix_size} + ${ipsec_overhead}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP GPE IPv4 over IPsec in 3-node circular topology
+| | ... | ${encr_alg} | ${auth_alg}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4ipsectptlispgpe-ip4base-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] Each DUT is configured with LISP and IPsec in each direction.\
+| | ... | IPsec is in transport mode. DUTs are configured with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${imix_size}= | Get Frame Size | ${framesize}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${imix_size} + ${ipsec_overhead}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add cryptodev to all DUTs | ${1}
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP GPE IPv4 over IPsec in 3-node circular topology
+| | ... | ${encr_alg} | ${auth_alg}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4ipsectptlispgpe-ip4base-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] Each DUT is configured with LISP and IPsec in each direction.\
+| | ... | IPsec is in transport mode. DUTs are configured with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP GPE IPv4 over IPsec in 3-node circular topology
+| | ... | ${encr_alg} | ${auth_alg}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4ipsectptlispgpe-ip4base-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] Each DUT is configured with LISP and IPsec in each direction.\
+| | ... | IPsec is in transport mode. DUTs are configured with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${2}
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP GPE IPv4 over IPsec in 3-node circular topology
+| | ... | ${encr_alg} | ${auth_alg}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4ipsectptlispgpe-ip4base-cbc-sha1-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] Each DUT is configured with LISP and IPsec in each direction.\
+| | ... | IPsec is in transport mode. DUTs are configured with 4 threads, 4 phy\
+| | ... | cores, 2 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${4}
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP GPE IPv4 over IPsec in 3-node circular topology
+| | ... | ${encr_alg} | ${auth_alg}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4ipsectptlispgpe-ip4base-cbc-sha1-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] Each DUT is configured with LISP and IPsec in each direction.\
+| | ... | IPsec is in transport mode. DUTs are configured with 4 threads, 4 phy\
+| | ... | cores, 2 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 40GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${encr_alg}= | Crypto Alg AES CBC 128
+| | ${auth_alg}= | Integ Alg SHA1 96
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Generate keys for IPSec | ${encr_alg} | ${auth_alg}
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add cryptodev to all DUTs | ${4}
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP GPE IPv4 over IPsec in 3-node circular topology
+| | ... | ${encr_alg} | ${auth_alg}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-copwhtlistbase-ndrchk.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-copwhtlistbase-ndrchk.robot
new file mode 100644
index 0000000000..5dc941725f
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-copwhtlistbase-ndrchk.robot
@@ -0,0 +1,260 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.Cop
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT1']}
+| ... | WITH NAME | dut1_v4
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT2']}
+| ... | WITH NAME | dut2_v4
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | FEATURE | COPWHLIST
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput IPv4 whitelist verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing, two static IPv4 /24 routes and IPv4 COP security whitelist
+| ... | ingress /24 filter entries applied on links TG - DUT1 and DUT2 - TG.
+| ... | DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 64 Byte frames using single trial throughput test
+| | ... | at 2x 4.1mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 4.1mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip4-ip4base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 1518 Byte frames using single trial throughput test
+| | ... | at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip4-ip4base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 9000 Byte frames using single trial throughput test
+| | ... | at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-64B-2t2c-ethip4-ip4base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 64 Byte frames using single trial throughput test
+| | ... | at 2x 7.2mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 7.2mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip4-ip4base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 1518 Byte frames using single trial throughput test
+| | ... | at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip4-ip4base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 9000 Byte frames using single trial throughput test
+| | ... | at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-64B-4t4c-ethip4-ip4base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 thread, 4 phy core, 2 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 64 Byte frames using single trial throughput test
+| | ... | at 2x 10.4mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 10.4mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip4-ip4base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 thread, 4 phy core, 2 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 1518 Byte frames using single trial throughput test
+| | ... | at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip4-ip4base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 thread, 4 phy core, 2 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 9000 Byte frames using single trial throughput test
+| | ... | at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-copwhtlistbase-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-copwhtlistbase-ndrpdrdisc.robot
new file mode 100644
index 0000000000..2afbb093c8
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-copwhtlistbase-ndrpdrdisc.robot
@@ -0,0 +1,569 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.Cop
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT1']}
+| ... | WITH NAME | dut1_v4
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT2']}
+| ... | WITH NAME | dut2_v4
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | FEATURE | COPWHLIST
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv4 whitelist test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing, two static IPv4 /24 routes and IPv4 COP security whitelist
+| ... | ingress /24 filter entries applied on links TG - DUT1 and DUT2 - TG.
+| ... | DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 64 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4-ip4base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 64 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4-ip4base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4-ip4base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip4-ip4base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip4-ip4base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4-ip4base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 64 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4-ip4base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 64 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip4-ip4base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip4-ip4base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip4-ip4base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip4-ip4base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4-ip4base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 64 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4-ip4base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 64 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip4-ip4base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip4-ip4base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip4-ip4base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip4-ip4base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 10.10.10.0 | 24 | 1 | local
+| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip4 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip4 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-iacldstbase-ndrchk.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-iacldstbase-ndrchk.robot
new file mode 100644
index 0000000000..3a070f8eaf
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-iacldstbase-ndrchk.robot
@@ -0,0 +1,332 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.Classify.Classify
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT1']}
+| ... | WITH NAME | dut1_v4
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT2']}
+| ... | WITH NAME | dut2_v4
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | FEATURE | IACLDST
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput IPv4 whitelist verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing, two static IPv4 /24 routes and IPv4 iAcl security whitelist
+| ... | ingress /24 filter entries applied on links TG - DUT1 and DUT2 - TG.
+| ... | DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 64 Byte frames using single trial throughput test
+| | ... | at 2x 3.6mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 3.6mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip4-ip4base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 1518 Byte frames using single trial throughput test
+| | ... | at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip4-ip4base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 9000 Byte frames using single trial throughput test
+| | ... | at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-64B-2t2c-ethip4-ip4base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 64 Byte frames using single trial throughput test
+| | ... | at 2x 6.3mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 6.3mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip4-ip4base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 1518 Byte frames using single trial throughput test
+| | ... | at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip4-ip4base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 9000 Byte frames using single trial throughput test
+| | ... | at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-64B-4t4c-ethip4-ip4base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 thread, 4 phy core, 2 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 64 Byte frames using single trial throughput test
+| | ... | at 2x 9.7mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 9.7mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip4-ip4base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 thread, 4 phy core, 2 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 1518 Byte frames using single trial throughput test
+| | ... | at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip4-ip4base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 thread, 4 phy core, 2 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 9000 Byte frames using single trial throughput test
+| | ... | at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-iacldstbase-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-iacldstbase-ndrpdrdisc.robot
new file mode 100644
index 0000000000..5d63633383
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-iacldstbase-ndrpdrdisc.robot
@@ -0,0 +1,713 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.Classify.Classify
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT1']}
+| ... | WITH NAME | dut1_v4
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT2']}
+| ... | WITH NAME | dut2_v4
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | FEATURE | IACLDST
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv4 iAcl whitelist test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing, two static IPv4 /24 routes and IPv4 iAcl security whitelist
+| ... | ingress /24 filter entries applied on links TG - DUT1 and DUT2 - TG.
+| ... | DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | static payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 64 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4-ip4base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 64 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4-ip4base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4-ip4base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip4-ip4base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip4-ip4base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4-ip4base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 64 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4-ip4base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 64 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip4-ip4base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip4-ip4base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip4-ip4base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip4-ip4base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4-ip4base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 64 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4-ip4base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 64 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip4-ip4base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip4-ip4base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip4-ip4base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip4-ip4base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut1} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut1} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 20.20.20.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut1} | ${dut1_if1} | ip4 | ${table_idx}
+| | ${table_idx} | ${skip_n} | ${match_n}= | And Vpp Creates Classify Table L3
+| | ... | ${dut2} | ip4 | dst
+| | And Vpp Configures Classify Session L3
+| | ... | ${dut2} | permit | ${table_idx} | ${skip_n} | ${match_n}
+| | ... | ip4 | dst | 10.10.10.2
+| | And Vpp Enable Input Acl Interface
+| | ... | ${dut2} | ${dut2_if2} | ip4 | ${table_idx}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ipolicemarkbase-ndrchk.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ipolicemarkbase-ndrchk.robot
new file mode 100644
index 0000000000..69c0f60cfc
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ipolicemarkbase-ndrchk.robot
@@ -0,0 +1,226 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.Policer
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT1']}
+| ... | WITH NAME | dut1_v4
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT2']}
+| ... | WITH NAME | dut2_v4
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | IP4FWD | BASE | DOT1Q
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput IPv4 policer verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 route entries and policer with 2R3C
+| ... | color-aware configuration. Policer is applied on links TG - DUT1 and
+| ... | DUT2 - TG. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544, RFC2698.
+
+*** Variables ***
+| ${cir}= | ${100}
+| ${eir}= | ${150}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4base-ipolicemarkbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 1 thread, 1 phy \
+| | ... | core, 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 \
+| | ... | Byte frames using single trial throughput test at 2x 3.6mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 3.6mpps
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip4-ip4base-ipolicemarkbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 1 thread, 1 phy \
+| | ... | core, 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 \
+| | ... | Byte frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip4-ip4base-ipolicemarkbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 1 thread, 1 phy \
+| | ... | core, 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 \
+| | ... | Byte frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-64B-2t2c-ethip4-ip4base-ipolicemarkbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 2 threads, 2 \
+| | ... | phy cores, 1 receive queue per NIC port. [Ver] Verify ref-NDR for \
+| | ... | 64 Byte frames using single trial throughput test at 2x 6.2mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 6.2mpps
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip4-ip4base-ipolicemarkbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 2 threads, 2 \
+| | ... | phy cores, 1 receive queue per NIC port. [Ver] Verify ref-NDR for \
+| | ... | 1518 Byte frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip4-ip4base-ipolicemarkbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 2 threads, 2 \
+| | ... | phy cores, 1 receive queue per NIC port. [Ver] Verify ref-NDR for \
+| | ... | 9000 Byte frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-64B-4t4c-ethip4-ip4base-ipolicemarkbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 4 thread, 4 \
+| | ... | phy core, 2 receive queue per NIC port. [Ver] Verify ref-NDR for 64 \
+| | ... | Byte frames using single trial throughput test at 2x 10.4mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 10.4mpps
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip4-ip4base-ipolicemarkbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 4 thread, 4 \
+| | ... | phy core, 2 receive queue per NIC port. [Ver] Verify ref-NDR for \
+| | ... | 1518 Byte frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip4-ip4base-ipolicemarkbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 4 thread, 4 \
+| | ... | phy core, 2 receive queue per NIC port. [Ver] Verify ref-NDR for \
+| | ... | 9000 Byte frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ipolicemarkbase-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ipolicemarkbase-ndrpdrdisc.robot
new file mode 100644
index 0000000000..2400a58d12
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ipolicemarkbase-ndrpdrdisc.robot
@@ -0,0 +1,517 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.Policer
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT1']}
+| ... | WITH NAME | dut1_v4
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT2']}
+| ... | WITH NAME | dut2_v4
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | IP4FWD | BASE | DOT1Q
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv4 policer test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 route entries and policer with 2R3C
+| ... | color-aware configuration. Policer is applied on links TG - DUT1 and
+| ... | DUT2 - TG. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | generated payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544, RFC2698.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+| ${cir}= | ${100}
+| ${eir}= | ${150}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4base-ipolicemarkbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 1 thread, 1 \
+| | ... | phy core, 1 receive queue per NIC port. [Ver] Find NDR for 64 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4-ip4base-ipolicemarkbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 1 thread, 1 \
+| | ... | phy core, 1 receive queue per NIC port. [Ver] Find PDR for 64 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4-ip4base-ipolicemarkbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 1 thread, 1 \
+| | ... | phy core, 1 receive queue per NIC port. [Ver] Find NDR for 1518 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4-ip4base-ipolicemarkbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 1 thread, 1 \
+| | ... | phy core, 1 receive queue per NIC port. [Ver] Find PDR for 1518 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip4-ip4base-ipolicemarkbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 1 thread, 1 \
+| | ... | phy core, 1 receive queue per NIC port. [Ver] Find NDR for 9000 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip4-ip4base-ipolicemarkbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 1 thread, 1 \
+| | ... | phy core, 1 receive queue per NIC port. [Ver] Find PDR for 9000 \
+| | ... | Byte frames using binary search start at 10GE linerate, step 5kpps, \
+| | ... | LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4-ip4base-ipolicemarkbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 2 threads, 2 \
+| | ... | phy cores, 1 receive queue per NIC port. [Ver] Find NDR for 64 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4-ip4base-ipolicemarkbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 2 threads, 2 \
+| | ... | phy cores, 1 receive queue per NIC port. [Ver] Find PDR for 64 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 100kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip4-ip4base-ipolicemarkbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 2 threads, 2 \
+| | ... | phy cores, 1 receive queue per NIC port. [Ver] Find NDR for 1518 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip4-ip4base-ipolicemarkbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 2 threads, 2 \
+| | ... | phy cores, 1 receive queue per NIC port. [Ver] Find PDR for 1518 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip4-ip4base-ipolicemarkbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 2 threads, 2 \
+| | ... | phy cores, 1 receive queue per NIC port. [Ver] Find NDR for 9000 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip4-ip4base-ipolicemarkbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 2 threads, 2 \
+| | ... | phy cores, 1 receive queue per NIC port. [Ver] Find PDR for 9000 \
+| | ... | Byte frames using binary search start at 10GE linerate, step 5kpps, \
+| | ... | LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4-ip4base-ipolicemarkbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 4 threads, 4 \
+| | ... | phy cores, 2 receive queue per NIC port. [Ver] Find NDR for 64 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4-ip4base-ipolicemarkbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 4 threads, 4 \
+| | ... | phy cores, 2 receive queue per NIC port. [Ver] Find PDR for 64 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip4-ip4base-ipolicemarkbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 4 threads, 4 \
+| | ... | phy cores, 2 receive queue per NIC port. [Ver] Find NDR for 1518 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip4-ip4base-ipolicemarkbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 4 threads, 4 \
+| | ... | phy cores, 2 receive queue per NIC port. [Ver] Find PDR for 1518 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip4-ip4base-ipolicemarkbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 4 threads, 4 \
+| | ... | phy cores, 2 receive queue per NIC port. [Ver] Find NDR for 9000 \
+| | ... | Byte frames using binary search start at 10GE linerate, step \
+| | ... | 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip4-ip4base-ipolicemarkbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and policer config with 4 threads, 4 \
+| | ... | phy cores, 2 receive queue per NIC port. [Ver] Find PDR for 9000 \
+| | ... | Byte frames using binary search start at 10GE linerate, step 5kpps, \
+| | ... | LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Variable | ${cb} | ${framesize}
+| | Set Test Variable | ${eb} | ${framesize}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding in 3-node circular topology
+| | And Initialize IPv4 policer 2r3c-'ca' in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ndrchk.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ndrchk.robot
new file mode 100644
index 0000000000..01dcedb4b8
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ndrchk.robot
@@ -0,0 +1,198 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.topology.Topology
+| Library | resources.libraries.python.NodePath
+| Library | resources.libraries.python.InterfaceUtil
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT1']}
+| ... | WITH NAME | dut1_v4
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT2']}
+| ... | WITH NAME | dut2_v4
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | BASE | IP4BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput IPv4 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 route entries. DUT1 and DUT2 tested with
+| ... | 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 4.6mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 4.6mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-64B-2t2c-ethip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 10.4mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 10.4mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-64B-4t4c-ethip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 10.4mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 10.4mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-9000B-4t4c-ethip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ndrpdrdisc.robot
new file mode 100644
index 0000000000..edf5ca0691
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-ndrpdrdisc.robot
@@ -0,0 +1,489 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.topology.Topology
+| Library | resources.libraries.python.NodePath
+| Library | resources.libraries.python.InterfaceUtil
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT1']}
+| ... | WITH NAME | dut1_v4
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT2']}
+| ... | WITH NAME | dut2_v4
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | BASE | IP4BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv4 routing test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 route entries. DUT1 and DUT2 tested with
+| ... | 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc19-IMIX-1t1c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame size
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc20-IMIX-1t1c-ethip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame size
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-pdrchk.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-pdrchk.robot
new file mode 100644
index 0000000000..3c49d6ba72
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-pdrchk.robot
@@ -0,0 +1,190 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.topology.Topology
+| Library | resources.libraries.python.NodePath
+| Library | resources.libraries.python.InterfaceUtil
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT1']}
+| ... | WITH NAME | dut1_v4
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT2']}
+| ... | WITH NAME | dut2_v4
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | BASE | IP4BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test
+| ...
+| Documentation | *Reference PDR throughput IPv4 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 route entries. DUT1 and DUT2 tested with
+| ... | 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-PDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-PDR value is periodically updated acording to
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Keywords ***
+| Check PDR for IPv4 routing
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with ${wt} thread(s), ${wt}\
+| | ... | phy core(s), ${rxq} receive queue(s) per NIC port.
+| | ... | [Ver] Verify ref-PDR for ${framesize} frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte\
+| | ... | frames using single trial throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${64} | rate=4.7mpps | wt=1 | rxq=1
+
+| tc02-1518B-1t1c-ethip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${1518} | rate=400000pps | wt=1 | rxq=1
+
+| tc03-9000B-1t1c-ethip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${9000} | rate=60000pps | wt=1 | rxq=1
+
+| tc04-64B-2t2c-ethip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 thread, 2 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 2T2C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${64} | rate=7.4mpps | wt=2 | rxq=1
+
+| tc05-1518B-2t2c-ethip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 thread, 2 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${1518} | rate=400000pps | wt=2 | rxq=1
+
+| tc06-9000B-2t2c-ethip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 thread, 2 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${9000} | rate=60000pps | wt=2 | rxq=1
+
+| tc07-64B-4t4c-ethip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 thread, 4 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${64} | rate=7.4mpps | wt=4 | rxq=2
+
+| tc08-1518B-4t4c-ethip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 thread, 4 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${1518} | rate=400000pps | wt=4 | rxq=2
+
+| tc08-9000B-4t4c-ethip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 thread, 4 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${9000} | rate=60000pps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-snat-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-snat-ndrpdrdisc.robot
new file mode 100644
index 0000000000..89aa2788ef
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4base-snat-ndrpdrdisc.robot
@@ -0,0 +1,227 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/ip/snat.robot
+| Resource | resources/libraries/robot/shared/traffic.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | FEATURE | SNAT | BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test with SNAT
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *SNAT performance test cases*
+| ...
+| ... | *High level description*
+| ...
+| ... | - NDR and PDR tests
+| ... | - 3-node topology, TG-DUT1-DUT2-TG, SNAT is enabled between DUTs.
+| ... | - Cores / threads: 1t1c and 2t2c
+| ... | - Framesize: 64B, 1518B, IMIX
+| ... | - Packet: ETH / IP(src, dst) / UDP(src_port, dst_port) / payload
+| ...
+| ... | *Low level description*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4-UDP for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 and IPv4/20 route entries. DUT1 and DUT2
+| ... | tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, one flow per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with UDP header and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4udp-1u1p
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4base-snat-1u-1p-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1 user and 1 port (session) per user.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4-ip4base-snat-1u-1p-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1 user and 1 port (session) per user.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4-ip4base-snat-1u-1p-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1 user and 1 port (session) per user.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4-ip4base-snat-1u-1p-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1 user and 1 port (session) per user.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4-ip4base-snat-1u-1p-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1 user and 1 port (session) per user.
+| | ... | [Ver] Find NDR for IMIX frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4-ip4base-snat-1u-1p-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1 user and 1 port (session) per user.
+| | ... | [Ver] Find PDR for IMIX frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale200k-ndrchk.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale200k-ndrchk.robot
new file mode 100644
index 0000000000..abb34c2794
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale200k-ndrchk.robot
@@ -0,0 +1,201 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | SCALE | FIB_200K
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput IPv4 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and 2x100k static IPv4 /32 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT confiiguration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 100k flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. Incrementing of IP.dst (IPv4 destination address)
+| ... | field is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | ${100000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${rts_per_flow}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 4.3mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 4.3mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip4-ip4scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip4-ip4scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-64B-2t2c-ethip4-ip4scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 9.1mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 9.1mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip4-ip4scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip4-ip4scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-64B-4t4c-ethip4-ip4scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 10.4mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 10.4mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip4-ip4scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip4-ip4scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale200k-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale200k-ndrpdrdisc.robot
new file mode 100644
index 0000000000..6d0034338f
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale200k-ndrpdrdisc.robot
@@ -0,0 +1,457 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | SCALE | FIB_200K
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv4 routing test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and 2x100k static IPv4 /32 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 100k flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | static payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+| ${rts_per_flow}= | 100000
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${rts_per_flow}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4-ip4scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4-ip4scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4-ip4scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip4-ip4scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip4-ip4scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4-ip4scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4-ip4scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip4-ip4scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip4-ip4scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip4-ip4scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip4-ip4scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4-ip4scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4-ip4scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip4-ip4scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip4-ip4scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip4-ip4scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip4-ip4scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale20k-ndrchk.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale20k-ndrchk.robot
new file mode 100644
index 0000000000..8303a4e139
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale20k-ndrchk.robot
@@ -0,0 +1,210 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | SCALE | FIB_20K
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput IPv4 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and 2x10k static IPv4 /32 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT confiiguration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 10k flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. Incrementing of IP.dst (IPv4 destination address)
+| ... | field is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | ${10000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${rts_per_flow}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 4.3mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 4.3mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip4-ip4scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip4-ip4scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc04-64B-2t2c-ethip4-ip4scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 9.0mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 9.0mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip4-ip4scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip4-ip4scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc07-64B-4t4c-ethip4-ip4scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 10.4mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 10.4mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip4-ip4scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip4-ip4scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale20k-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale20k-ndrpdrdisc.robot
new file mode 100644
index 0000000000..a4c5b31bd7
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale20k-ndrpdrdisc.robot
@@ -0,0 +1,457 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | SCALE | FIB_20K
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv4 routing test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and 2x10k static IPv4 /32 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 10k flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | static payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+| ${rts_per_flow}= | ${10000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${rts_per_flow}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4-ip4scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4-ip4scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4-ip4scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip4-ip4scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip4-ip4scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4-ip4scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4-ip4scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip4-ip4scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip4-ip4scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip4-ip4scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip4-ip4scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4-ip4scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4-ip4scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip4-ip4scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip4-ip4scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip4-ip4scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip4-ip4scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale2m-ndrchk.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale2m-ndrchk.robot
new file mode 100644
index 0000000000..84fc806d23
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale2m-ndrchk.robot
@@ -0,0 +1,210 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | SCALE | FIB_2M
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput IPv4 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and 2x1M static IPv4 /32 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT confiiguration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 1M flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. Incrementing of IP.dst (IPv4 destination address)
+| ... | field is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | ${1000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${rts_per_flow}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 3.7mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 3.7mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip4-ip4scale2m-ndrchk
+| | [Documentation]
+| | ... | Verify ref-NDR for 1518 Byte frames using single trial throughput
+| | ... | test. DUT runs IPv4 routing config with 1 thread, 1 phy core, 1
+| | ... | receive queue per NIC port at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip4-ip4scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc04-64B-2t2c-ethip4-ip4scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 7.6mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 7.6mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip4-ip4scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip4-ip4scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc07-64B-4t4c-ethip4-ip4scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 10.4mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 10.4mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip4-ip4scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip4-ip4scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale2m-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale2m-ndrpdrdisc.robot
new file mode 100644
index 0000000000..920bb8f42d
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale2m-ndrpdrdisc.robot
@@ -0,0 +1,457 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | SCALE | FIB_2M
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv4 routing test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and 2x1M static IPv4 /32 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 1M flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | static payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+| ${rts_per_flow}= | ${1000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${rts_per_flow}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4-ip4scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4-ip4scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4-ip4scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip4-ip4scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip4-ip4scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4-ip4scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4-ip4scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip4-ip4scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip4-ip4scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip4-ip4scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip4-ip4scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4-ip4scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4-ip4scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip4-ip4scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip4-ip4scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip4-ip4scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip4-ip4scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale2m-pdrchk.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale2m-pdrchk.robot
new file mode 100644
index 0000000000..0a5133aff1
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4-ip4scale2m-pdrchk.robot
@@ -0,0 +1,185 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | SCALE | FIB_2M
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test
+| ...
+| Documentation | *Reference PDR throughput IPv4 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and 2x1M static IPv4 /32 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG verifies throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. Ref-PDR value is periodically updated acording
+| ... | to formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 1M flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. Incrementing of IP.dst (IPv4 destination address)
+| ... | field is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | ${1000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4dst${rts_per_flow}
+
+*** Keywords ***
+| Check PDR for IPv4 routing
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with ${wt} thread, ${wt} phy\
+| | ... | core, ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${64} | rate=3.0mpps | wt=1 | rxq=1
+
+| tc02-1518B-1t1c-ethip4-ip4scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | trial throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${1518} | rate=600000pps | wt=1 | rxq=1
+
+| tc03-9000B-1t1c-ethip4-ip4scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${9000} | rate=100000pps | wt=1 | rxq=1
+
+| tc04-64B-2t2c-ethip4-ip4scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${64} | rate=7.0mpps | wt=2 | rxq=1
+
+| tc05-1518B-2t2c-ethip4-ip4scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${1518} | rate=500000pps | wt=2 | rxq=1
+
+| tc06-9000B-2t2c-ethip4-ip4scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${9000} | rate=100000pps | wt=2 | rxq=1
+
+| tc07-64B-4t4c-ethip4-ip4scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${64} | rate=10.0mpps | wt=4 | rxq=2
+
+| tc08-1518B-4t4c-ethip4-ip4scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${1518} | rate=500000pps | wt=4 | rxq=2
+
+| tc09-9000B-4t4c-ethip4-ip4scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing
+| | framesize=${9000} | rate=100000pps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4base-udpsrcscale15-snat-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4base-udpsrcscale15-snat-ndrpdrdisc.robot
new file mode 100644
index 0000000000..4eb3862e0f
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4base-udpsrcscale15-snat-ndrpdrdisc.robot
@@ -0,0 +1,229 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/ip/snat.robot
+| Resource | resources/libraries/robot/shared/traffic.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | FEATURE | SNAT | BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test with SNAT
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *SNAT performance test cases*
+| ...
+| ... | *High level description*
+| ...
+| ... | - NDR and PDR tests
+| ... | - 3-node topology, TG-DUT1-DUT2-TG, SNAT is enabled between DUTs.
+| ... | - Cores / threads: 1t1c and 2t2c
+| ... | - Framesize: 64B, 1518B, IMIX
+| ... | - Packet: ETH / IP(src, dst) / UDP(src_port, dst_port) / payload
+| ... | - scale: src: 1 user, 10 users, 100 users, ..., 4000 up to the memory
+| ... | limit; 15 ports per user
+| ...
+| ... | *Low level description*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4-UDP for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 and IPv4/20 route entries. DUT1 and DUT2
+| ... | tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, one flow per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with UDP header and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4udp-1u15p
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4udp-ip4base-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1 user and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4udp-ip4base-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1 user and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4udp-ip4base-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1 user and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4udp-ip4base-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1 user and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4udp-ip4base-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1 user and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for IMIX frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4udp-ip4base-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1 user and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for IMIX frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale10-udpsrcscale15-snat-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale10-udpsrcscale15-snat-ndrpdrdisc.robot
new file mode 100644
index 0000000000..28b44ebfc6
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale10-udpsrcscale15-snat-ndrpdrdisc.robot
@@ -0,0 +1,230 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/ip/snat.robot
+| Resource | resources/libraries/robot/shared/traffic.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | FEATURE | SNAT | SRC_USER_10
+| ... | SCALE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test with SNAT
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *SNAT performance test cases*
+| ...
+| ... | *High level description*
+| ...
+| ... | - NDR and PDR tests
+| ... | - 3-node topology, TG-DUT1-DUT2-TG, SNAT is enabled between DUTs.
+| ... | - Cores / threads: 1t1c and 2t2c
+| ... | - Framesize: 64B, 1518B, IMIX
+| ... | - Packet: ETH / IP(src, dst) / UDP(src_port, dst_port) / payload
+| ... | - scale: src: 1 user, 10 users, 100 users, ..., 4000 up to the memory
+| ... | limit; 15 ports per user
+| ...
+| ... | *Low level description*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4-UDP for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 and IPv4/20 route entries. DUT1 and DUT2
+| ... | tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, one flow per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with UDP header and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4udp-10u15p
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4udp-ip4scale10-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 10 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4udp-ip4scale10-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 10 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4udp-ip4scale10-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 10 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4udp-ip4scale10-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 10 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4udp-ip4scale10-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 10 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for IMIX frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4udp-ip4scale10-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 10 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for IMIX frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale100-udpsrcscale15-snat-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale100-udpsrcscale15-snat-ndrpdrdisc.robot
new file mode 100644
index 0000000000..9f683679e2
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale100-udpsrcscale15-snat-ndrpdrdisc.robot
@@ -0,0 +1,230 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/ip/snat.robot
+| Resource | resources/libraries/robot/shared/traffic.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | FEATURE | SNAT | SRC_USER_100
+| ... | SCALE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test with SNAT
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *SNAT performance test cases*
+| ...
+| ... | *High level description*
+| ...
+| ... | - NDR and PDR tests
+| ... | - 3-node topology, TG-DUT1-DUT2-TG, SNAT is enabled between DUTs.
+| ... | - Cores / threads: 1t1c and 2t2c
+| ... | - Framesize: 64B, 1518B, IMIX
+| ... | - Packet: ETH / IP(src, dst) / UDP(src_port, dst_port) / payload
+| ... | - scale: src: 1 user, 10 users, 100 users, ..., 4000 up to the memory
+| ... | limit; 15 ports per user
+| ...
+| ... | *Low level description*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4-UDP for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 and IPv4/20 route entries. DUT1 and DUT2
+| ... | tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, one flow per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with UDP header and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4udp-100u15p
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4udp-ip4scale100-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 100 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4udp-ip4scale100-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 100 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4udp-ip4scale100-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 100 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4udp-ip4scale100-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 100 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4udp-ip4scale100-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 100 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for IMIX frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4udp-ip4scale100-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 100 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for IMIX frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale1000-udpsrcscale15-snat-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale1000-udpsrcscale15-snat-ndrpdrdisc.robot
new file mode 100644
index 0000000000..73ede552b2
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale1000-udpsrcscale15-snat-ndrpdrdisc.robot
@@ -0,0 +1,230 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/ip/snat.robot
+| Resource | resources/libraries/robot/shared/traffic.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | FEATURE | SNAT | SRC_USER_1000
+| ... | SCALE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test with SNAT
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *SNAT performance test cases*
+| ...
+| ... | *High level description*
+| ...
+| ... | - NDR and PDR tests
+| ... | - 3-node topology, TG-DUT1-DUT2-TG, SNAT is enabled between DUTs.
+| ... | - Cores / threads: 1t1c and 2t2c
+| ... | - Framesize: 64B, 1518B, IMIX
+| ... | - Packet: ETH / IP(src, dst) / UDP(src_port, dst_port) / payload
+| ... | - scale: src: 1 user, 10 users, 100 users, ..., 4000 up to the memory
+| ... | limit; 15 ports per user
+| ...
+| ... | *Low level description*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4-UDP for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 and IPv4/20 route entries. DUT1 and DUT2
+| ... | tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, one flow per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with UDP header and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4udp-1000u15p
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4udp-ip4scale1000-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1000 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4udp-ip4scale1000-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1000 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4udp-ip4scale1000-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1000 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4udp-ip4scale1000-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1000 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4udp-ip4scale1000-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1000 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for IMIX frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4udp-ip4scale1000-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 1000 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for IMIX frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale2000-udpsrcscale15-snat-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale2000-udpsrcscale15-snat-ndrpdrdisc.robot
new file mode 100644
index 0000000000..bce900f0ec
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale2000-udpsrcscale15-snat-ndrpdrdisc.robot
@@ -0,0 +1,124 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/ip/snat.robot
+| Resource | resources/libraries/robot/shared/traffic.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | FEATURE | SNAT | SRC_USER_2000
+| ... | SCALE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test with SNAT
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *SNAT performance test cases*
+| ...
+| ... | *High level description*
+| ...
+| ... | - NDR and PDR tests
+| ... | - 3-node topology, TG-DUT1-DUT2-TG, SNAT is enabled between DUTs.
+| ... | - Cores / threads: 1t1c and 2t2c
+| ... | - Framesize: 64B, 1518B, IMIX
+| ... | - Packet: ETH / IP(src, dst) / UDP(src_port, dst_port) / payload
+| ... | - scale: src: 1 user, 10 users, 100 users, ..., 4000 up to the memory
+| ... | limit; 15 ports per user
+| ...
+| ... | *Low level description*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4-UDP for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 and IPv4/20 route entries. DUT1 and DUT2
+| ... | tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, one flow per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with UDP header and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4udp-2000u15p
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4udp-ip4scale2000-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 2000 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4udp-ip4scale2000-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 2000 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale4000-udpsrcscale15-snat-ndrpdrdisc.robot b/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale4000-udpsrcscale15-snat-ndrpdrdisc.robot
new file mode 100644
index 0000000000..2b91618470
--- /dev/null
+++ b/tests/vpp/perf/ip4/10ge2p1x520-ethip4udp-ip4scale4000-udpsrcscale15-snat-ndrpdrdisc.robot
@@ -0,0 +1,124 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/ip/snat.robot
+| Resource | resources/libraries/robot/shared/traffic.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | FEATURE | SNAT | SRC_USER_4000
+| ... | SCALE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test with SNAT
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *SNAT performance test cases*
+| ...
+| ... | *High level description*
+| ...
+| ... | - NDR and PDR tests
+| ... | - 3-node topology, TG-DUT1-DUT2-TG, SNAT is enabled between DUTs.
+| ... | - Cores / threads: 1t1c and 2t2c
+| ... | - Framesize: 64B, 1518B, IMIX
+| ... | - Packet: ETH / IP(src, dst) / UDP(src_port, dst_port) / payload
+| ... | - scale: src: 1 user, 10 users, 100 users, ..., 4000 up to the memory
+| ... | limit; 15 ports per user
+| ...
+| ... | *Low level description*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4-UDP for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 and IPv4/20 route entries. DUT1 and DUT2
+| ... | tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, one flow per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with UDP header and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4udp-4000u15p
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4udp-ip4scale4000-udpsrcscale15-snat-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 4000 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4udp-ip4scale4000-udpsrcscale15-snat-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. SNAT is configured between DUTs -\
+| | ... | 4000 users and 15 ports (sessions) per user.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Add SNAT to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize SNAT in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4/40ge2p1xl710-ethip4-ip4base-ndrpdrdisc.robot b/tests/vpp/perf/ip4/40ge2p1xl710-ethip4-ip4base-ndrpdrdisc.robot
new file mode 100644
index 0000000000..1dbda15832
--- /dev/null
+++ b/tests/vpp/perf/ip4/40ge2p1xl710-ethip4-ip4base-ndrpdrdisc.robot
@@ -0,0 +1,257 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.topology.Topology
+| Library | resources.libraries.python.NodePath
+| Library | resources.libraries.python.InterfaceUtil
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT1']}
+| ... | WITH NAME | dut1_v4
+| Library | resources.libraries.python.IPv4Setup.Dut | ${nodes['DUT2']}
+| ... | WITH NAME | dut2_v4
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-XL710 | ETH | IP4FWD | BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-XL710
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv4 routing test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 route entries. DUT1 and DUT2 tested with
+| ... | 2p40GE NIC XL710 by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 18.75Mpps rate, step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc03-1518B-1t1c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 24.5G rate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc07-64B-2t2c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 18.75Mpps rate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc09-1518B-2t2c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 24.5G rate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc13-64B-4t4c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 18.75Mpps rate, step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc15-1518B-4t4c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 24.5G rate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc19-IMIX-1t1c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame size
+| | ... | using binary search start at 24.5G rate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc20-IMIX-2t2c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 thread, 2 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame size
+| | ... | using binary search start at 24.5G rate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc21-IMIX-4t4c-ethip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 thread, 4 phy core, \
+| | ... | 2 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame size
+| | ... | using binary search start at 24.5G rate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv4 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
diff --git a/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip4-ip4base-ndrchk.robot b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip4-ip4base-ndrchk.robot
new file mode 100644
index 0000000000..79f0bbc772
--- /dev/null
+++ b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip4-ip4base-ndrchk.robot
@@ -0,0 +1,266 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/overlay/lisp_static_adjacency.robot
+| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | IP4FWD | ENCAP | LISP | IP4UNRLAY | IP4OVRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput Lisp tunnel verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology\
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4-LISP-IPv4 on DUT1-DUT2\
+| ... | Eth-IPv4 on TG-DUTn for IPv4 routing over LISPoIPv4 tunnel.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4\
+| ... | routing and static routes. LISPoIPv4 tunnel is configured between\
+| ... | DUT1 and DUT2. DUT1 and DUT2 tested with 2p10GE NIC X520\
+| ... | Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies\
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet\
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to\
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long\
+| ... | performance tests for the same DUT confiiguration. Test packets are\
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3\
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with\
+| ... | all packets containing Ethernet header, IPv4 header with\
+| ... | IP protocol=61 and generated payload.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4lispip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial throughput\
+| | ... | test at 2x 1.9mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 1.9mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1480B-1t1c-ethip4lispip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1480 Byte frames using single trial\
+| | ... | throughput test at 2x 740000pps.
+| | [Tags] | 1480B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1480}
+| | ${rate}= | Set Variable | 740000pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip4lispip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-64B-2t2c-ethip4lispip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x 3.6mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 3.6mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1480B-2t2c-ethip4lispip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1480 Byte frames using single trial\
+| | ... | throughput test at 2x 740000pps.
+| | [Tags] | 1480B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1480}
+| | ${rate}= | Set Variable | 740000pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip4lispip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-64B-4t4c-ethip4lispip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x 3.5mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 3.5mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1480B-4t4c-ethip4lispip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1480 Byte frames using single trial\
+| | ... | throughput test at 2x 740000pps.
+| | [Tags] | 1480B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1480}
+| | ${rate}= | Set Variable | 740000pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip4lispip4-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip4-ip4base-ndrpdrdisc.robot b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip4-ip4base-ndrpdrdisc.robot
new file mode 100644
index 0000000000..2dc5be4981
--- /dev/null
+++ b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip4-ip4base-ndrpdrdisc.robot
@@ -0,0 +1,595 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/overlay/lisp_static_adjacency.robot
+| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | IP4FWD | ENCAP | LISP | IP4UNRLAY | IP4OVRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC6830: Pkt throughput Lisp test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology\
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4-LISP-IPv4 on DUT1-DUT2,\
+| ... | Eth-IPv4 on TG-DUTn for IPv4 routing over LISPoIPv4 tunnel.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4\
+| ... | routing and static routes. LISPoIPv4 tunnel is configured between DUT1\
+| ... | and DUT2. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop\
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\
+| ... | of packets transmitted. NDR and PDR are discovered for different\
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | *[Ref] Applicable standard specifications:* RFC6830.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4lispip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4lispip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1480B-1t1c-ethip4lispip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 1480 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps.
+| | [Tags] | 1480B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1480}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1480B-1t1c-ethip4lispip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 1480 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1480B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1480}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip4lispip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip4lispip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4lispip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4lispip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1480B-2t2c-ethip4lispip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 1480 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps.
+| | [Tags] | 1480B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1480}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1480B-2t2c-ethip4lispip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 1480 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1480B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1480}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip4lispip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip4lispip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4lispip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4lispip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1480B-4t4c-ethip4lispip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 1480 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps.
+| | [Tags] | 1480B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1480}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1480B-4t4c-ethip4lispip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 1480 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1480B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1480}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip4lispip4-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip4lispip4-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip4-ip4base-pdrchk.robot b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip4-ip4base-pdrchk.robot
new file mode 100644
index 0000000000..bebbfbc938
--- /dev/null
+++ b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip4-ip4base-pdrchk.robot
@@ -0,0 +1,193 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/overlay/lisp_static_adjacency.robot
+| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | IP4FWD | ENCAP | LISP | IP4UNRLAY | IP4OVRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test
+| ...
+| Documentation | *Reference PDR throughput Lisp tunnel verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology\
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4-LISP-IPv4 on DUT1-DUT2\
+| ... | Eth-IPv4 on TG-DUTn for IPv4 routing over LISPoIPv4 tunnel.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4\
+| ... | routing and static routes. LISPoIPv4 tunnel is configured between\
+| ... | DUT1 and DUT2. DUT1 and DUT2 tested with 2p10GE NIC X520\
+| ... | Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies\
+| ... | DUTs' throughput at ref-PDR (reference Non Drop Rate) with zero packet\
+| ... | loss tolerance. Ref-PDR value is periodically updated acording to\
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long\
+| ... | performance tests for the same DUT confiiguration. Test packets are\
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3\
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with\
+| ... | all packets containing Ethernet header, IPv4 header with\
+| ... | IP protocol=61 and generated payload.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Keywords ***
+| Check PDR for Lisp IPv4 over IPv4 forwarding
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with ${wt} thread, ${wt} phy core,\
+| | ... | ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4} | ${dut1_to_tg_ip4} | ${dut2_to_dut1_ip4}
+| | ... | ${dut2_to_tg_ip4} | ${prefix4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4_eid} | ${dut2_ip4_eid}
+| | ... | ${dut1_ip4_static_adjacency} | ${dut2_ip4_static_adjacency}
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4lispip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv4 forwarding
+| | framesize=${64} | rate=1.9mpps | wt=1 | rxq=1
+
+| tc02-1480B-1t1c-ethip4lispip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1480 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1480B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv4 forwarding
+| | framesize=${1480} | rate=370000pps | wt=1 | rxq=1
+
+| tc03-9000B-1t1c-ethip4lispip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv4 forwarding
+| | framesize=${9000} | rate=60000pps | wt=1 | rxq=1
+
+| tc04-64B-2t2c-ethip4lispip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv4 forwarding
+| | framesize=${64} | rate=3.6mpps | wt=2 | rxq=1
+
+| tc05-1480B-2t2c-ethip4lispip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1480 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1480B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv4 forwarding
+| | framesize=${1480} | rate=370000pps | wt=2 | rxq=1
+
+| tc06-9000B-2t2c-ethip4lispip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv4 forwarding
+| | framesize=${9000} | rate=60000pps | wt=2 | rxq=1
+
+| tc07-64B-4t4c-ethip4lispip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv4 forwarding
+| | framesize=${64} | rate=3.5mpps | wt=4 | rxq=2
+
+| tc08-1480B-4t4c-ethip4lispip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1480 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1480B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv4 forwarding
+| | framesize=${1480} | rate=370000pps | wt=4 | rxq=2
+
+| tc09-9000B-4t4c-ethip4lispip4-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv4 forwarding
+| | framesize=${9000} | rate=60000pps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip6-ip4base-ndrchk.robot b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip6-ip4base-ndrchk.robot
new file mode 100644
index 0000000000..cdcc200b55
--- /dev/null
+++ b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip6-ip4base-ndrchk.robot
@@ -0,0 +1,266 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/overlay/lisp_static_adjacency.robot
+| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | IP4FWD | ENCAP | LISP | IP6UNRLAY | IP4OVRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput Lisp tunnel verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4-LISP-IPv6 on DUT1-DUT2,\
+| ... | Eth-IPv4 on TG-DUTn for IPv4 routing over LISPoIPv6 tunnel.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4\
+| ... | routing and static routes. LISPoIPv6 tunnel is configured between\
+| ... | DUT1 and DUT2. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic\
+| ... | by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies\
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet\
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to\
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT confiiguration. Test packets are\
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3\
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with\
+| ... | all packets containing Ethernet header, IPv4 header or IPv6 header with\
+| ... | IP protocol=61 and generated payload.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4lispip6-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x 1.53mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 1.53mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1460B-1t1c-ethip4lispip6-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x 720000pps.
+| | [Tags] | 1460B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1460}
+| | ${rate}= | Set Variable | 720000pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip4lispip6-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-64B-2t2c-ethip4lispip6-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x 3.25mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 3.25mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1460B-2t2c-ethip4lispip6-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x 720000pps.
+| | [Tags] | 1460B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1460}
+| | ${rate}= | Set Variable | 720000pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip4lispip6-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-64B-4t4c-ethip4lispip6-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x 3.12mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 3.12mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1460B-4t4c-ethip4lispip6-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x 720000pps.
+| | [Tags] | 1460B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1460}
+| | ${rate}= | Set Variable | 720000pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip4lispip6-ip4base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip6-ip4base-ndrpdrdisc.robot b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip6-ip4base-ndrpdrdisc.robot
new file mode 100644
index 0000000000..a7f730ac40
--- /dev/null
+++ b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip6-ip4base-ndrpdrdisc.robot
@@ -0,0 +1,596 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/overlay/lisp_static_adjacency.robot
+| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | IP4FWD | ENCAP | LISP | IP6UNRLAY | IP4OVRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC6830: Pkt throughput Lisp test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology\
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4-LISP-IPv6 on DUT1-DUT2 and\
+| ... | Eth-IPv4 on TG-DUTn for IPv4 routing over LISPoIPv6 tunnel.\
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4\
+| ... | routing and static routes. LISPoIPv6 tunnel is configured\
+| ... | between DUT1 and DUT2. DUT1 and DUT2 tested with 2p10GE NIC X520\
+| ... | Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop\
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\
+| ... | of packets transmitted. NDR and PDR are discovered for different\
+| ... | Ethernet L2 frame sizes using either binary search or linear search\
+| ... | *[Ref] Applicable standard specifications:* RFC6830.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4lispip6-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4lispip6-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1460B-1t1c-ethip4lispip6-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 1460B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1460B-1t1c-ethip4lispip6-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1460B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip4lispip6-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip4lispip6-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4lispip6-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4lispip6-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1460B-2t2c-ethip4lispip6-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps.
+| | [Tags] | 1460B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1460B-2t2c-ethip4lispip6-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1460B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip4lispip6-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip4lispip6-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4lispip6-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4lispip6-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1460B-4t4c-ethip4lispip6-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps.
+| | [Tags] | 1460B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1460B-4t4c-ethip4lispip6-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1460B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip4lispip6-ip4base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip4lispip6-ip4base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip6-ip4base-pdrchk.robot b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip6-ip4base-pdrchk.robot
new file mode 100644
index 0000000000..f8190674a7
--- /dev/null
+++ b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4lispip6-ip4base-pdrchk.robot
@@ -0,0 +1,193 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/overlay/lisp_static_adjacency.robot
+| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | IP4FWD | ENCAP | LISP | IP6UNRLAY | IP4OVRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test
+| ...
+| Documentation | *Reference PDR throughput Lisp tunnel verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4-LISP-IPv6 on DUT1-DUT2,\
+| ... | Eth-IPv4 on TG-DUTn for IPv4 routing over LISPoIPv6 tunnel.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4\
+| ... | routing and static routes. LISPoIPv6 tunnel is configured between\
+| ... | DUT1 and DUT2. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic\
+| ... | by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies\
+| ... | DUTs' throughput at ref-PDR (reference Non Drop Rate) with zero packet\
+| ... | loss tolerance. Ref-PDR value is periodically updated acording to\
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long
+| ... | performance tests for the same DUT confiiguration. Test packets are\
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3\
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with\
+| ... | all packets containing Ethernet header, IPv4 header or IPv6 header with\
+| ... | IP protocol=61 and generated payload.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Keywords ***
+| Check PDR for Lisp IPv4 over IPv6 forwarding
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with ${wt} thread, ${wt} phy core,\
+| | ... | ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv4 over IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip4o6} | ${dut1_to_tg_ip4o6} | ${dut2_to_dut1_ip4o6}
+| | ... | ${dut2_to_tg_ip4o6} | ${tg_prefix4o6} | ${dut_prefix4o6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip4o6_eid} | ${dut2_ip4o6_eid}
+| | ... | ${dut1_ip4o6_static_adjacency} | ${dut2_ip4o6_static_adjacency}
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4lispip6-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv6 forwarding
+| | framesize=${64} | rate=1.53mpps | wt=1 | rxq=1
+
+| tc02-1460B-1t1c-ethip4lispip6-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1460B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv6 forwarding
+| | framesize=${1460} | rate=360000pps | wt=1 | rxq=1
+
+| tc03-9000B-1t1c-ethip4lispip6-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv6 forwarding
+| | framesize=${9000} | rate=60000pps | wt=1 | rxq=1
+
+| tc04-64B-2t2c-ethip4lispip6-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv6 forwarding
+| | framesize=${64} | rate=3.25mpps | wt=2 | rxq=1
+
+| tc05-1460B-2t2c-ethip4lispip6-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1460B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv6 forwarding
+| | framesize=${1460} | rate=360000pps | wt=2 | rxq=1
+
+| tc06-9000B-2t2c-ethip4lispip6-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv6 forwarding
+| | framesize=${9000} | rate=60000pps | wt=2 | rxq=1
+
+| tc07-64B-4t4c-ethip4lispip6-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv6 forwarding
+| | framesize=${64} | rate=3.12mpps | wt=4 | rxq=2
+
+| tc08-1460B-4t4c-ethip4lispip6-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1460B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv6 forwarding
+| | framesize=${1460} | rate=360000pps | wt=4 | rxq=2
+
+| tc09-9000B-4t4c-ethip4lispip6-ip4base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv4 over IPv6 forwarding
+| | framesize=${9000} | rate=60000pps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-ndrpdrdisc.robot b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-ndrpdrdisc.robot
new file mode 100644
index 0000000000..f41f619375
--- /dev/null
+++ b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-ndrpdrdisc.robot
@@ -0,0 +1,487 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | L2BDMACLRN | ENCAP | VXLAN | L2OVRLAY | IP4UNRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Packet throughput L2BD with VXLANoIPv4 test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 bridge domain on TG-DUTn.
+| ... | Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUT1 and DUT2.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge
+| ... | domain. VXLAN tunnels are configured between L2BDs on DUT1 and DUT2.
+| ... | DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | generated payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544, RFC7348.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+| ${vxlan_overhead} | ${50}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4vxlan-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4vxlan-l2bdbasemaclrn-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4vxlan-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4vxlan-l2bdbasemaclrn-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip4vxlan-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 5kpps.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip4vxlan-l2bdbasemaclrn-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 5kpps, LT=0.5%.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4vxlan-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4vxlan-l2bdbasemaclrn-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip4vxlan-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip4vxlan-l2bdbasemaclrn-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip4vxlan-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 5kpps.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip4vxlan-l2bdbasemaclrn-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 5kpps, LT=0.5%.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4vxlan-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4vxlan-l2bdbasemaclrn-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip4vxlan-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip4vxlan-l2bdbasemaclrn-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip4vxlan-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 5kpps.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip4vxlan-l2bdbasemaclrn-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD forwarding config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 5kpps, LT=0.5%.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-pdrchk.robot b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-pdrchk.robot
new file mode 100644
index 0000000000..dee56213c7
--- /dev/null
+++ b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-pdrchk.robot
@@ -0,0 +1,186 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | L2BDMACLRN | ENCAP | VXLAN | L2OVRLAY | IP4UNRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance pdrchk test
+| ...
+| Documentation | *RFC2544: Packet throughput L2BD with VXLANoIPv4 test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 bridge domain on TG-DUTn.
+| ... | Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUT1 and DUT2.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge
+| ... | domain. VXLAN tunnels are configured between L2BDs on DUT1 and DUT2.
+| ... | DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG verifies throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. Ref-PDR value is periodically updated acording
+| ... | to formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and generated payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544, RFC7348.
+
+*** Variables ***
+| ${vxlan_overhead} | ${50}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Keywords ***
+| Check PDR for L2BD forwarding with VXLAN over IPv4
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config config with ${wt} thread, ${wt} phy\
+| | ... | core, ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4vxlan-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2BD forwarding with VXLAN over IPv4
+| | framesize=${64 + ${vxlan_overhead}} | rate=2.4mpps | wt=1 | rxq=1
+
+| tc02-1518B-1t1c-ethip4vxlan-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2BD forwarding with VXLAN over IPv4
+| | framesize=${1518 + ${vxlan_overhead}} | rate=700000pps | wt=1 | rxq=1
+
+| tc03-9000B-1t1c-ethip4vxlan-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2BD forwarding with VXLAN over IPv4
+| | framesize=${9000 + ${vxlan_overhead}} | rate=100000pps | wt=1 | rxq=1
+
+| tc04-64B-2t2c-ethip4vxlan-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2BD forwarding with VXLAN over IPv4
+| | framesize=${64 + ${vxlan_overhead}} | rate=5.2mpps | wt=2 | rxq=1
+
+| tc05-1518B-2t2c-ethip4vxlan-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2BD forwarding with VXLAN over IPv4
+| | framesize=${1518 + ${vxlan_overhead}} | rate=650000pps | wt=2 | rxq=1
+
+| tc06-9000B-2t2c-ethip4vxlan-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2BD forwarding with VXLAN over IPv4
+| | framesize=${9000 + ${vxlan_overhead}} | rate=100000pps | wt=2 | rxq=1
+
+| tc07-64B-4t4c-ethip4vxlan-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2BD forwarding with VXLAN over IPv4
+| | framesize=${64 + ${vxlan_overhead}} | rate=6.5mpps | wt=4 | rxq=2
+
+| tc08-1518B-4t4c-ethip4vxlan-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2BD forwarding with VXLAN over IPv4
+| | framesize=${1518 + ${vxlan_overhead}} | rate=700000pps | wt=4 | rxq=2
+
+| tc09-9000B-4t4c-ethip4vxlan-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2BD forwarding with VXLAN over IPv4
+| | framesize=${9000 + ${vxlan_overhead}} | rate=100000pps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2xcbase-ndrchk.robot b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2xcbase-ndrchk.robot
new file mode 100644
index 0000000000..1a07df20a3
--- /dev/null
+++ b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2xcbase-ndrchk.robot
@@ -0,0 +1,190 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | L2XCFWD | ENCAP | VXLAN | L2OVRLAY | IP4UNRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput L2XC with VXLANoIPv4 verify test\
+| ... | cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 cross-connect on TG-DUTn.
+| ... | Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUT1 and DUT2.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. VXLAN tunnels are configured between L2XCs on DUT1 and DUT2.
+| ... | DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and generated payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544, RFC7348.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4vxlan-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 2.4mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 2.4mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip4vxlan-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 787153pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 787153pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip4vxlan-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 137816pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 137816pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-64B-2t2c-ethip4vxlan-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 5.2mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 5.2mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip4vxlan-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 787153pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 787153pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip4vxlan-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 137816pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 137816pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-64B-4t4c-ethip4vxlan-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 8.4mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 8.4mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip4vxlan-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 787153pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 787153pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip4vxlan-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 137816pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 137816pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2xcbase-ndrpdrdisc.robot b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2xcbase-ndrpdrdisc.robot
new file mode 100644
index 0000000000..8a4d7422b6
--- /dev/null
+++ b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2xcbase-ndrpdrdisc.robot
@@ -0,0 +1,432 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | L2XCFWD | ENCAP | VXLAN | L2OVRLAY | IP4UNRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2XC with VXLANoIPv4 test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 cross-connect on TG-DUTn.
+| ... | Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUT1 and DUT2.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. VXLAN tunnels are configured between L2XCs on DUT1 and DUT2.
+| ... | DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | generated payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544, RFC7348.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4vxlan-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4vxlan-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4vxlan-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4vxlan-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip4vxlan-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip4vxlan-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4vxlan-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4vxlan-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip4vxlan-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip4vxlan-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip4vxlan-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip4vxlan-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4vxlan-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4vxlan-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip4vxlan-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip4vxlan-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip4vxlan-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip4vxlan-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 50}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2xcbase-pdrchk.robot b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2xcbase-pdrchk.robot
new file mode 100644
index 0000000000..cfd119b300
--- /dev/null
+++ b/tests/vpp/perf/ip4_tunnels/10ge2p1x520-ethip4vxlan-l2xcbase-pdrchk.robot
@@ -0,0 +1,186 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | L2XCFWD | ENCAP | VXLAN | L2OVRLAY | IP4UNRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test
+| ...
+| Documentation | *Reference PDR throughput L2XC with VXLANoIPv4 verify test\
+| ... | cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 cross-connect on TG-DUTn.
+| ... | Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUT1 and DUT2.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. VXLAN tunnels are configured between L2XCs on DUT1 and DUT2.
+| ... | DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-PDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-PDR value is periodically updated acording to
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 254 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and generated payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544, RFC7348.
+
+*** Variables ***
+| ${vxlan_overhead} | ${50}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Keywords ***
+| Check PDR for L2 xconnect with VXLAN over IPv4
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with ${wt} thread, ${wt} phy core,\
+| | ... | ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4vxlan-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect with VXLAN over IPv4
+| | framesize=${64 + ${vxlan_overhead}} | rate=2.0mpps | wt=1 | rxq=1
+
+| tc02-1518B-1t1c-ethip4vxlan-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect with VXLAN over IPv4
+| | framesize=${1518 + ${vxlan_overhead}} | rate=500000pps | wt=1 | rxq=1
+
+| tc03-9000B-1t1c-ethip4vxlan-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect with VXLAN over IPv4
+| | framesize=${9000 + ${vxlan_overhead}} | rate=100000pps | wt=1 | rxq=1
+
+| tc04-64B-2t2c-ethip4vxlan-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect with VXLAN over IPv4
+| | framesize=${64 + ${vxlan_overhead}} | rate=5.0mpps | wt=2 | rxq=1
+
+| tc05-1518B-2t2c-ethip4vxlan-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect with VXLAN over IPv4
+| | framesize=${1518 + ${vxlan_overhead}} | rate=500000pps | wt=2 | rxq=1
+
+| tc06-9000B-2t2c-ethip4vxlan-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect with VXLAN over IPv4
+| | framesize=${9000 + ${vxlan_overhead}} | rate=100000pps | wt=2 | rxq=1
+
+| tc07-64B-4t4c-ethip4vxlan-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect with VXLAN over IPv4
+| | framesize=${64 + ${vxlan_overhead}} | rate=6.5mpps | wt=4 | rxq=2
+
+| tc08-1518B-4t4c-ethip4vxlan-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect with VXLAN over IPv4
+| | framesize=${1518 + ${vxlan_overhead}} | rate=500000pps | wt=4 | rxq=2
+
+| tc09-9000B-4t4c-ethip4vxlan-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect with VXLAN over IPv4
+| | framesize=${9000 + ${vxlan_overhead}} | rate=100000pps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-copwhtlistbase-ndrchk.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-copwhtlistbase-ndrchk.robot
new file mode 100644
index 0000000000..c619b6530b
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-copwhtlistbase-ndrchk.robot
@@ -0,0 +1,256 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.Cop
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | FEATURE | COPWHLIST
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput IPv6 whitelist verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing, two static IPv6 /64 routes and IPv6 COP security whitelist
+| ... | ingress /24 filter entries applied on links TG - DUT1 and DUT2 - TG.
+| ... | DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv6 header and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 78 Byte frames using single trial throughput test
+| | ... | at 2x 3.1mpps.
+| | [Tags] | 78B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 3.1mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip6-ip6base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 1518 Byte frames using single trial throughput test
+| | ... | at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip6-ip6base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 9000 Byte frames using single trial throughput test
+| | ... | at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-78B-2t2c-ethip6-ip6base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 78 Byte frames using single trial throughput test
+| | ... | at 2x 5.9mpps.
+| | [Tags] | 78B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 5.9mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip6-ip6base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 1518 Byte frames using single trial throughput test
+| | ... | at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip6-ip6base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 9000 Byte frames using single trial throughput test
+| | ... | at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-78B-4t4c-ethip6-ip6base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 78 Byte frames using single trial throughput test
+| | ... | at 2x 9.4mpps.
+| | [Tags] | 78B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 9.4mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip6-ip6base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 1518 Byte frames using single trial throughput test
+| | ... | at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip6-ip6base-copwhtlistbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 9000 Byte frames using single trial throughput test
+| | ... | at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-copwhtlistbase-ndrpdrdisc.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-copwhtlistbase-ndrpdrdisc.robot
new file mode 100644
index 0000000000..6808603262
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-copwhtlistbase-ndrpdrdisc.robot
@@ -0,0 +1,565 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.Cop
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | FEATURE | COPWHLIST
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv6 whitelist test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing, two static IPv6 /64 routes and IPv6 COP security whitelist
+| ... | ingress /64 filter entries applied on links TG - DUT1 and DUT2 - TG.
+| ... | DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv6 header and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 78 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps.
+| | [Tags] | 78B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-78B-1t1c-ethip6-ip6base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 78 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip6-ip6base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip6-ip6base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip6-ip6base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip6-ip6base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-78B-2t2c-ethip6-ip6base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 78 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps.
+| | [Tags] | 78B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-78B-2t2c-ethip6-ip6base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 78 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip6-ip6base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip6-ip6base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip6-ip6base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip6-ip6base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-78B-4t4c-ethip6-ip6base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 78 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps.
+| | [Tags] | 78B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-78B-4t4c-ethip6-ip6base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 78 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip6-ip6base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip6-ip6base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip6-ip6base-copwhtlistbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip6-ip6base-copwhtlistbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv6 forwarding in 3-node circular topology
+| | And Add fib table | ${dut1} | 2001:1:: | 64 | 1 | local
+| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 | local
+| | And COP Add whitelist Entry | ${dut1} | ${dut1_if1} | ip6 | 1
+| | And COP Add whitelist Entry | ${dut2} | ${dut2_if2} | ip6 | 1
+| | And COP interface enable or disable | ${dut1} | ${dut1_if1} | enable
+| | And COP interface enable or disable | ${dut2} | ${dut2_if2} | enable
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-iacldstbase-ndrchk.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-iacldstbase-ndrchk.robot
new file mode 100644
index 0000000000..b82a356f1b
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-iacldstbase-ndrchk.robot
@@ -0,0 +1,210 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.Classify.Classify
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | FEATURE | IACLDST
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput IPv6 whitelist verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing, two static IPv6 /64 routes and IPv6 iAcl security whitelist
+| ... | ingress /64 filter entries applied on links TG - DUT1 and DUT2 - TG.
+| ... | DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT confiiguration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv6 header and generated
+| ... | payload.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 78 Byte frames using single trial throughput test
+| | ... | at 2x 3.2mpps.
+| | [Tags] | 78B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 3.2mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip6-ip6base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 1518 Byte frames using single trial throughput test
+| | ... | at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip6-ip6base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 9000 Byte frames using single trial throughput test
+| | ... | at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-78B-2t2c-ethip6-ip6base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 78 Byte frames using single trial throughput test
+| | ... | at 2x 5.7mpps.
+| | [Tags] | 78B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 5.7mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip6-ip6base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 1518 Byte frames using single trial throughput test
+| | ... | at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip6-ip6base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 9000 Byte frames using single trial throughput test
+| | ... | at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-78B-4t4c-ethip6-ip6base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 thread, 4 phy core, 2 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 78 Byte frames using single trial throughput test
+| | ... | at 2x 9.2mpps.
+| | [Tags] | 78B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 9.2mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip6-ip6base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 thread, 4 phy core, 2 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 1518 Byte frames using single trial throughput test
+| | ... | at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip6-ip6base-iacldstbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 thread, 4 phy core, 2 receive queue per NIC port. [Ver] Verify
+| | ... | ref-NDR for 9000 Byte frames using single trial throughput test
+| | ... | at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-iacldstbase-ndrpdrdisc.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-iacldstbase-ndrpdrdisc.robot
new file mode 100644
index 0000000000..6ee98fbcc4
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-iacldstbase-ndrpdrdisc.robot
@@ -0,0 +1,474 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.Classify.Classify
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | FEATURE | IACLDST
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv6 iAcl whitelist test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing, two static IPv6 /64 routes and IPv6 iAcl security whitelist
+| ... | ingress /64 filter entries applied on links TG - DUT1 and DUT2 - TG.
+| ... | DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv6 header and generated payload. MAC
+| ... | addresses are matching MAC addresses of the TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 78 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps.
+| | [Tags] | 78B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-78B-1t1c-ethip6-ip6base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 78 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip6-ip6base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip6-ip6base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip6-ip6base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip6-ip6base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-78B-2t2c-ethip6-ip6base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 78 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps.
+| | [Tags] | 78B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-78B-2t2c-ethip6-ip6base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 78 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip6-ip6base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip6-ip6base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip6-ip6base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip6-ip6base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-78B-4t4c-ethip6-ip6base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 78 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps.
+| | [Tags] | 78B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-78B-4t4c-ethip6-ip6base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 78 Byte frames using binary search start at 10GE linerate,
+| | ... | step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip6-ip6base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip6-ip6base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 1518 Byte frames using binary search start at 10GE linerate,
+| | ... | step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip6-ip6base-iacldstbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find NDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip6-ip6base-iacldstbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing and whitelist filters config with \
+| | ... | 4 threads, 4 phy cores, 2 receive queue per NIC port. [Ver] Find PDR
+| | ... | for 9000 Byte frames using binary search start at 10GE linerate,
+| | ... | step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | And Initialize IPv6 iAcl whitelist in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-ndrchk.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-ndrchk.robot
new file mode 100644
index 0000000000..8011baf632
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-ndrchk.robot
@@ -0,0 +1,191 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | BASE | IP6BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput IPv6 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing and two static IPv6 /64 route entries. DUT1 and DUT2 tested with
+| ... | 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv6 header and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 78 Byte
+| | ... | frames using single trial throughput test at 2x 3.9mpps.
+| | [Tags] | 78B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 3.9mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-78B-2t2c-ethip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 78 Byte
+| | ... | frames using single trial throughput test at 2x 8.1mpps.
+| | [Tags] | 78B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 8.1mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-78B-4t4c-ethip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 78 Byte
+| | ... | frames using single trial throughput test at 2x 9.4mpps.
+| | [Tags] | 78B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 9.4mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-ndrpdrdisc.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-ndrpdrdisc.robot
new file mode 100644
index 0000000000..3999677ffb
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-ndrpdrdisc.robot
@@ -0,0 +1,436 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | BASE | IP6BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv6 routing test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing and two static IPv6 /64 route entries. DUT1 and DUT2 tested with
+| ... | 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv6 header and static payload.
+| ... | MAC addresses are matching MAC addresses of the TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-78B-1t1c-ethip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-78B-2t2c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-78B-2t2c-ethip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-78B-4t4c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-78B-4t4c-ethip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-pdrchk.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-pdrchk.robot
new file mode 100644
index 0000000000..5ca64f5790
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6base-pdrchk.robot
@@ -0,0 +1,183 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | BASE | IP6BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test
+| ...
+| Documentation | *Reference PDR throughput IPv6 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing and two static IPv6 /64 route entries. DUT1 and DUT2 tested with
+| ... | 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-PDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-PDR value is periodically updated acording to
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv6 header and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Keywords ***
+| Check PDR for IPv6 routing
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with ${wt} thread, ${wt} phy core,\
+| | ... | ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 78B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 routing
+| | framesize=${78} | rate=3.9mpps | wt=1 | rxq=1
+
+| tc02-1518B-1t1c-ethip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 routing
+| | framesize=${1518} | rate=400000pps | wt=1 | rxq=1
+
+| tc03-9000B-1t1c-ethip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 routing
+| | framesize=${9000} | rate=60000pps | wt=1 | rxq=1
+
+| tc04-78B-2t2c-ethip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 78B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 routing
+| | framesize=${78} | rate=6.0mpps | wt=2 | rxq=1
+
+| tc05-1518B-2t2c-ethip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 routing
+| | framesize=${1518} | rate=400000pps | wt=2 | rxq=1
+
+| tc06-9000B-2t2c-ethip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 routing
+| | framesize=${9000} | rate=60000pps | wt=2 | rxq=1
+
+| tc07-78B-4t4c-ethip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 78B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 routing
+| | framesize=${78} | rate=6.1mpps | wt=4 | rxq=2
+
+| tc08-1518B-4t4c-ethip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 routing
+| | framesize=${1518} | rate=400000pps | wt=4 | rxq=2
+
+| tc09-9000B-4t4c-ethip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 routing
+| | framesize=${9000} | rate=60000pps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale200k-ndrchk.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale200k-ndrchk.robot
new file mode 100644
index 0000000000..a6f8c980e1
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale200k-ndrchk.robot
@@ -0,0 +1,201 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | SCALE | FIB_200K
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput IPv6 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing and 2x100k static IPv6 /64 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT confiiguration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 100k flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv6 header and static
+| ... | payload. Incrementing of IP.dst (IPv6 destination address) field is
+| ... | applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | ${100000}
+
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6dst${rts_per_flow}
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 78 Byte
+| | ... | frames using single trial throughput test at 2x 2.3mpps.
+| | [Tags] | 78B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 2.3mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip6-ip6scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip6-ip6scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-78B-2t2c-ethip6-ip6scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 78 Byte
+| | ... | frames using single trial throughput test at 2x 5.0mpps.
+| | [Tags] | 78B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 5.0mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip6-ip6scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip6-ip6scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-78B-4t4c-ethip6-ip6scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 78 Byte
+| | ... | frames using single trial throughput test at 2x 5.7mpps.
+| | [Tags] | 78B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 5.7mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip6-ip6scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip6-ip6scale200k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale200k-ndrpdrdisc.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale200k-ndrpdrdisc.robot
new file mode 100644
index 0000000000..f64102384d
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale200k-ndrpdrdisc.robot
@@ -0,0 +1,457 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | SCALE | FIB_200K
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv6 routing test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing and 2x100k static IPv6 /64 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 100k flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv6 header withq IP and static payload.
+| ... | MAC addresses are matching MAC addresses of the TG node interfaces.
+| ... | Incrementing of IP.dst (IPv6 destination address) field is applied to
+| ... | both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+| ${rts_per_flow}= | ${100000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6dst${rts_per_flow}
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-78B-1t1c-ethip6-ip6scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip6-ip6scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip6-ip6scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip6-ip6scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip6-ip6scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-78B-2t2c-ethip6-ip6scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-78B-2t2c-ethip6-ip6scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip6-ip6scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip6-ip6scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip6-ip6scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip6-ip6scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-78B-4t4c-ethip6-ip6scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-78B-4t4c-ethip6-ip6scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip6-ip6scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip6-ip6scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip6-ip6scale200k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip6-ip6scale200k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale20k-ndrchk.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale20k-ndrchk.robot
new file mode 100644
index 0000000000..ebf642f9b5
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale20k-ndrchk.robot
@@ -0,0 +1,201 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | SCALE | FIB_20K
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput IPv6 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing and 2x10k static IPv6 /64 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT confiiguration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 10k flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv6 header and static
+| ... | payload. Incrementing of IP.dst (IPv6 destination address) field is
+| ... | applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | ${10000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6dst${rts_per_flow}
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 78 Byte
+| | ... | frames using single trial throughput test at 2x 2.9mpps.
+| | [Tags] | 78B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 2.9mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip6-ip6scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip6-ip6scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-78B-2t2c-ethip6-ip6scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 78 Byte
+| | ... | frames using single trial throughput test at 2x 7.0mpps.
+| | [Tags] | 78B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 7.0mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip6-ip6scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip6-ip6scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-78B-4t4c-ethip6-ip6scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 78 Byte
+| | ... | frames using single trial throughput test at 2x 9.6mpps.
+| | [Tags] | 78B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 10.0mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip6-ip6scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip6-ip6scale20k-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale20k-ndrpdrdisc.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale20k-ndrpdrdisc.robot
new file mode 100644
index 0000000000..4337c0d6a3
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale20k-ndrpdrdisc.robot
@@ -0,0 +1,457 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | SCALE | FIB_20K
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv6 routing test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing and 2x10k static IPv6 /64 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 10k flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv6 header withq IP and static payload.
+| ... | MAC addresses are matching MAC addresses of the TG node interfaces.
+| ... | Incrementing of IP.dst (IPv6 destination address) field is applied to
+| ... | both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+| ${rts_per_flow}= | ${10000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6dst${rts_per_flow}
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-78B-1t1c-ethip6-ip6scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip6-ip6scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip6-ip6scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip6-ip6scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip6-ip6scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-78B-2t2c-ethip6-ip6scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-78B-2t2c-ethip6-ip6scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip6-ip6scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip6-ip6scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip6-ip6scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip6-ip6scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-78B-4t4c-ethip6-ip6scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-78B-4t4c-ethip6-ip6scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip6-ip6scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip6-ip6scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip6-ip6scale20k-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip6-ip6scale20k-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale2m-ndrchk.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale2m-ndrchk.robot
new file mode 100644
index 0000000000..c3411cebd8
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale2m-ndrchk.robot
@@ -0,0 +1,210 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | SCALE | FIB_2M
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput IPv6 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing and 2x1M static IPv6 /64 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT confiiguration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 1M flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv6 header and static
+| ... | payload. Incrementing of IP.dst (IPv6 destination address) field is
+| ... | applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | ${1000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6dst${rts_per_flow}
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 78 Byte
+| | ... | frames using single trial throughput test at 2x 1.8mpps.
+| | [Tags] | 78B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 1.8mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc02-1518B-1t1c-ethip6-ip6scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip6-ip6scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc04-78B-2t2c-ethip6-ip6scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 78 Byte
+| | ... | frames using single trial throughput test at 2x 4.0mpps.
+| | [Tags] | 78B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 4.0mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc05-1518B-2t2c-ethip6-ip6scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip6-ip6scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc07-78B-4t4c-ethip6-ip6scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 78 Byte
+| | ... | frames using single trial throughput test at 2x 6.8mpps.
+| | [Tags] | 78B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 6.8mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc08-1518B-4t4c-ethip6-ip6scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip6-ip6scale2m-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize}
+| | ... | ${traffic_profile}
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale2m-ndrpdrdisc.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale2m-ndrpdrdisc.robot
new file mode 100644
index 0000000000..1259fd9130
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale2m-ndrpdrdisc.robot
@@ -0,0 +1,457 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | SCALE | FIB_2M
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv6 routing test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing and 2x1M static IPv6 /64 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 1M flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv6 header withq IP and static payload.
+| ... | MAC addresses are matching MAC addresses of the TG node interfaces.
+| ... | Incrementing of IP.dst (IPv6 destination address) field is applied to
+| ... | both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+| ${rts_per_flow}= | ${1000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6dst${rts_per_flow}
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-78B-1t1c-ethip6-ip6scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip6-ip6scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip6-ip6scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip6-ip6scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip6-ip6scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-78B-2t2c-ethip6-ip6scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-78B-2t2c-ethip6-ip6scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip6-ip6scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip6-ip6scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip6-ip6scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip6-ip6scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-78B-4t4c-ethip6-ip6scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-78B-4t4c-ethip6-ip6scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 78 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip6-ip6scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip6-ip6scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip6-ip6scale2m-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip6-ip6scale2m-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale2m-pdrchk.robot b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale2m-pdrchk.robot
new file mode 100644
index 0000000000..977e73ff2a
--- /dev/null
+++ b/tests/vpp/perf/ip6/10ge2p1x520-ethip6-ip6scale2m-pdrchk.robot
@@ -0,0 +1,185 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP6FWD | SCALE | FIB_2M
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test
+| ...
+| Documentation | *Reference PDR throughput IPv6 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing and 2x1M static IPv6 /64 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-PDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-PDR value is periodically updated acording to
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long
+| ... | performance tests for the same DUT confiiguration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 1M flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv6 header and static
+| ... | payload. Incrementing of IP.dst (IPv6 destination address) field is
+| ... | applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | ${1000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6dst${rts_per_flow}
+
+*** Keywords ***
+| Check PDR for IPv6 scaled routing
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with ${wt} thread, ${wt} phy core,\
+| | ... | ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding with scaling in 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 78B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 scaled routing
+| | framesize=${78} | rate=1.8mpps | wt=1 | rxq=1
+
+| tc02-1518B-1t1c-ethip6-ip6scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 scaled routing
+| | framesize=${1518} | rate=400000pps | wt=1 | rxq=1
+
+| tc03-9000B-1t1c-ethip6-ip6scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 scaled routing
+| | framesize=${9000} | rate=60000pps | wt=1 | rxq=1
+
+| tc04-78B-2t2c-ethip6-ip6scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 78B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 scaled routing
+| | framesize=${78} | rate=4.0mpps | wt=2 | rxq=1
+
+| tc05-1518B-2t2c-ethip6-ip6scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 scaled routing
+| | framesize=${1518} | rate=400000pps | wt=2 | rxq=1
+
+| tc06-9000B-2t2c-ethip6-ip6scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 scaled routing
+| | framesize=${900} | rate=138580pps | wt=2 | rxq=1
+
+| tc07-78B-4t4c-ethip6-ip6scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 78B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 scaled routing
+| | framesize=${78} | rate=6.4mpps | wt=4 | rxq=2
+
+| tc08-1518B-4t4c-ethip6-ip6scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 scaled routing
+| | framesize=${1518} | rate=400000pps | wt=4 | rxq=2
+
+| tc09-9000B-4t4c-ethip6-ip6scale2m-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv6 scaled routing
+| | framesize=${9000} | rate=60000pps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/ip6/40ge2p1xl710-ethip6-ip6base-ndrpdrdisc.robot b/tests/vpp/perf/ip6/40ge2p1xl710-ethip6-ip6base-ndrpdrdisc.robot
new file mode 100644
index 0000000000..4c45fdd266
--- /dev/null
+++ b/tests/vpp/perf/ip6/40ge2p1xl710-ethip6-ip6base-ndrpdrdisc.robot
@@ -0,0 +1,183 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-XL710 | ETH | IP6FWD | BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-XL710
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv6 routing test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6 for IPv6 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6
+| ... | routing and two static IPv6 /64 route entries. DUT1 and DUT2 tested with
+| ... | 2p40GE NIC XL710 by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv6 header and static payload.
+| ... | MAC addresses are matching MAC addresses of the TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 18.75Mpps rate, step 100kpps.
+| | [Tags] | 78B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc03-1518B-1t1c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 24.5G rate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc07-78B-2t2c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 18.75Mpps rate, step 100kpps.
+| | [Tags] | 78B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc09-1518B-2t2c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 24.5G rate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc13-78B-4t4c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 78 Byte frames
+| | ... | using binary search start at 18.75Mpps rate, step 100kpps.
+| | [Tags] | 78B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc15-1518B-4t4c-ethip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 24.5G rate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize IPv6 forwarding in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
diff --git a/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip4-ip6base-ndrchk.robot b/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip4-ip6base-ndrchk.robot
new file mode 100644
index 0000000000..6448d4390f
--- /dev/null
+++ b/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip4-ip6base-ndrchk.robot
@@ -0,0 +1,266 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/overlay/lisp_static_adjacency.robot
+| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | IP6FWD | ENCAP | LISP | IP4UNRLAY | IP6OVRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput Lisp tunnel verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6-LISP-IPv4 on DUT1-DUT2,\
+| ... | Eth-IPv6 on TG-DUTn for IPv6 routing over LISPoIPv4 tunnel.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6\
+| ... | routing and static routes. LISPoIPv4 tunnel is configured between\
+| ... | DUT1 and DUT2. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic\
+| ... | by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies\
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet\
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to\
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long\
+| ... | performance tests for the same DUT confiiguration. Test packets are\
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3\
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with\
+| ... | all packets containing Ethernet header, IPv4 header or IPv6 header with\
+| ... | IP protocol=61 and generated payload.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6lispip4-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x 1.75mpps.
+| | [Tags] | 78B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 1.75mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1460B-1t1c-ethip6lispip4-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x 720000pps.
+| | [Tags] | 1460B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1460}
+| | ${rate}= | Set Variable | 720000pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip6lispip4-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-78B-2t2c-ethip6lispip4-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x 3.42mpps.
+| | [Tags] | 78B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 3.42mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1460B-2t2c-ethip6lispip4-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x 720000pps.
+| | [Tags] | 1460B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1460}
+| | ${rate}= | Set Variable | 720000pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip6lispip4-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-78B-4t4c-ethip6lispip4-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x 3.42mpps.
+| | [Tags] | 78B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 3.42mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1460B-4t4c-ethip6lispip4-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x 720000pps.
+| | [Tags] | 1460B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1460}
+| | ${rate}= | Set Variable | 720000pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip6lispip4-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip4-ip6base-ndrpdrdisc.robot b/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip4-ip6base-ndrpdrdisc.robot
new file mode 100644
index 0000000000..bcf0711c94
--- /dev/null
+++ b/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip4-ip6base-ndrpdrdisc.robot
@@ -0,0 +1,596 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/overlay/lisp_static_adjacency.robot
+| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | IP6FWD | ENCAP | LISP | IP4UNRLAY | IP6OVRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC6830: Pkt throughput Lisp test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology\
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6-LISP-IPv4 on DUT1-DUT2,\
+| ... | Eth-IPv6 on TG-DUTn for IPv6 routing over LISPoIPv4 tunnel.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6\
+| ... | routing and static routes. LISPoIPv4 tunnel is configured between\
+| ... | DUT1 and DUT2. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic\
+| ... | by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop\
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\
+| ... | of packets transmitted. NDR and PDR are discovered for different\
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | *[Ref] Applicable standard specifications:* RFC6830.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6lispip4-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 78 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-78B-1t1c-ethip6lispip4-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 78 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1460B-1t1c-ethip6lispip4-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps.
+| | [Tags] | 1460B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1460B-1t1c-ethip6lispip4-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1460B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip6lispip4-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip6lispip4-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-78B-2t2c-ethip6lispip4-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 78 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-78B-2t2c-ethip6lispip4-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 78 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1460B-2t2c-ethip6lispip4-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps.
+| | [Tags] | 1460B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1460B-2t2c-ethip6lispip4-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1460B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip6lispip4-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip6lispip4-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-78B-4t4c-ethip6lispip4-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 78 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-78B-4t4c-ethip6lispip4-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 78 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1460B-4t4c-ethip6lispip4-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps.
+| | [Tags] | 1460B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1460B-4t4c-ethip6lispip4-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1460B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip6lispip4-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip6lispip4-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 48}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip4-ip6base-pdrchk.robot b/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip4-ip6base-pdrchk.robot
new file mode 100644
index 0000000000..55c5a19e70
--- /dev/null
+++ b/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip4-ip6base-pdrchk.robot
@@ -0,0 +1,193 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/overlay/lisp_static_adjacency.robot
+| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | IP6FWD | ENCAP | LISP | IP4UNRLAY | IP6OVRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test
+| ...
+| Documentation | *Reference PDR throughput Lisp tunnel verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6-LISP-IPv4 on DUT1-DUT2,\
+| ... | Eth-IPv6 on TG-DUTn for IPv6 routing over LISPoIPv4 tunnel.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6\
+| ... | routing and static routes. LISPoIPv4 tunnel is configured between\
+| ... | DUT1 and DUT2. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic\
+| ... | by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies\
+| ... | DUTs' throughput at ref-PDR (reference Non Drop Rate) with zero packet\
+| ... | loss tolerance. Ref-PDR value is periodically updated acording to\
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long\
+| ... | performance tests for the same DUT confiiguration. Test packets are\
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3\
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with\
+| ... | all packets containing Ethernet header, IPv4 header or IPv6 header with\
+| ... | IP protocol=61 and generated payload.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Keywords ***
+| Check PDR for Lisp IPv6 over IPv4 forwarding
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with ${wt} thread, ${wt} phy core,\
+| | ... | ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 over IPv4 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6o4} | ${dut1_to_tg_ip6o4} | ${dut2_to_dut1_ip6o4}
+| | ... | ${dut2_to_tg_ip6o4} | ${tg_prefix6o4} | ${dut_prefix6o4}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6o4_eid} | ${dut2_ip6o4_eid}
+| | ... | ${dut1_ip6o4_static_adjacency} | ${dut2_ip6o4_static_adjacency}
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6lispip4-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 78B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv4 forwarding
+| | framesize=${78} | rate=1.75mpps | wt=1 | rxq=1
+
+| tc02-1460B-1t1c-ethip6lispip4-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1460B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv4 forwarding
+| | framesize=${1460} | rate=37000pps | wt=1 | rxq=1
+
+| tc03-9000B-1t1c-ethip6lispip4-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv4 forwarding
+| | framesize=${9000} | rate=60000pps | wt=1 | rxq=1
+
+| tc04-78B-2t2c-ethip6lispip4-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 78B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv4 forwarding
+| | framesize=${78} | rate=3.42mpps | wt=2 | rxq=1
+
+| tc05-1460B-2t2c-ethip6lispip4-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1460B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv4 forwarding
+| | framesize=${1460} | rate=370000pps | wt=2 | rxq=1
+
+| tc06-9000B-2t2c-ethip6lispip4-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv4 forwarding
+| | framesize=${9000} | rate=60000pps | wt=2 | rxq=1
+
+| tc07-78B-4t4c-ethip6lispip4-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 78B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv4 forwarding
+| | framesize=${78} | rate=3.42mpps | wt=4 | rxq=2
+
+| tc08-1460B-4t4c-ethip6lispip4-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1460B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv4 forwarding
+| | framesize=${1460} | rate=370000pps | wt=4 | rxq=2
+
+| tc09-9000B-4t4c-ethip6lispip4-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv4 forwarding
+| | framesize=${9000} | rate=60000pps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip6-ip6base-ndrchk.robot b/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip6-ip6base-ndrchk.robot
new file mode 100644
index 0000000000..2e44791318
--- /dev/null
+++ b/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip6-ip6base-ndrchk.robot
@@ -0,0 +1,266 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/overlay/lisp_static_adjacency.robot
+| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | IP6FWD | ENCAP | LISP | IP6UNRLAY | IP6OVRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput Lisp tunnel verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology\
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6-LISP-IPv6 on DUT1-DUT2,\
+| ... | Eth-IPv6-ICMPv6 on TG-DUTn for IPv6 routing over LISPoIPv6 tunnel.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6\
+| ... | routing and static routes. LISPoIPv6 tunnel is configured\
+| ... | between DUT1 and DUT2. DUT1 and DUT2 tested with 2p10GE NIC X520\
+| ... | Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies\
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet\
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to\
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long\
+| ... | performance tests for the same DUT confiiguration. Test packets are\
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3\
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with\
+| ... | all packets containing Ethernet header, IPv6 header with\
+| ... | IP protocol=61 and generated payload.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6lispip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x 1.56mpps.
+| | [Tags] | 78B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 1.56mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1460B-1t1c-ethip6lispip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x 740000pps.
+| | [Tags] | 1460B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1460}
+| | ${rate}= | Set Variable | 740000pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-ethip6lispip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-78B-2t2c-ethip6lispip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 thread, 2 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x 3.2mpps.
+| | [Tags] | 78B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 3.2mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1460B-2t2c-ethip6lispip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x 740000pps.
+| | [Tags] | 1460B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1460}
+| | ${rate}= | Set Variable | 740000pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-ethip6lispip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-78B-4t4c-ethip6lispip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x 3.2mpps.
+| | [Tags] | 78B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${78}
+| | ${rate}= | Set Variable | 3.2mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1460B-4t4c-ethip6lispip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x 740000pps.
+| | [Tags] | 1460B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1460}
+| | ${rate}= | Set Variable | 740000pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-ethip6lispip6-ip6base-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip6-ip6base-ndrpdrdisc.robot b/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip6-ip6base-ndrpdrdisc.robot
new file mode 100644
index 0000000000..f29f4e4626
--- /dev/null
+++ b/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip6-ip6base-ndrpdrdisc.robot
@@ -0,0 +1,596 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/overlay/lisp_static_adjacency.robot
+| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | IP6FWD | ENCAP | LISP | IP6UNRLAY | IP6OVRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC6830: Pkt throughput Lisp test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology\
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6-LISP-IPv6 on DUT1-DUT2,\
+| ... | Eth-IPv6 on TG-DUTn for IPv6 routing over LISPoIPv6 tunnel.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6\
+| ... | routing and static routes. LISPoIPv6 tunnel is configured between\
+| ... | DUT1 and DUT2. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic\
+| ... | by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop\
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\
+| ... | of packets transmitted. NDR and PDR are discovered for different\
+| ... | Ethernet L2 frame sizes using either binary search or linear search\
+| ... | *[Ref] Applicable standard specifications:* RFC6830.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6lispip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 78 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-78B-1t1c-ethip6lispip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 78 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1460B-1t1c-ethip6lispip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps.
+| | [Tags] | 1460B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1460B-1t1c-ethip6lispip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1460B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-ethip6lispip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-ethip6lispip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 1 thread, 1 phy core, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-78B-2t2c-ethip6lispip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 78 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-78B-2t2c-ethip6lispip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 78 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1460B-2t2c-ethip6lispip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps.
+| | [Tags] | 1460B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1460B-2t2c-ethip6lispip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1460B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-ethip6lispip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-ethip6lispip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 2 threads, 2 phy cores, 1 receive queue per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-78B-4t4c-ethip6lispip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 78 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 78B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-78B-4t4c-ethip6lispip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 78 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 78B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${78}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1460B-4t4c-ethip6lispip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 100kpps.
+| | [Tags] | 1460B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1460B-4t4c-ethip6lispip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 1460 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1460B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1460}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-ethip6lispip6-ip6base-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-ethip6lispip6-ip6base-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv6 LISP remote static mappings and whitelist\
+| | ... | filters config with 4 threads, 4 phy cores, 2 receive queues per NIC\
+| | ... | port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start\
+| | ... | at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip6-ip6base-pdrchk.robot b/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip6-ip6base-pdrchk.robot
new file mode 100644
index 0000000000..717efedf45
--- /dev/null
+++ b/tests/vpp/perf/ip6_tunnels/10ge2p1x520-ethip6lispip6-ip6base-pdrchk.robot
@@ -0,0 +1,193 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/overlay/lisp_static_adjacency.robot
+| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | IP6FWD | ENCAP | LISP | IP6UNRLAY | IP6OVRLAY
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test
+| ...
+| Documentation | *Reference PDR throughput Lisp tunnel verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology\
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv6-LISP-IPv6 on DUT1-DUT2,\
+| ... | Eth-IPv6-ICMPv6 on TG-DUTn for IPv6 routing over LISPoIPv6 tunnel.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv6\
+| ... | routing and static routes. LISPoIPv6 tunnel is configured\
+| ... | between DUT1 and DUT2. DUT1 and DUT2 tested with 2p10GE NIC X520\
+| ... | Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies\
+| ... | DUTs' throughput at ref-PDR (reference Non Drop Rate) with zero packet\
+| ... | loss tolerance. Ref-PDR value is periodically updated acording to\
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long\
+| ... | performance tests for the same DUT confiiguration. Test packets are\
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3\
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with\
+| ... | all packets containing Ethernet header, IPv6 header with\
+| ... | IP protocol=61 and generated payload.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip6-ip6src253
+
+*** Keywords ***
+| Check PDR for Lisp IPv6 over IPv6 forwarding
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with ${wt} thread, ${wt} phy core,\
+| | ... | ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize LISP IPv6 forwarding in 3-node circular topology
+| | ... | ${dut1_to_dut2_ip6} | ${dut1_to_tg_ip6} | ${dut2_to_dut1_ip6}
+| | ... | ${dut2_to_tg_ip6} | ${prefix6}
+| | And Configure LISP topology in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${NONE}
+| | ... | ${dut2} | ${dut2_if1} | ${NONE}
+| | ... | ${duts_locator_set} | ${dut1_ip6_eid} | ${dut2_ip6_eid}
+| | ... | ${dut1_ip6_static_adjacency} | ${dut2_ip6_static_adjacency}
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-78B-1t1c-ethip6lispip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 78B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv6 forwarding
+| | framesize=${78} | rate=1.56mpps | wt=1 | rxq=1
+
+| tc02-1460B-1t1c-ethip6lispip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1460B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv6 forwarding
+| | framesize=${1460} | rate=370000pps | wt=1 | rxq=1
+
+| tc03-9000B-1t1c-ethip6lispip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv6 forwarding
+| | framesize=${9000} | rate=60000pps | wt=1 | rxq=1
+
+| tc04-78B-2t2c-ethip6lispip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 thread, 2 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 78B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv6 forwarding
+| | framesize=${78} | rate=3.2mpps | wt=2 | rxq=1
+
+| tc05-1460B-2t2c-ethip6lispip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1460B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv6 forwarding
+| | framesize=${1460} | rate=370000pps | wt=2 | rxq=1
+
+| tc06-9000B-2t2c-ethip6lispip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv6 forwarding
+| | framesize=${9000} | rate=60000pps | wt=2 | rxq=1
+
+| tc07-78B-4t4c-ethip6lispip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 78 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 78B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv6 forwarding
+| | framesize=${78} | rate=3.2mpps | wt=4 | rxq=2
+
+| tc08-1460B-4t4c-ethip6lispip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1460 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1460B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv6 forwarding
+| | framesize=${1460} | rate=370000pps | wt=4 | rxq=2
+
+| tc09-9000B-4t4c-ethip6lispip6-ip6base-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs LISP tunnel config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for Lisp IPv6 over IPv6 forwarding
+| | framesize=${9000} | rate=60000pps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/l2/10ge2p1vic1227-eth-l2bdbasemaclrn-ndrpdrdisc.robot b/tests/vpp/perf/l2/10ge2p1vic1227-eth-l2bdbasemaclrn-ndrpdrdisc.robot
new file mode 100644
index 0000000000..51eb235f24
--- /dev/null
+++ b/tests/vpp/perf/l2/10ge2p1vic1227-eth-l2bdbasemaclrn-ndrpdrdisc.robot
@@ -0,0 +1,288 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Cisco-VIC-1227 | ETH | L2BDMACLRN | BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Cisco-VIC-1227
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2BD test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology\
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-\
+| ... | domain and MAC learning enabled. DUT1 and DUT2 tested with 2p10GE NIC
+| ... | Cisco-VIC-1227.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop\
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\
+| ... | of packets transmitted. NDR and PDR are discovered for different\
+| ... | Ethernet L2 frame sizes using either binary search or linear search\
+| ... | algorithms with configured starting rate and final step that determines\
+| ... | throughput measurement resolution. Test packets are generated by TG on\
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups\
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets\
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static\
+| ... | payload. MAC addresses are matching MAC addresses of the TG node\
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Cisco-VIC-1227 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Keywords ***
+| L2 Bridge Domain NDR Binary Search
+| | [Arguments] | ${framesize} | ${min_rate} | ${wt} | ${rxq}
+| | # Test Variables required for test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${min_rate}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Documentation | [Cfg] DUT runs L2BD switching config with ${wt}\
+| | Set Test Documentation | thread, ${wt} phy core, ${rxq}\ | append=True
+| | Set Test Documentation | receive queue per NIC port. | append=True
+| | Set Test Documentation | [Ver] Find NDR for ${framesize} Byte\ | append=True
+| | Set Test Documentation | frames using binary search start at\ | append=True
+| | Set Test Documentation | 10GE linerate, step ${threshold}pps. | append=True
+| | Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | Add PCI devices to DUTs in 3-node single link topology
+| | Run Keyword If | ${framesize} < ${1522} | Add no multi seg to all DUTs
+| | Apply startup configuration on all VPP DUTs
+| | Initialize L2 bridge domain in 3-node circular topology
+| | Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| L2 Bridge Domain PDR Binary Search
+| | [Arguments] | ${framesize} | ${min_rate} | ${wt} | ${rxq}
+| | # Test Variables required for test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${min_rate}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Set Test Documentation | [Cfg] DUT runs L2BD switching config with ${wt}\
+| | Set Test Documentation | thread, ${wt} phy core, ${rxq}\ | append=True
+| | Set Test Documentation | receive queue per NIC port. | append=True
+| | Set Test Documentation | [Ver] Find PDR for ${framesize} Byte\ | append=True
+| | Set Test Documentation | frames using binary search start at\ | append=True
+| | Set Test Documentation | 10GE linerate, step ${threshold}pps, | append=True
+| | Set Test Documentation | LT=${perf_pdr_loss_acceptance} | append=True
+| | Set Test Documentation | ${perf_pdr_loss_acceptance_type}. | append=True
+| | Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | Add PCI devices to DUTs in 3-node single link topology
+| | Run Keyword If | ${framesize} < ${1522} | Add no multi seg to all DUTs
+| | Apply startup configuration on all VPP DUTs
+| | Initialize L2 bridge domain in 3-node circular topology
+| | Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=1 | rxq=1
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc02-64B-1t1c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=1 | rxq=1
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc03-1518B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=1 | rxq=1
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc04-1518B-1t1c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=1 | rxq=1
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc05-9000B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=1 | rxq=1
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc06-9000B-1t1c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=1 | rxq=1
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc07-64B-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=2 | rxq=1
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc08-64B-2t2c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=2 | rxq=1
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc09-1518B-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=2 | rxq=1
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc10-1518B-2t2c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=2 | rxq=1
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc11-9000B-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=2 | rxq=1
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc12-9000B-2t2c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=2 | rxq=1
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc13-64B-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=4 | rxq=2
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc14-64B-4t4c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=4 | rxq=2
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc15-1518B-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=4 | rxq=2
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc16-1518B-4t4c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=4 | rxq=2
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc17-9000B-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=4 | rxq=2
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc18-9000B-4t4c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=4 | rxq=2
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Template] | L2 Bridge Domain PDR Binary Search
diff --git a/tests/vpp/perf/l2/10ge2p1x520-dot1ad-l2xcbase-ndrchk.robot b/tests/vpp/perf/l2/10ge2p1x520-dot1ad-l2xcbase-ndrchk.robot
new file mode 100644
index 0000000000..4b4162581a
--- /dev/null
+++ b/tests/vpp/perf/l2/10ge2p1x520-dot1ad-l2xcbase-ndrchk.robot
@@ -0,0 +1,279 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/l2/tagging.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | L2XCFWD | BASE | DOT1AD
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput L2XC with 802.1ad verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 xconnect.
+| ... | 802.1ad tagging is applied on link between DUT1 and DUT2 with inner 4B
+| ... | vlan tag (id=100) and outer 4B vlan tag (id=200).
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${subid}= | 10
+| ${outer_vlan_id}= | 100
+| ${inner_vlan_id}= | 200
+| ${type_subif}= | two_tags
+| ${tag_rewrite}= | pop-2
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-dot1ad-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC with 802.1ad config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 4.0mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 4.0mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1514B-1t1c-dot1ad-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC with 802.1ad config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1514 Byte
+| | ... | frames using single trial throughput test at 2x 810635pps.
+| | [Tags] | 1514B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1514}
+| | ${rate}= | Set Variable | 810635pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-dot1ad-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC with 802.1ad config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138458pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138458pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-64B-2t2c-dot1ad-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC with 802.1ad config with 2 thread, 2 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 8.5mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 8.5mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1514B-2t2c-dot1ad-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC with 802.1ad config with 2 thread, 2 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1514 Byte
+| | ... | frames using single trial throughput test at 2x 810635pps.
+| | [Tags] | 1514B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1514}
+| | ${rate}= | Set Variable | 810635pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-dot1ad-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC with 802.1ad config with 2 thread, 2 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138458pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138458pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-64B-4t4c-dot1ad-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC with 802.1ad config with 4 thread, 4 phy core, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 8.6mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 8.6mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1514B-4t4c-dot1ad-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC with 802.1ad config with 4 thread, 4 phy core, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 1514 Byte
+| | ... | frames using single trial throughput test at 2x 810635pps.
+| | [Tags] | 1514B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1514}
+| | ${rate}= | Set Variable | 810635pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-dot1ad-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC with 802.1ad config with 4 thread, 4 phy core, \
+| | ... | 2 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138458pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138458pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/l2/10ge2p1x520-dot1ad-l2xcbase-ndrpdrdisc.robot b/tests/vpp/perf/l2/10ge2p1x520-dot1ad-l2xcbase-ndrpdrdisc.robot
new file mode 100644
index 0000000000..17b907c91b
--- /dev/null
+++ b/tests/vpp/perf/l2/10ge2p1x520-dot1ad-l2xcbase-ndrpdrdisc.robot
@@ -0,0 +1,606 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/l2/tagging.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | L2XCFWD | BASE | DOT1AD
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2XC with 802.1ad test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 xconnect.
+| ... | 802.1ad tagging is applied on link between DUT1 and DUT2 with inner 4B
+| ... | vlan tag (id=100) and outer 4B vlan tag (id=200).
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | static payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${subid}= | 10
+| ${outer_vlan_id}= | 100
+| ${inner_vlan_id}= | 200
+| ${type_subif}= | two_tags
+| ${tag_rewrite}= | pop-2
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-dot1ad-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-dot1ad-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1514B-1t1c-dot1ad-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1514 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1514B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1514}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1514B-1t1c-dot1ad-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1514 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1514B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1514}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-dot1ad-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-dot1ad-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-dot1ad-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-dot1ad-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1514B-2t2c-dot1ad-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1514 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1514B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1514}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1514B-2t2c-dot1ad-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1514 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1514B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1514}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-dot1ad-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-dot1ad-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-dot1ad-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-dot1ad-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1514B-4t4c-dot1ad-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1514 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1514B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1514}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1514B-4t4c-dot1ad-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1514 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1514B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1514}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-dot1ad-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-dot1ad-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC forwarding config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 8}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | ... | ${outer_vlan_id} | ${inner_vlan_id} | ${type_subif}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/l2/10ge2p1x520-dot1q-l2xcbase-ndrchk.robot b/tests/vpp/perf/l2/10ge2p1x520-dot1q-l2xcbase-ndrchk.robot
new file mode 100644
index 0000000000..d3a67f3777
--- /dev/null
+++ b/tests/vpp/perf/l2/10ge2p1x520-dot1q-l2xcbase-ndrchk.robot
@@ -0,0 +1,266 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/l2/tagging.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | L2XCFWD | BASE | DOT1Q
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput L2XC with 802.1q verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 cross connect. 802.1q
+| ... | tagging is applied on link between DUT1 and DUT2.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${subid}= | 10
+| ${tag_rewrite}= | pop-1
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-dot1q-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 4.0mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 4.0mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-dot1q-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 720000pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 720000pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-dot1q-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-64B-2t2c-dot1q-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 8.7mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 8.7mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-dot1q-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 720000pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 720000pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-dot1q-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-64B-4t4c-dot1q-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 10.3mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 10.3mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-dot1q-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 720000pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 720000pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-dot1q-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 120000pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 120000pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/l2/10ge2p1x520-dot1q-l2xcbase-ndrpdrdisc.robot b/tests/vpp/perf/l2/10ge2p1x520-dot1q-l2xcbase-ndrpdrdisc.robot
new file mode 100644
index 0000000000..a5a8580f29
--- /dev/null
+++ b/tests/vpp/perf/l2/10ge2p1x520-dot1q-l2xcbase-ndrpdrdisc.robot
@@ -0,0 +1,584 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Resource | resources/libraries/robot/l2/tagging.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | L2XCFWD | BASE | DOT1Q
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2XC with 802.1q test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 cross connect. 802.1q
+| ... | tagging is applied on link between DUT1 and DUT2.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${subid}= | 10
+| ${tag_rewrite}= | pop-1
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-dot1q-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-dot1q-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-dot1q-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-dot1q-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-dot1q-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-dot1q-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-dot1q-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-dot1q-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-dot1q-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-dot1q-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-dot1q-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-dot1q-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-dot1q-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-dot1q-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-dot1q-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-dot1q-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-dot1q-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-dot1q-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + 4}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Set interfaces in path in 3-node circular topology up
+| | When Initialize VLAN dot1q sub-interfaces in 3-node circular topology
+| | ... | ${dut1} | ${dut1_if2} | ${dut2} | ${dut2_if1} | ${subid}
+| | And Configure L2 tag rewrite method on interfaces
+| | ... | ${dut1} | ${subif_index_1} | ${dut2} | ${subif_index_2}
+| | ... | ${tag_rewrite}
+| | And Connect interfaces and VLAN sub-interfaces using L2XC
+| | ... | ${dut1} | ${dut1_if1} | ${subif_index_1}
+| | ... | ${dut2} | ${dut2_if2} | ${subif_index_2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/l2/10ge2p1x520-eth-l2bdbasemaclrn-ndrchk.robot b/tests/vpp/perf/l2/10ge2p1x520-eth-l2bdbasemaclrn-ndrchk.robot
new file mode 100644
index 0000000000..598d45f6b6
--- /dev/null
+++ b/tests/vpp/perf/l2/10ge2p1x520-eth-l2bdbasemaclrn-ndrchk.robot
@@ -0,0 +1,192 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.NodePath
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | L2BDMACLRN | BASE | L2BDBASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput L2BD verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-
+| ... | domain and MAC learning enabled. DUT1 and DUT2 tested with 2p10GE NIC
+| ... | X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2bdbasemaclrn-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 4.9mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 4.9mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-eth-l2bdbasemaclrn-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-eth-l2bdbasemaclrn-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-64B-2t2c-eth-l2bdbasemaclrn-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 10.4mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 10.4mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-eth-l2bdbasemaclrn-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-eth-l2bdbasemaclrn-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-64B-4t4c-eth-l2bdbasemaclrn-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 10.4mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 10.4mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-eth-l2bdbasemaclrn-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-eth-l2bdbasemaclrn-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/l2/10ge2p1x520-eth-l2bdbasemaclrn-ndrpdrdisc.robot b/tests/vpp/perf/l2/10ge2p1x520-eth-l2bdbasemaclrn-ndrpdrdisc.robot
new file mode 100644
index 0000000000..5234981c14
--- /dev/null
+++ b/tests/vpp/perf/l2/10ge2p1x520-eth-l2bdbasemaclrn-ndrpdrdisc.robot
@@ -0,0 +1,274 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | L2BDMACLRN | BASE | L2BDBASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2BD test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology\
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-\
+| ... | domain and MAC learning enabled. DUT1 and DUT2 tested with 2p10GE NI
+| ... | X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop\
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\
+| ... | of packets transmitted. NDR and PDR are discovered for different\
+| ... | Ethernet L2 frame sizes using either binary search or linear search\
+| ... | algorithms with configured starting rate and final step that determines\
+| ... | throughput measurement resolution. Test packets are generated by TG on\
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups\
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets\
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static\
+| ... | payload. MAC addresses are matching MAC addresses of the TG node\
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Keywords ***
+| L2 Bridge Domain NDR Binary Search
+| | [Arguments] | ${framesize} | ${min_rate} | ${wt} | ${rxq}
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${min_rate}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | Add PCI devices to DUTs in 3-node single link topology
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | Run Keyword If | ${get_framesize} < ${1522} | Add no multi seg to all DUTs
+| | Apply startup configuration on all VPP DUTs
+| | Initialize L2 bridge domain in 3-node circular topology
+| | Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| L2 Bridge Domain PDR Binary Search
+| | [Arguments] | ${framesize} | ${min_rate} | ${wt} | ${rxq}
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${min_rate}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | Add PCI devices to DUTs in 3-node single link topology
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | Run Keyword If | ${get_framesize} < ${1522} | Add no multi seg to all DUTs
+| | Apply startup configuration on all VPP DUTs
+| | Initialize L2 bridge domain in 3-node circular topology
+| | Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=1 | rxq=1
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc02-64B-1t1c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=1 | rxq=1
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc03-1518B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=1 | rxq=1
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc04-1518B-1t1c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=1 | rxq=1
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc05-9000B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=1 | rxq=1
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc06-9000B-1t1c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=1 | rxq=1
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc07-64B-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=2 | rxq=1
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc08-64B-2t2c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=2 | rxq=1
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc09-1518B-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=2 | rxq=1
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc10-1518B-2t2c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=2 | rxq=1
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc11-9000B-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=2 | rxq=1
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc12-9000B-2t2c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=2 | rxq=1
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc13-64B-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=4 | rxq=2
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc14-64B-4t4c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=4 | rxq=2
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc15-1518B-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=4 | rxq=2
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc16-1518B-4t4c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=4 | rxq=2
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc17-9000B-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=4 | rxq=2
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc18-9000B-4t4c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=4 | rxq=2
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
diff --git a/tests/vpp/perf/l2/10ge2p1x520-eth-l2bdbasemaclrn-pdrchk.robot b/tests/vpp/perf/l2/10ge2p1x520-eth-l2bdbasemaclrn-pdrchk.robot
new file mode 100644
index 0000000000..58512392f7
--- /dev/null
+++ b/tests/vpp/perf/l2/10ge2p1x520-eth-l2bdbasemaclrn-pdrchk.robot
@@ -0,0 +1,184 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.NodePath
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | L2BDMACLRN | BASE | L2BDBASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test
+| ...
+| Documentation | *Reference PDR throughput L2BD verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-
+| ... | domain and MAC learning enabled. DUT1 and DUT2 tested with 2p10GE NIC
+| ... | X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-PDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-PDR value is periodically updated acording to
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 254 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Keywords ***
+| Check PDR for L2 bridge domain
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with ${wt} thread, ${wt} phy core,\
+| | ... | ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain
+| | framesize=${64} | rate=4.9mpps | wt=1 | rxq=1
+
+| tc02-1518B-1t1c-eth-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain
+| | framesize=${1518} | rate=772105pps | wt=1 | rxq=1
+
+| tc03-9000B-1t1c-eth-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain
+| | framesize=${9000} | rate=110000pps | wt=1 | rxq=1
+
+| tc04-64B-2t2c-eth-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 2 thread, 2 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain
+| | framesize=${64} | rate=10.4mpps | wt=2 | rxq=1
+
+| tc05-1518B-2t2c-eth-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 2 thread, 2 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain
+| | framesize=${1518} | rate=812743pps | wt=2 | rxq=1
+
+| tc06-9000B-2t2c-eth-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 2 thread, 2 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain
+| | framesize=${9000} | rate=100000pps | wt=2 | rxq=1
+
+| tc07-64B-4t4c-eth-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 4 thread, 4 phy core,\
+| | ... | 2 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain
+| | framesize=${64} | rate=10.4mpps | wt=4 | rxq=2
+
+| tc08-1518B-4t4c-eth-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 4 thread, 4 phy core,\
+| | ... | 2 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain
+| | framesize=${1518} | rate=812743pps | wt=4 | rxq=2
+
+| tc09-9000B-4t4c-eth-l2bdbasemaclrn-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD config with 4 thread, 4 phy core,\
+| | ... | 2 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain
+| | framesize=${9000} | rate=138580pps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/l2/10ge2p1x520-eth-l2xcbase-ndrchk.robot b/tests/vpp/perf/l2/10ge2p1x520-eth-l2xcbase-ndrchk.robot
new file mode 100644
index 0000000000..13a2b7577f
--- /dev/null
+++ b/tests/vpp/perf/l2/10ge2p1x520-eth-l2xcbase-ndrchk.robot
@@ -0,0 +1,192 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.InterfaceUtil
+| Library | resources.libraries.python.NodePath
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | L2XCFWD | BASE | L2XCBASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance ndrchk test
+| ...
+| Documentation | *Reference NDR throughput L2XC verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 cross connect.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 5.9mpps.
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 5.9mpps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc02-1518B-1t1c-eth-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc03-9000B-1t1c-eth-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc04-64B-2t2c-eth-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 10.4mpps.
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 10.4mpps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc05-1518B-2t2c-eth-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc06-9000B-2t2c-eth-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc07-64B-4t4c-eth-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test at 2x 10.4mpps.
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${64}
+| | ${rate}= | Set Variable | 10.4mpps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc08-1518B-4t4c-eth-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test at 2x 812743pps.
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${1518}
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
+
+| tc09-9000B-4t4c-eth-l2xcbase-ndrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test at 2x 138580pps.
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ${framesize}= | Set Variable | ${9000}
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Traffic should pass with no loss | ${perf_trial_duration} | ${rate}
+| | ... | ${framesize} | ${traffic_profile}
diff --git a/tests/vpp/perf/l2/10ge2p1x520-eth-l2xcbase-ndrpdrdisc.robot b/tests/vpp/perf/l2/10ge2p1x520-eth-l2xcbase-ndrpdrdisc.robot
new file mode 100644
index 0000000000..4c0e0a18b2
--- /dev/null
+++ b/tests/vpp/perf/l2/10ge2p1x520-eth-l2xcbase-ndrpdrdisc.robot
@@ -0,0 +1,483 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.InterfaceUtil
+| Library | resources.libraries.python.NodePath
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | L2XCFWD | BASE | L2XCBASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2XC test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 cross connect.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-eth-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-eth-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-9000B-1t1c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-9000B-1t1c-eth-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-eth-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-eth-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-9000B-2t2c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-9000B-2t2c-eth-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-eth-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-eth-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-9000B-4t4c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-9000B-4t4c-eth-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc19-IMIX-1t1c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame size
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc20-IMIX-1t1c-eth-l2xcbase-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame size
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/l2/10ge2p1x520-eth-l2xcbase-pdrchk.robot b/tests/vpp/perf/l2/10ge2p1x520-eth-l2xcbase-pdrchk.robot
new file mode 100644
index 0000000000..72589f6ca1
--- /dev/null
+++ b/tests/vpp/perf/l2/10ge2p1x520-eth-l2xcbase-pdrchk.robot
@@ -0,0 +1,184 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.InterfaceUtil
+| Library | resources.libraries.python.NodePath
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | L2XCFWD | BASE | L2XCBASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test
+| ...
+| Documentation | *Reference PDR throughput L2XC verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 cross connect.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. DUT1 and DUT2 tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-PDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-PDR value is periodically updated acording to
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 254 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Keywords ***
+| Check PDR for L2 xconnect
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with ${wt} thread, ${wt} phy core,\
+| | ... | ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect
+| | framesize=${64} | rate=5.9mpps | wt=1 | rxq=1
+
+| tc02-1518B-1t1c-eth-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect
+| | framesize=${1518} | rate=812743pps | wt=1 | rxq=1
+
+| tc03-9000B-1t1c-eth-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect
+| | framesize=${9000} | rate=138580pps | wt=1 | rxq=1
+
+| tc04-64B-2t2c-eth-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect
+| | framesize=${64} | rate=10.4mpps | wt=2 | rxq=1
+
+| tc05-1518B-2t2c-eth-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect
+| | framesize=${1518} | rate=812743pps | wt=2 | rxq=1
+
+| tc06-9000B-2t2c-eth-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect
+| | framesize=${9000} | rate=100000pps | wt=2 | rxq=1
+
+| tc07-64B-4t4c-eth-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect
+| | framesize=${64} | rate=10.4mpps | wt=4 | rxq=2
+
+| tc08-1518B-4t4c-eth-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect
+| | framesize=${1518} | rate=812743pps | wt=4 | rxq=2
+
+| tc09-9000B-4t4c-eth-l2xcbase-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-NDR for 9000 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 9000B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 xconnect
+| | framesize=${9000} | rate=138580pps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/l2/10ge2p1x710-eth-l2bdbasemaclrn-ndrpdrdisc.robot b/tests/vpp/perf/l2/10ge2p1x710-eth-l2bdbasemaclrn-ndrpdrdisc.robot
new file mode 100644
index 0000000000..166e1c7a4f
--- /dev/null
+++ b/tests/vpp/perf/l2/10ge2p1x710-eth-l2bdbasemaclrn-ndrpdrdisc.robot
@@ -0,0 +1,274 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X710 | ETH | L2BDMACLRN | BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X710
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2BD test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology\
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-\
+| ... | domain and MAC learning enabled. DUT1 and DUT2 tested with 2p10GE NIC
+| ... | X710 by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop\
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\
+| ... | of packets transmitted. NDR and PDR are discovered for different\
+| ... | Ethernet L2 frame sizes using either binary search or linear search\
+| ... | algorithms with configured starting rate and final step that determines\
+| ... | throughput measurement resolution. Test packets are generated by TG on\
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups\
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets\
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static\
+| ... | payload. MAC addresses are matching MAC addresses of the TG node\
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X710 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Keywords ***
+| L2 Bridge Domain NDR Binary Search
+| | [Arguments] | ${framesize} | ${min_rate} | ${wt} | ${rxq}
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${min_rate}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | Add PCI devices to DUTs in 3-node single link topology
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | Run Keyword If | ${get_framesize} < ${1522} | Add no multi seg to all DUTs
+| | Apply startup configuration on all VPP DUTs
+| | Initialize L2 bridge domain in 3-node circular topology
+| | Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| L2 Bridge Domain PDR Binary Search
+| | [Arguments] | ${framesize} | ${min_rate} | ${wt} | ${rxq}
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${min_rate}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | Add PCI devices to DUTs in 3-node single link topology
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | Run Keyword If | ${get_framesize} < ${1522} | Add no multi seg to all DUTs
+| | Apply startup configuration on all VPP DUTs
+| | Initialize L2 bridge domain in 3-node circular topology
+| | Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=1 | rxq=1
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc02-64B-1t1c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=1 | rxq=1
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc03-1518B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=1 | rxq=1
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc04-1518B-1t1c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=1 | rxq=1
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc05-9000B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=1 | rxq=1
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc06-9000B-1t1c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=1 | rxq=1
+| | [Tags] | 9000B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc07-64B-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=2 | rxq=1
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc08-64B-2t2c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=2 | rxq=1
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc09-1518B-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=2 | rxq=1
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc10-1518B-2t2c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=2 | rxq=1
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc11-9000B-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=2 | rxq=1
+| | [Tags] | 9000B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc12-9000B-2t2c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=2 | rxq=1
+| | [Tags] | 9000B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc13-64B-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=4 | rxq=2
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc14-64B-4t4c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=4 | rxq=2
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 100kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc15-1518B-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=4 | rxq=2
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc16-1518B-4t4c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=4 | rxq=2
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
+
+| tc17-9000B-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=4 | rxq=2
+| | [Tags] | 9000B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc18-9000B-4t4c-eth-l2bdbasemaclrn-pdrdisc
+| | ... | framesize=${9000} | min_rate=${10000} | wt=4 | rxq=2
+| | [Tags] | 9000B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 9000 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | [Template] | L2 Bridge Domain PDR Binary Search
diff --git a/tests/vpp/perf/l2/40ge2p1vic1385-eth-l2bdbasemaclrn-ndrpdrdisc.robot b/tests/vpp/perf/l2/40ge2p1vic1385-eth-l2bdbasemaclrn-ndrpdrdisc.robot
new file mode 100644
index 0000000000..6d39fc169a
--- /dev/null
+++ b/tests/vpp/perf/l2/40ge2p1vic1385-eth-l2bdbasemaclrn-ndrpdrdisc.robot
@@ -0,0 +1,352 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.NodePath
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Cisco-VIC-1385 | ETH | L2BDMACLRN | BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Cisco-VIC-1385
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2BD test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-
+| ... | domain and MAC learning enabled. DUT1 and DUT2 tested with 2p40GE NIC
+| ... | NIC_Cisco-VIC-1385.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# XL710-DA2 (TG) bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps (TG) limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at\
+| | ... | 18.75Mpps rate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-eth-l2bdbasemaclrn-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at\
+| | ... | 18.75Mpps rate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at\
+| | ... | 24.5G rate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc05-9000B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at\
+| | ... | 24.5G rate, step 10kpps.
+| | ...
+| | [Tags] | 9000B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${9000}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc07-64B-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at\
+| | ... | 18.75Mpps rate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-eth-l2bdbasemaclrn-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at\
+| | ... | 18.75Mpps rate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDR | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at\
+| | ... | 24.5G rate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc13-64B-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at\
+| | ... | 18.75Mpps rate, step 100kpps.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc15-1518B-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at\
+| | ... | 24.5G rate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc19-IMIX-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frame size using binary search start at\
+| | ... | 24.5G rate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc20-IMIX-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 thread, 2 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frame size using binary search start at\
+| | ... | 24.5G rate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc21-IMIX-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 thread, 4 phy core,\
+| | ... | 2 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frame size using binary search start at\
+| | ... | 24.5G rate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 bridge domain in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
diff --git a/tests/vpp/perf/l2/40ge2p1xl710-eth-l2bdbasemaclrn-ndrpdrdisc.robot b/tests/vpp/perf/l2/40ge2p1xl710-eth-l2bdbasemaclrn-ndrpdrdisc.robot
new file mode 100644
index 0000000000..595bc029ce
--- /dev/null
+++ b/tests/vpp/perf/l2/40ge2p1xl710-eth-l2bdbasemaclrn-ndrpdrdisc.robot
@@ -0,0 +1,232 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.NodePath
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-XL710 | ETH | L2BDMACLRN | BASE
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-XL710
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2BD test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology\
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-\
+| ... | domain and MAC learning enabled. DUT1 and DUT2 tested with 2p40GE NIC\
+| ... | XL710-DA2 by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop\
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop\
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage\
+| ... | of packets transmitted. NDR and PDR are discovered for different\
+| ... | Ethernet L2 frame sizes using either binary search or linear search\
+| ... | algorithms with configured starting rate and final step that determines\
+| ... | throughput measurement resolution. Test packets are generated by TG on\
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups\
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets\
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static\
+| ... | payload. MAC addresses are matching MAC addresses of the TG node\
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Keywords ***
+| L2 Bridge Domain NDR Binary Search BW limit
+| | [Arguments] | ${framesize} | ${min_rate} | ${wt} | ${rxq} | ${s_limit}
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${min_rate}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | Add PCI devices to DUTs in 3-node single link topology
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | Run Keyword If | ${get_framesize} < ${1522} | Add no multi seg to all DUTs
+| | Apply startup configuration on all VPP DUTs
+| | Initialize L2 bridge domain in 3-node circular topology
+| | Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| L2 Bridge Domain NDR Binary Search
+| | [Arguments] | ${framesize} | ${min_rate} | ${wt} | ${rxq} | ${s_limit}
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${min_rate}
+| | ${max_rate}= | Set Variable | ${s_limit}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | Add PCI devices to DUTs in 3-node single link topology
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | Run Keyword If | ${get_framesize} < ${1522} | Add no multi seg to all DUTs
+| | Apply startup configuration on all VPP DUTs
+| | Initialize L2 bridge domain in 3-node circular topology
+| | Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| L2 Bridge Domain PDR Binary Search BW limit
+| | [Arguments] | ${framesize} | ${min_rate} | ${wt} | ${rxq} | ${s_limit}
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${min_rate}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | Add PCI devices to DUTs in 3-node single link topology
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | Run Keyword If | ${get_framesize} < ${1522} | Add no multi seg to all DUTs
+| | Apply startup configuration on all VPP DUTs
+| | Initialize L2 bridge domain in 3-node circular topology
+| | Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| L2 Bridge Domain PDR Binary Search
+| | [Arguments] | ${framesize} | ${min_rate} | ${wt} | ${rxq} | ${s_limit}
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${min_rate}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | Add PCI devices to DUTs in 3-node single link topology
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | Run Keyword If | ${get_framesize} < ${1522} | Add no multi seg to all DUTs
+| | Apply startup configuration on all VPP DUTs
+| | Initialize L2 bridge domain in 3-node circular topology
+| | Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=1 | rxq=1
+| | ... | s_limit=${s_18.75Mpps}
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at\
+| | ... | 18.75Mpps rate, step 100kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc03-1518B-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=1 | rxq=1
+| | ... | s_limit=${s_24.5G}
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at\
+| | ... | 24.5G rate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search BW limit
+
+| tc07-64B-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=2 | rxq=1
+| | ... | s_limit=${s_18.75Mpps}
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at\
+| | ... | 18.75Mpps rate, step 100kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc09-1518B-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=2 | rxq=1
+| | ... | s_limit=${s_24.5G}
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at\
+| | ... | 24.5G rate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search BW limit
+
+| tc13-64B-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${64} | min_rate=${100000} | wt=4 | rxq=2
+| | ... | s_limit=${s_18.75Mpps}
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at\
+| | ... | 18.75Mpps rate, step 100kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search
+
+| tc15-1518B-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=${1518} | min_rate=${10000} | wt=4 | rxq=2
+| | ... | s_limit=${s_24.5G}
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at\
+| | ... | 24.5G rate, step 10kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search BW limit
+
+| tc19-IMIX-1t1c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=IMIX_v4_1 | min_rate=${100000} | wt=1 | rxq=1
+| | ... | s_limit=${s_24.5G}
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 1 thread, 1 phy core, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frames using binary search start at\
+| | ... | 24.5G rate, step 100kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search BW limit
+
+| tc20-IMIX-2t2c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=IMIX_v4_1 | min_rate=${100000} | wt=2 | rxq=1
+| | ... | s_limit=${s_24.5G}
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 2 threads, 2 phy cores, 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frames using binary search start at\
+| | ... | 24.5G rate, step 100kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search BW limit
+
+| tc21-IMIX-4t4c-eth-l2bdbasemaclrn-ndrdisc
+| | ... | framesize=IMIX_v4_1 | min_rate=${100000} | wt=4 | rxq=2
+| | ... | s_limit=${s_24.5G}
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with with\
+| | ... | 4 threads, 4 phy cores, 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frames using binary search start at\
+| | ... | 24.5G rate, step 100kpps.
+| | [Template] | L2 Bridge Domain NDR Binary Search BW limit
diff --git a/tests/vpp/perf/l2/40ge2p1xl710-eth-l2xcbase-ndrpdrdisc.robot b/tests/vpp/perf/l2/40ge2p1xl710-eth-l2xcbase-ndrpdrdisc.robot
new file mode 100644
index 0000000000..4c6f279e8d
--- /dev/null
+++ b/tests/vpp/perf/l2/40ge2p1xl710-eth-l2xcbase-ndrpdrdisc.robot
@@ -0,0 +1,251 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.InterfaceUtil
+| Library | resources.libraries.python.NodePath
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-XL710 | ETH | L2XCFWD | BASE
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-XL710
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| ...
+| Test Teardown | Tear down performance discovery test | ${min_rate}pps
+| ... | ${framesize} | ${traffic_profile}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2XC test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 cross connect.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. DUT1 and DUT2 tested with 2p40GE NIC XL710 by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 18.75Mpps rate, step 100kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc03-1518B-1t1c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 24.5G rate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc07-64B-2t2c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 18.75Mpps rate, step 100kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc09-1518B-2t2c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 24.5G rate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc13-64B-4t4c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 18.75Mpps rate, step 100kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc15-1518B-4t4c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 24.5G rate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc19-IMIX-1t1c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame size
+| | ... | using binary search start at 24.5G rate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc20-IMIX-2t2c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 thread, 2 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame size
+| | ... | using binary search start at 24.5G rate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc21-IMIX-4t4c-eth-l2xcbase-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 thread, 4 phy core, \
+| | ... | 2 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame size
+| | ... | using binary search start at 24.5G rate, step 100kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${100000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Initialize L2 xconnect in 3-node circular topology
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..9b88f7f2d1
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot
@@ -0,0 +1,705 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | DOT1Q | L2BDMACLRN | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Packet throughput L2BD test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. 802.1q
+| ... | tagging is applied on link between DUT1 and DUT2.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-
+| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via
+| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting
+| ... | vhost-user interfaces using 5 cores pinned to cpus 5-9 and 2048M
+| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores
+| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to
+| ... | io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with 2p10GE NIC X520
+| ... | Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${subid}= | 10
+| ${tag_rewrite}= | pop-1
+| ${vlan_overhead}= | ${4}
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Socket names
+| ${bd_id1}= | 1
+| ${bd_id2}= | 2
+| ${sock1}= | /tmp/sock-1-${bd_id1}
+| ${sock2}= | /tmp/sock-1-${bd_id2}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 framesize using binary search start at\
+| | ... | 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${357.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 framesize using binary search start at\
+| | ... | 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${357.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 2T2C | STHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 2T2C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 framesize using binary search start at\
+| | ... | 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | STHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${357.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 framesize using binary search start at\
+| | ... | 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${357.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 4T4C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 4T4C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 4T4C | STHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 4T4C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 framesize using binary search start at\
+| | ... | 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | STHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${357.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 framesize using binary search start at\
+| | ... | 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${357.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrchk.robot b/tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrchk.robot
new file mode 100644
index 0000000000..bab866795a
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrchk.robot
@@ -0,0 +1,212 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | DOT1Q | L2BDMACLRN | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test with vhost and VM with dpdk-testpmd
+| ... | ${rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Packet throughput L2BD test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. 802.1q
+| ... | tagging is applied on link between DUT1 and DUT2.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-
+| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via
+| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting
+| ... | vhost-user interfaces using 5 cores pinned to cpus 5-9 and 2048M
+| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores
+| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to
+| ... | io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with 2p10GE NIC X520
+| ... | Niantic by Intel.
+| ... | *[Ver] TG verification:* TG verifies throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. Ref-PDR value is periodically updated acording to
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${subid}= | 10
+| ${tag_rewrite}= | pop-1
+# Socket names
+| ${bd_id1}= | 1
+| ${bd_id2}= | 2
+| ${sock1}= | /tmp/sock-1-${bd_id1}
+| ${sock2}= | /tmp/sock-1-${bd_id2}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Keywords ***
+| Check PDR for L2 bridge domain with vhost and VM with dpdk-testpmd
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with ${wt} thread, ${wt} phy\
+| | ... | core, ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VLAN in a 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2} | ${subid}
+| | ... | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain with vhost and VM with dpdk-testpmd
+| | framesize=${64} | rate=1.2mpps | wt=1 | rxq=1
+
+| tc02-1518B-1t1c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain with vhost and VM with dpdk-testpmd
+| | framesize=${1518} | rate=350000pps | wt=1 | rxq=1
+
+| tc03-IMIX-1t1c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain with vhost and VM with dpdk-testpmd
+| | framesize=IMIX_v4_1 | rate=0.9mpps | wt=1 | rxq=1
+
+| tc04-64B-2t2c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain with vhost and VM with dpdk-testpmd
+| | framesize=${64} | rate=2.2mpps | wt=2 | rxq=1
+
+| tc05-1518B-2t2c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain with vhost and VM with dpdk-testpmd
+| | framesize=${1518} | rate=0.6mpps | wt=2 | rxq=1
+
+| tc06-IMIX-2t2c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain with vhost and VM with dpdk-testpmd
+| | framesize=IMIX_v4_1 | rate=1.5pps | wt=2 | rxq=1
+
+| tc07-64B-4t4c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy\
+| | ... | cores, 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain with vhost and VM with dpdk-testpmd
+| | framesize=${64} | rate=3.4mpps | wt=4 | rxq=2
+
+| tc08-1518B-4t4c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy\
+| | ... | cores, 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain with vhost and VM with dpdk-testpmd
+| | framesize=${1518} | rate=0.7mpps | wt=4 | rxq=2
+
+| tc09-IMIX-4t4c-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy\
+| | ... | cores, 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2 bridge domain with vhost and VM with dpdk-testpmd
+| | framesize=IMIX_v4_1 | rate=2.2mpps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2xcbase-eth-2vhost-1vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2xcbase-eth-2vhost-1vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..258ebbbe57
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2xcbase-eth-2vhost-1vm-ndrpdrdisc.robot
@@ -0,0 +1,684 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.NodePath
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | DOT1Q | L2XCFWD | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2XC test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 cross connect. 802.1q
+| ... | tagging is applied on link between DUT1 and DUT2.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. Qemu Guest is connected to VPP via vhost-user interfaces.
+| ... | Guest is running DPDK testpmd interconnecting vhost-user interfaces
+| ... | using 5 cores pinned to cpus 5-9 and 2048M memory. Testpmd is using
+| ... | socket-mem=1024M (512x2M hugepages), 5 cores (1 main core and 4 cores
+| ... | dedicated for io), forwarding mode is set to io, rxd/txd=256,
+| ... | burst=64. DUT1, DUT2 are tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${subid}= | 10
+| ${tag_rewrite}= | pop-1
+| ${vlan_overhead}= | ${4}
+# Socket names
+| ${sock1}= | /tmp/sock-1
+| ${sock2}= | /tmp/sock-2
+# X520-DA2 bandwidth limit
+| ${s_limit}= | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 framesize using binary search start at\
+| | ... | 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${357.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 framesize using binary search start at\
+| | ... | 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${357.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 2T2C | STHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 2T2C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 framesize using binary search start at\
+| | ... | 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | STHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${357.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 framesize using binary search start at\
+| | ... | 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${357.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 4T4C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 4T4C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 4T4C | STHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 4T4C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 framesize using binary search start at\
+| | ... | 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | STHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${357.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 framesize using binary search start at\
+| | ... | 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${357.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2xcbase-eth-2vhost-1vm-pdrchk.robot b/tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2xcbase-eth-2vhost-1vm-pdrchk.robot
new file mode 100644
index 0000000000..68b11c8348
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x520-dot1q-l2xcbase-eth-2vhost-1vm-pdrchk.robot
@@ -0,0 +1,209 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | DOT1Q | L2XCFWD | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test with vhost and VM with dpdk-testpmd
+| ... | ${rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Packet throughput L2XC test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. 802.1q
+| ... | tagging is applied on link between DUT1 and DUT2.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. Qemu Guest is connected to VPP via
+| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting
+| ... | vhost-user interfaces using 5 cores pinned to cpus 5-9 and 2048M
+| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores
+| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to
+| ... | io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with 2p10GE NIC X520
+| ... | Niantic by Intel.
+| ... | *[Ver] TG verification:* TG verifies throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. Ref-PDR value is periodically updated acording to
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${subid}= | 10
+| ${tag_rewrite}= | pop-1
+# Socket names
+| ${sock1}= | /tmp/sock-1
+| ${sock2}= | /tmp/sock-2
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Keywords ***
+| Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with ${wt} thread, ${wt} phy\
+| | ... | core, ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User and VLAN in 3-node circular topology
+| | ... | ${sock1} | ${sock2} | ${subid} | ${tag_rewrite}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-dot1q-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=${64} | rate=1.4mpps | wt=1 | rxq=1
+
+| tc02-1518B-1t1c-dot1q-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=${1518} | rate=330000pps | wt=1 | rxq=1
+
+| tc03-IMIX-1t1c-dot1q-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=IMIX_v4_1 | rate=1.0mpps | wt=1 | rxq=1
+
+| tc04-64B-2t2c-dot1q-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=${64} | rate=1.4mpps | wt=2 | rxq=1
+
+| tc05-1518B-2t2c-dot1q-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=${1518} | rate=0.7mpps | wt=2 | rxq=1
+
+| tc06-IMIX-2t2c-dot1q-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=IMIX_v4_1 | rate=1.6mpps | wt=2 | rxq=1
+
+| tc07-64B-4t4c-dot1q-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy\
+| | ... | cores, 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=${64} | rate=4.0mpps | wt=4 | rxq=2
+
+| tc08-1518B-4t4c-dot1q-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy\
+| | ... | cores, 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=${1518} | rate=0.7mpps | wt=4 | rxq=2
+
+| tc09-IMIX-4t4c-dot1q-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy\
+| | ... | cores, 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=IMIX_v4_1 | rate=2.4mpps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..6bb145325a
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot
@@ -0,0 +1,623 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | L2BDMACLRN | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2BD test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-
+| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via
+| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting
+| ... | vhost-user interfaces using 5 cores pinned to cpus 5-9 and 2048M
+| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores
+| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to
+| ... | io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with 2p10GE NIC X520
+| ... | Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Socket names
+| ${bd_id1}= | 1
+| ${bd_id2}= | 2
+| ${sock1}= | /tmp/sock-1-${bd_id1}
+| ${sock2}= | /tmp/sock-1-${bd_id2}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 5kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..ade616825a
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrpdrdisc.robot
@@ -0,0 +1,548 @@
+# Copyright (c) 2016 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | L2BDMACLRN | BASE | VHOST | VM
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| Test Setup | Set up performance test
+| Test Teardown | Run Keywords
+| ... | Show Bridge Domain Data On All DUTs
+| ... | AND | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| Documentation | *RFC2544: Pkt throughput L2BD test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-
+| ... | domains and MAC learning enabled. Qemu Guests are connected to VPP via
+| ... | vhost-user interfaces. Guests are running DPDK testpmd interconnecting
+| ... | vhost-user interfaces using 5 cores pinned to cpus 6-10 and 11-15 and
+| ... | 2048M memory. Testpmd is using socket-mem=1024M (512x2M hugepages),
+| ... | 5 cores (1 main core and 4 cores dedicated for io), forwarding mode is
+| ... | set to io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with 2p10GE NIC
+| ... | X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit}= | ${10000000000}
+# CPU settings
+| ${system_cpus}= | ${1}
+| ${vpp_cpus}= | ${5}
+| ${vm_cpus}= | ${5}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2xcbase-eth-2vhost-1vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2xcbase-eth-2vhost-1vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..d231afffb7
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2xcbase-eth-2vhost-1vm-ndrpdrdisc.robot
@@ -0,0 +1,621 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.NodePath
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | L2XCFWD | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2XC test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 cross connect.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. Qemu Guest is connected to VPP via vhost-user interfaces.
+| ... | Guest is running DPDK testpmd interconnecting vhost-user interfaces
+| ... | using 5 cores pinned to cpus 5-9 and 2048M memory. Testpmd is using
+| ... | socket-mem=1024M (512x2M hugepages), 5 cores (1 main core and 4 cores
+| ... | dedicated for io), forwarding mode is set to io, rxd/txd=256,
+| ... | burst=64. DUT1, DUT2 are tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${bd_id1}= | 1
+| ${bd_id2}= | 2
+| ${sock1}= | /tmp/sock-1-${bd_id1}
+| ${sock2}= | /tmp/sock-1-${bd_id2}
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-eth-l2xcbase-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-eth-l2xcbase-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2xcbase-eth-2vhost-1vm-pdrchk.robot b/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2xcbase-eth-2vhost-1vm-pdrchk.robot
new file mode 100644
index 0000000000..4fb03188bb
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2xcbase-eth-2vhost-1vm-pdrchk.robot
@@ -0,0 +1,208 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | L2XCFWD | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test with vhost and VM with dpdk-testpmd
+| ... | ${rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Packet throughput L2XC test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4. 802.1q
+| ... | tagging is applied on link between DUT1 and DUT2.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connect. Qemu Guest is connected to VPP via
+| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting
+| ... | vhost-user interfaces using 5 cores pinned to cpus 5-9 and 2048M
+| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores
+| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to
+| ... | io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with 2p10GE NIC X520
+| ... | Niantic by Intel.
+| ... | *[Ver] TG verification:* TG verifies throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. Ref-PDR value is periodically updated acording to
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${bd_id1}= | 1
+| ${bd_id2}= | 2
+| ${sock1}= | /tmp/sock-1-${bd_id1}
+| ${sock2}= | /tmp/sock-1-${bd_id2}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Keywords ***
+| Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with ${wt} thread, ${wt} phy\
+| | ... | core, ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=${64} | rate=1.6mpps | wt=1 | rxq=1
+
+| tc02-1518B-1t1c-eth-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=${1518} | rate=400000pps | wt=1 | rxq=1
+
+| tc03-IMIX-1t1c-eth-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=IMIX_v4_1 | rate=1.0mpps | wt=1 | rxq=1
+
+| tc04-64B-2t2c-eth-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=${64} | rate=3.0mpps | wt=2 | rxq=1
+
+| tc05-1518B-2t2c-eth-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=${1518} | rate=0.7mpps | wt=2 | rxq=1
+
+| tc06-IMIX-2t2c-eth-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=IMIX_v4_1 | rate=2.0mpps | wt=2 | rxq=1
+
+| tc07-64B-4t4c-eth-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy\
+| | ... | cores, 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=${64} | rate=4.2mpps | wt=4 | rxq=2
+
+| tc08-1518B-4t4c-eth-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy\
+| | ... | cores, 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=${1518} | rate=0.7mpps | wt=4 | rxq=2
+
+| tc09-IMIX-4t4c-eth-l2xcbase-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy\
+| | ... | cores, 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for L2XC with vhost and VM with dpdk-testpmd
+| | framesize=IMIX_v4_1 | rate=3.0mpps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2xcbase-eth-4vhost-2vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2xcbase-eth-4vhost-2vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..ed2635ae02
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x520-eth-l2xcbase-eth-4vhost-2vm-ndrpdrdisc.robot
@@ -0,0 +1,549 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.NodePath
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | L2XCFWD | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2XC test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 cross connect.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connects. Qemu Guests are connected to VPP via vhost-user interfaces.
+| ... | Guests are running DPDK testpmd interconnecting vhost-user interfaces
+| ... | using 5 cores pinned to cpus 6-10 and 11-15 and 2048M memory. Testpmd is
+| ... | using socket-mem=1024M (512x2M hugepages), 5 cores (1 main core and 4
+| ... | cores dedicated for io), forwarding mode is set to io, rxd/txd=256,
+| ... | burst=64. DUT1, DUT2 are tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+#CPU settings
+| ${system_cpus}= | ${1}
+| ${vpp_cpus}= | ${5}
+| ${vm_cpus}= | ${5}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4-ip4base-eth-2vhost-1vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4-ip4base-eth-2vhost-1vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..0b19414fb5
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4-ip4base-eth-2vhost-1vm-ndrpdrdisc.robot
@@ -0,0 +1,658 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv4 test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 route entries. Qemu Guest is connected
+| ... | to VPP via vhost-user interfaces. Guest is running DPDK testpmd
+| ... | interconnecting vhost-user interfaces using 5 cores pinned to cpus 5-9
+| ... | and 2048M memory. Testpmd is using socket-mem=1024M (512x2M hugepages),
+| ... | 5 cores (1 main core and 4 cores dedicated for io), forwarding mode is
+| ... | set to mac, rxd/txd=256, burst=64. DUT1, DUT2 are tested with 2p10GE
+| ... | NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+| ${sock1}= | /tmp/sock-1
+| ${sock2}= | /tmp/sock-2
+| ${fib_table_1}= | 100
+| ${fib_table_2}= | 101
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4base-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4-ip4base-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4-ip4base-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4-ip4base-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4-ip4base-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4-ip4base-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4-ip4base-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4-ip4base-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip4-ip4base-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip4-ip4base-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-ethip4-ip4base-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-ethip4-ip4base-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4-ip4base-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4-ip4base-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip4-ip4base-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip4-ip4base-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-ethip4-ip4base-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-ethip4-ip4base-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4-ip4base-eth-2vhost-1vm-pdrchk.robot b/tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4-ip4base-eth-2vhost-1vm-pdrchk.robot
new file mode 100644
index 0000000000..46a44cb58b
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4-ip4base-eth-2vhost-1vm-pdrchk.robot
@@ -0,0 +1,209 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PDRCHK
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance pdrchk test with vhost and VM with dpdk-testpmd
+| ... | ${rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv4 test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and two static IPv4 /24 route entries. Qemu Guest is connected
+| ... | to VPP via vhost-user interfaces. Guest is running DPDK testpmd
+| ... | interconnecting vhost-user interfaces using 5 cores pinned to cpus 5-9
+| ... | and 2048M memory. Testpmd is using socket-mem=1024M (512x2M hugepages),
+| ... | 5 cores (1 main core and 4 cores dedicated for io), forwarding mode is
+| ... | set to mac, rxd/txd=256, burst=64. DUT1, DUT2 are tested with 2p10GE
+| ... | NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG verifies throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. Ref-PDR value is periodically updated acording to
+| ... | formula: ref-PDR = 0.9x PDR, where PDR is found in RFC2544 long
+| ... | performance tests for the same DUT configuration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 253 flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and static payload. MAC addresses are matching MAC addresses of the
+| ... | TG node interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${sock1}= | /tmp/sock-1
+| ${sock2}= | /tmp/sock-2
+| ${fib_table_1}= | 100
+| ${fib_table_2}= | 101
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Keywords ***
+| Check PDR for IPv4 routing with vhost and VM with dpdk-testpmd
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with ${wt} thread, ${wt} phy\
+| | ... | core, ${rxq} receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for ${framesize} Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Arguments] | ${framesize} | ${rate} | ${wt} | ${rxq}
+| | ...
+| | # Test Variables required for test and test teardown
+| | Set Test Variable | ${framesize}
+| | Set Test Variable | ${rate}
+| | ${get_framesize}= | Get Frame Size | ${framesize}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | ...
+| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Run Keyword If | ${get_framesize} < ${1522}
+| | ... | Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost in 3-node circular topology
+| | ... | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1 | ${dut1_vif1_mac}
+| | ... | ${dut1_vif2_mac}
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd-mac connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1 | ${dut2_vif1_mac}
+| | ... | ${dut2_vif2_mac}
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Traffic should pass with partial loss | ${perf_trial_duration}
+| | ... | ${rate} | ${framesize} | ${traffic_profile}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-ip4base-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing with vhost and VM with dpdk-testpmd
+| | framesize=${64} | rate=2.5mpps | wt=1 | rxq=1
+
+| tc02-1518B-1t1c-eth-ip4base-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing with vhost and VM with dpdk-testpmd
+| | framesize=${1518} | rate=300000pps | wt=1 | rxq=1
+
+| tc03-IMIX-1t1c-eth-ip4base-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy\
+| | ... | core, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing with vhost and VM with dpdk-testpmd
+| | framesize=IMIX_v4_1 | rate=0.9mpps | wt=1 | rxq=1
+
+| tc04-64B-2t2c-eth-ip4base-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing with vhost and VM with dpdk-testpmd
+| | framesize=${64} | rate=3.0mpps | wt=2 | rxq=1
+
+| tc05-1518B-2t2c-eth-ip4base-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing with vhost and VM with dpdk-testpmd
+| | framesize=${1518} | rate=350000pps | wt=2 | rxq=1
+
+| tc06-IMIX-2t2c-eth-ip4base-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy\
+| | ... | cores, 1 receive queue per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing with vhost and VM with dpdk-testpmd
+| | framesize=IMIX_v4_1 | rate=2.0mpps | wt=2 | rxq=1
+
+| tc07-64B-4t4c-eth-ip4base-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy\
+| | ... | cores, 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing with vhost and VM with dpdk-testpmd
+| | framesize=${64} | rate=3.8mpps | wt=4 | rxq=2
+
+| tc08-1518B-4t4c-eth-ip4base-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy\
+| | ... | cores, 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 1518 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing with vhost and VM with dpdk-testpmd
+| | framesize=${1518} | rate=350000pps | wt=4 | rxq=2
+
+| tc09-IMIX-4t4c-eth-ip4base-eth-2vhost-1vm-pdrchk
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy\
+| | ... | cores, 2 receive queues per NIC port.
+| | ... | [Ver] Verify ref-PDR for 64 Byte frames using single trial\
+| | ... | throughput test at 2x ${rate}.
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD
+| | ...
+| | [Template] | Check PDR for IPv4 routing with vhost and VM with dpdk-testpmd
+| | framesize=IMIX_v4_1 | rate=3.0mpps | wt=4 | rxq=2
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4-ip4base-eth-4vhost-2vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4-ip4base-eth-4vhost-2vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..5e1c9afca7
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4-ip4base-eth-4vhost-2vm-ndrpdrdisc.robot
@@ -0,0 +1,550 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | ETH | IP4FWD | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Pkt throughput IPv4 test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and static IPv4 /24 route entries. Qemu Guests are connected
+| ... | to VPP via vhost-user interfaces. Guests are running DPDK testpmd
+| ... | interconnecting vhost-user interfaces using 5 cores pinned to cpus 6-10
+| ... | and 11-15 and 2048M memory. Testpmd is using socket-mem=1024M (512x2M
+| ... | hugepages), 5 cores (1 main core and 4 cores dedicated for io),
+| ... | forwarding mode is set to mac, rxd/txd=256, burst=64. DUT1, DUT2 are
+| ... | tested with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit}= | ${10000000000}
+# CPU settings
+| ${system_cpus}= | ${1}
+| ${vpp_cpus}= | ${5}
+| ${vm_cpus}= | ${5}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip4-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip4-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-ethip4-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-ethip4-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip4-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip4-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-ethip4-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-ethip4-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | [Tags] | IMIX | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..b24e0ca3c6
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot
@@ -0,0 +1,675 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X520-DA2 | L2BDMACLRN | ENCAP | VXLAN | L2OVRLAY | IP4UNRLAY
+| ... | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X520-DA2
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Packet throughput L2BD test cases with VXLANoIPv4
+| ... | and vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUT1 and DUT2.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-
+| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via
+| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting
+| ... | vhost-user interfaces using 5 cores pinned to cpus 5-9 and 2048M
+| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores
+| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to
+| ... | io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with 2p10GE NIC X520
+| ... | Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544, RFC7348.
+
+*** Variables ***
+# X520-DA2 bandwidth limit
+| ${s_limit} | ${10000000000}
+| ${vxlan_overhead} | ${50}
+# Socket names
+| ${bd_id1}= | 1
+| ${bd_id2}= | 2
+| ${sock1}= | /tmp/sock-1-${bd_id1}
+| ${sock2}= | /tmp/sock-1-${bd_id2}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 framesize using binary search start at\
+| | ... | 10GE linerate, step 5kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${403.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 frame size using binary search start at\
+| | ... | 10GE linerate, step 5kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${403.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frame size using binary search start at\
+| | ... | 10GE linerate, step 5kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${403.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 frame size using binary search start at\
+| | ... | 10GE linerate, step 5kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${403.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 64 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for 1518 Byte frames using binary search start at 10GE\
+| | ... | linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize + ${vxlan_overhead}}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find NDR for IMIX_v4_1 frame size using binary search start at\
+| | ... | 10GE linerate, step 5kpps.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${403.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port.
+| | ... | [Ver] Find PDR for IMIX_v4_1 frame size using binary search start at\
+| | ... | 10GE linerate, step 5kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${avg_framesize}= | Set Variable | ${403.833}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${avg_framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User and VXLANoIPv4 in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..734b624666
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/10ge2p1x710-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot
@@ -0,0 +1,658 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-X710 | ETH | L2BDMACLRN | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-X710
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Pkt throughput L2BD test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-
+| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via
+| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting
+| ... | vhost-user interfaces using 5 cores pinned to cpus on NUMA1 and 2048M
+| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores
+| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to
+| ... | io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with 2p10GE NIC X710
+| ... | by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# X710 bandwidth limit
+| ${s_limit} | ${10000000000}
+# Socket names
+| ${bd_id1}= | 1
+| ${bd_id2}= | 2
+| ${sock1}= | /tmp/sock-1-${bd_id1}
+| ${sock2}= | /tmp/sock-1-${bd_id2}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames\
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames\
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame\
+| | ... | size using binary search start at 10GE linerate, step 5kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame\
+| | ... | size using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames\
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames\
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame\
+| | ... | size using binary search start at 10GE linerate, step 5kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame\
+| | ... | size using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames\
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames\
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for IMIX_v4_1 frame\
+| | ... | size using binary search start at 10GE linerate, step 5kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for IMIX_v4_1 frame\
+| | ... | size using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/vm_vhost/40ge2p1xl710-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/40ge2p1xl710-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..b16a699e6a
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/40ge2p1xl710-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrpdrdisc.robot
@@ -0,0 +1,660 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-XL710 | ETH | L2BDMACLRN | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-XL710
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Packet throughput L2BD test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-
+| ... | domain and MAC learning enabled. Qemu Guest is connected to VPP via
+| ... | vhost-user interfaces. Guest is running DPDK testpmd interconnecting
+| ... | vhost-user interfaces using 5 cores pinned to cpus on NUMA1 and 2048M
+| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores
+| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to
+| ... | io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with 2p40GE NIC XL710
+| ... | by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+# Socket names
+| ${bd_id1}= | 1
+| ${bd_id2}= | 2
+| ${sock1}= | /tmp/sock-1-${bd_id1}
+| ${sock2}= | /tmp/sock-1-${bd_id2}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame\
+| | ... | size using binary search start at 40GE linerate, step 5kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core,\
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame\
+| | ... | size using binary search start at 40GE linerate, step 5kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame\
+| | ... | size using binary search start at 40GE linerate, step 5kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores,\
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame\
+| | ... | size using binary search start at 40GE linerate, step 5kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames\
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for IMIX_v4_1 frame\
+| | ... | size using binary search start at 40GE linerate, step 5kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-eth-l2bdbasemaclrn-eth-2vhost-1vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores,\
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for IMIX_v4_1 frame\
+| | ... | size using binary search start at 40GE linerate, step 5kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User in 3-node circular topology
+| | ... | ${bd_id1} | ${bd_id2} | ${sock1} | ${sock2}
+| | ${vm1}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut1} | ${sock1} | ${sock2} | DUT1_VM1
+| | Set To Dictionary | ${dut1_vm_refs} | DUT1_VM1 | ${vm1}
+| | ${vm2}= | And Configure guest VM with dpdk-testpmd connected via vhost-user
+| | ... | ${dut2} | ${sock1} | ${sock2} | DUT2_VM1
+| | Set To Dictionary | ${dut2_vm_refs} | DUT2_VM1 | ${vm2}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold} | ${perf_pdr_loss_acceptance}
+| | ... | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/vm_vhost/40ge2p1xl710-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/40ge2p1xl710-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..c9fc7292b5
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/40ge2p1xl710-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrpdrdisc.robot
@@ -0,0 +1,588 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-XL710 | ETH | L2BDMACLRN | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-XL710
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Packet throughput L2BD test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 bridge-
+| ... | domain and MAC learning enabled. Qemu Guests are connected to VPP via
+| ... | vhost-user interfaces. Guests are running DPDK testpmd interconnecting
+| ... | vhost-user interfaces using 5 cores pinned to cpus on NUMA1 and 2048M
+| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores
+| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to
+| ... | io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with 2p40GE NIC XL710
+| ... | by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+# CPU settings
+| ${system_cpus}= | ${1}
+| ${vpp_cpus}= | ${5}
+| ${vm_cpus}= | ${5}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-eth-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-eth-l2bdbasemaclrn-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2BD switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 bridge domains with Vhost-User for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/vm_vhost/40ge2p1xl710-eth-l2xcbase-eth-4vhost-2vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/40ge2p1xl710-eth-l2xcbase-eth-4vhost-2vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..7d79bc2e78
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/40ge2p1xl710-eth-l2xcbase-eth-4vhost-2vm-ndrpdrdisc.robot
@@ -0,0 +1,588 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-XL710 | ETH | L2XCFWD | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L2 | Intel-XL710
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Packet throughput L2XC test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for L2 switching of IPv4.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with L2 cross-
+| ... | connects. Qemu Guests are connected to VPP via vhost-user interfaces.
+| ... | Guests are running DPDK testpmd interconnecting vhost-user interfaces
+| ... | using 5 cores pinned to cpus on NUMA1 (cpus 24-28 and 29-34) and 2048M
+| ... | memory. Testpmd is using socket-mem=1024M (512x2M hugepages), 5 cores
+| ... | (1 main core and 4 cores dedicated for io), forwarding mode is set to
+| ... | io, rxd/txd=256, burst=64. DUT1, DUT2 are tested with 2p40GE NIC XL710
+| ... | by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+# CPU settings
+| ${system_cpus}= | ${1}
+| ${vpp_cpus}= | ${5}
+| ${vm_cpus}= | ${5}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src254
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-eth-l2xcbase-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-eth-l2xcbase-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs L2XC switching config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize L2 xconnect with Vhost-User for '2' in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
diff --git a/tests/vpp/perf/vm_vhost/40ge2p1xl710-ethip4-ip4base-eth-4vhost-2vm-ndrpdrdisc.robot b/tests/vpp/perf/vm_vhost/40ge2p1xl710-ethip4-ip4base-eth-4vhost-2vm-ndrpdrdisc.robot
new file mode 100644
index 0000000000..30e68e1f15
--- /dev/null
+++ b/tests/vpp/perf/vm_vhost/40ge2p1xl710-ethip4-ip4base-eth-4vhost-2vm-ndrpdrdisc.robot
@@ -0,0 +1,588 @@
+# Copyright (c) 2017 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance/performance_setup.robot
+| Library | resources.libraries.python.QemuUtils
+| ...
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDRDISC
+| ... | NIC_Intel-XL710 | ETH | IP4FWD | BASE | VHOST | VM
+| ...
+| Suite Setup | Set up 3-node performance topology with DUT's NIC model
+| ... | L3 | Intel-XL710
+| Suite Teardown | Tear down 3-node performance topology
+| ...
+| Test Setup | Set up performance test
+| Test Teardown | Tear down performance test with vhost and VM with dpdk-testpmd
+| ... | ${min_rate}pps | ${framesize} | ${traffic_profile}
+| ... | dut1_node=${dut1} | dut1_vm_refs=${dut1_vm_refs}
+| ... | dut2_node=${dut2} | dut2_vm_refs=${dut2_vm_refs}
+| ...
+| Documentation | *RFC2544: Packet throughput IP14 test cases with vhost*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 forIPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and static IPv4 /24 route entries. Qemu Guests are connected to
+| ... | VPP via vhost-user interfaces. Guests are running DPDK testpmd
+| ... | interconnecting vhost-user interfaces using 5 cores pinned to cpus on
+| ... | NUMA1 (cpus 24-28 and 29-34) and 2048M memory. Testpmd is using
+| ... | socket-mem=1024M (512x2M hugepages), 5 cores (1 main core and 4 cores
+| ... | dedicated for io), forwarding mode is set to io, rxd/txd=256, burst=64.
+| ... | DUT1, DUT2 are tested with 2p40GE NIC XL710 by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 253 flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and static
+| ... | payload. MAC addresses are matching MAC addresses of the TG node
+| ... | interfaces.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+# XL710-DA2 bandwidth limit ~49Gbps/2=24.5Gbps
+| ${s_24.5G} | ${24500000000}
+# XL710-DA2 Mpps limit 37.5Mpps/2=18.75Mpps
+| ${s_18.75Mpps} | ${18750000}
+# CPU settings
+| ${system_cpus}= | ${1}
+| ${vpp_cpus}= | ${5}
+| ${vm_cpus}= | ${5}
+# Traffic profile:
+| ${traffic_profile} | trex-sl-3n-ethip4-ip4src253
+
+*** Test Cases ***
+| tc01-64B-1t1c-eth-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc02-64B-1t1c-eth-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc03-1518B-1t1c-eth-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc04-1518B-1t1c-eth-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc05-IMIX-1t1c-eth-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc06-IMIX-1t1c-eth-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 1T1C | STHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '1' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc07-64B-2t2c-eth-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc08-64B-2t2c-eth-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc09-1518B-2t2c-eth-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc10-1518B-2t2c-eth-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc11-IMIX-2t2c-eth-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc12-IMIX-2t2c-eth-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 2T2C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '2' worker threads and '1' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc13-64B-4t4c-eth-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | NDRDISC
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc14-64B-4t4c-eth-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 64B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${64}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Set Variable | ${s_18.75Mpps}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc15-1518B-4t4c-eth-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc16-1518B-4t4c-eth-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames \
+| | ... | using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ...
+| | [Tags] | 1518B | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | ${1518}
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${framesize}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}
+
+| tc17-IMIX-4t4c-eth-ip4base-eth-4vhost-2vm-ndrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD | NDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find NDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+
+| tc18-IMIX-4t4c-eth-ip4base-eth-4vhost-2vm-pdrdisc
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for IMIX_v4_1 frame \
+| | ... | size using binary search start at 40GE linerate, step 10kpps, LT=0.5%.
+| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B)
+| | ...
+| | [Tags] | IMIX | 4T4C | MTHREAD | PDRDISC | SKIP_PATCH
+| | ...
+| | ${framesize}= | Set Variable | IMIX_v4_1
+| | ${min_rate}= | Set Variable | ${10000}
+| | ${max_rate}= | Calculate pps | ${s_24.5G} | ${353.83333}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | ${dut1_vm_refs}= | Create Dictionary
+| | ${dut2_vm_refs}= | Create Dictionary
+| | Set Test Variable | ${dut1_vm_refs}
+| | Set Test Variable | ${dut2_vm_refs}
+| | Given Add '4' worker threads and '2' rxqueues in 3-node single-link circular topology
+| | And Add PCI devices to DUTs in 3-node single link topology
+| | And Add no multi seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | When Initialize IPv4 forwarding with vhost for '2' VMs in 3-node circular topology
+| | And Configure '2' guest VMs with dpdk-testpmd-mac connected via vhost-user in 3-node circular topology
+| | Then Find PDR using binary search and pps
+| | ... | ${framesize} | ${binary_min} | ${binary_max} | ${traffic_profile}
+| | ... | ${min_rate} | ${max_rate} | ${threshold}
+| | ... | ${perf_pdr_loss_acceptance} | ${perf_pdr_loss_acceptance_type}