diff options
Diffstat (limited to 'tests/vpp/perf')
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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 20.20.20.0 | 24 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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 |Â local +| | And Add fib table | ${dut2} | 2001:2:: | 64 | 1 |Â 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} |