7.1. container_memif¶
7.1.1. 10ge2p1x710-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr suite¶
RFC2544: Pkt throughput L2BD with memif and IEEE 802.1Q test cases [Top] Network Topologies: TG-DUT1-DUT2-TG 3-node circular topologywith single links between nodes. [Enc] Packet Encapsulations: Eth-IPv4 for L2 bridge domain. IEEE802.1Q tagging is applied on link between DUT1 and DUT2. [Cfg] DUT configuration: DUT1 and DUT2 are configured with two L2bridge domains and MAC learning enabled. Container is connected to VPPvia Memif interface. Container is running same VPP version as runningon DUT. Container is limited via cgroup to use 3 cores allocated frompool of isolated CPUs. There are no memory constraints. DUT1 and DUT2tested with ${nic_name}.[Ver] TG verification: TG finds and reports throughput NDR (Non DropRate) with zero packet loss tolerance and throughput PDR (Partial DropRate) with non-zero packet loss tolerance (LT) expressed in percentageof packets transmitted. NDR and PDR are discovered for differentEthernet L2 frame sizes using MLRsearch library.Test packets aregenerated by TG on links to DUTs. TG traffic profile contains two L3flow-groups (flow-group per direction, 254 flows per flow-group) withall packets containing Ethernet header, IPv4 header with IPprotocol=61 and static payload. MAC addresses are matching MACaddresses of the TG node interfaces. [Ref] Applicable standard specifications: RFC2544.
# Copyright (c) 2021 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
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | DOT1Q | L2BDMACLRN | BASE | MEMIF | DOCKER
| ... | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | dot1q-l2bdbasemaclrn-eth-2memif-1dcr
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | container
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | memif_plugin.so
| ${crypto_type}= | ${None}
| ${nic_name}= | Intel-X710
| ${nic_driver}= | vfio-pci
| ${nic_rxq_size}= | 0
| ${nic_txq_size}= | 0
| ${nic_pfs}= | 2
| ${nic_vfs}= | 0
| ${osi_layer}= | L2
| ${overhead}= | ${4}
| ${subid}= | 10
| ${tag_rewrite}= | pop-1
| ${bd_id1}= | 1
| ${bd_id2}= | 2
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-ethip4-ip4src254
# Container
| ${container_engine}= | Docker
| ${container_chain_topology}= | chain7.1.1.1. Local Template¶
[Cfg] Each DUT runs L2BD switching with VLAN and uses ${phy_cores}physical core(s) for worker threads. [Ver] Measure NDR and PDR values using MLRsearch algorithm.Arguments: - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). Type: integer, string - phy_cores - Number of physical cores. Type: integer - rxq - Number of RX queues, default value: ${None}. Type: integer
Set Test Variable \${frame_size}
Given Set Max Rate And Jumbo
And Add worker threads to all DUTs ${phy_cores} ${rxq}
And Pre-initialize layer driver ${nic_driver}
And Apply startup configuration on all VPP DUTs
When Initialize layer driver ${nic_driver}
And Initialize layer interface
And Start containers for test
And Initialize L2 Bridge Domain with memif pairs and VLAN in circular topology ${bd_id1} ${bd_id2} ${subid} ${tag_rewrite}
Then Find NDR and PDR intervals using optimized search
7.1.1.2. 64B-1c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.1.1.3. 64B-2c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.1.1.4. 64B-4c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.1.1.5. 1518B-1c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.1.1.6. 1518B-2c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.1.1.7. 1518B-4c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.1.1.8. 9000B-1c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.1.1.9. 9000B-2c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.1.1.10. 9000B-4c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.1.1.11. IMIX-1c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.1.1.12. IMIX-2c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.1.1.13. IMIX-4c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.1.2. 10ge2p1x710-eth-l2bdbasemaclrn-eth-2memif-1lxc-ndrpdr suite¶
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 bridge domain. [Cfg] DUT configuration: DUT1 and DUT2 are configured with two L2 bridge domains and MAC learning enabled. LXC is connected to VPP via Memif interface. LXC is running same VPP version as running on DUT. LXC is limited via cgroup to use 3 cores allocated from pool of isolated CPUs. There are no memory constraints. DUT1 and DUT2 tested with ${nic_name}.[Ver] TG verification: TG finds and reports throughput NDR (Non DropRate) with zero packet loss tolerance and throughput PDR (Partial DropRate) with non-zero packet loss tolerance (LT) expressed in percentageof packets transmitted. NDR and PDR are discovered for differentEthernet L2 frame sizes using MLRsearch library.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.
# Copyright (c) 2021 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
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | ETH | L2BDMACLRN | BASE | MEMIF | LXC
| ... | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | eth-l2bdbasemaclrn-eth-2memif-1lxc
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | container
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | memif_plugin.so
| ${crypto_type}= | ${None}
| ${nic_name}= | Intel-X710
| ${nic_driver}= | vfio-pci
| ${nic_rxq_size}= | 0
| ${nic_txq_size}= | 0
| ${nic_pfs}= | 2
| ${nic_vfs}= | 0
| ${osi_layer}= | L2
| ${overhead}= | ${0}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-ethip4-ip4src254
# Container
| ${container_engine}= | LXC
| ${container_chain_topology}= | chain7.1.2.1. Local Template¶
[Cfg] DUT runs L2BD switching config. Each DUT uses ${phy_cores} physical core(s) for worker threads. [Ver] Measure NDR and PDR values using MLRsearch algorithm.Arguments: - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). Type: integer, string - phy_cores - Number of physical cores. Type: integer - rxq - Number of RX queues, default value: ${None}. Type: integer
Set Test Variable \${frame_size}
Given Set Max Rate And Jumbo
And Add worker threads to all DUTs ${phy_cores} ${rxq}
And Pre-initialize layer driver ${nic_driver}
And Apply startup configuration on all VPP DUTs
When Initialize layer driver ${nic_driver}
And Initialize layer interface
And Start containers for test
And Initialize L2 Bridge Domain for multiple chains with memif pairs
Then Find NDR and PDR intervals using optimized search
7.1.2.2. 64B-1c-eth-l2bdbasemaclrn-eth-2memif-1lxc-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.1.2.3. 64B-2c-eth-l2bdbasemaclrn-eth-2memif-1lxc-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.1.2.4. 64B-4c-eth-l2bdbasemaclrn-eth-2memif-1lxc-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.1.2.5. 1518B-1c-eth-l2bdbasemaclrn-eth-2memif-1lxc-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.1.2.6. 1518B-2c-eth-l2bdbasemaclrn-eth-2memif-1lxc-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.1.2.7. 1518B-4c-eth-l2bdbasemaclrn-eth-2memif-1lxc-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.1.2.8. 9000B-1c-eth-l2bdbasemaclrn-eth-2memif-1lxc-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.1.2.9. 9000B-2c-eth-l2bdbasemaclrn-eth-2memif-1lxc-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.1.2.10. 9000B-4c-eth-l2bdbasemaclrn-eth-2memif-1lxc-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.1.2.11. IMIX-1c-eth-l2bdbasemaclrn-eth-2memif-1lxc-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.1.2.12. IMIX-2c-eth-l2bdbasemaclrn-eth-2memif-1lxc-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.1.2.13. IMIX-4c-eth-l2bdbasemaclrn-eth-2memif-1lxc-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.1.3. 10ge2p1x710-eth-l2xcbase-eth-2memif-1dcr-ndrpdr suite¶
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 ${nic_name}.Container is connected to VPP via Memif interface. Container is running same VPP version as running on DUT. Container is limited via cgroup to use 3 cores allocated from pool of isolated CPUs. There are no memory contraints. [Ver] TG verification: TG finds and reports throughput NDR (Non DropRate) with zero packet loss tolerance and throughput PDR (Partial DropRate) with non-zero packet loss tolerance (LT) expressed in percentageof packets transmitted. NDR and PDR are discovered for differentEthernet L2 frame sizes using MLRsearch library.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.
# Copyright (c) 2021 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
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | ETH | L2XCFWD | BASE | MEMIF | DOCKER
| ... | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | eth-l2xcbase-eth-2memif-1dcr
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | container
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | memif_plugin.so
| ${crypto_type}= | ${None}
| ${nic_name}= | Intel-X710
| ${nic_driver}= | vfio-pci
| ${nic_rxq_size}= | 0
| ${nic_txq_size}= | 0
| ${nic_pfs}= | 2
| ${nic_vfs}= | 0
| ${osi_layer}= | L2
| ${overhead}= | ${0}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-ethip4-ip4src254
# Container
| ${container_engine}= | Docker
| ${container_chain_topology}= | chain7.1.3.1. Local Template¶
[Cfg] DUT runs L2XC switching config. Each DUT uses ${phy_cores} physical core(s) for worker threads. [Ver] Measure NDR and PDR values using MLRsearch algorithm.Arguments: - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). Type: integer, string - phy_cores - Number of physical cores. Type: integer - rxq - Number of RX queues, default value: ${None}. Type: integer
Set Test Variable \${frame_size}
Given Set Max Rate And Jumbo
And Add worker threads to all DUTs ${phy_cores} ${rxq}
And Pre-initialize layer driver ${nic_driver}
And Apply startup configuration on all VPP DUTs
When Initialize layer driver ${nic_driver}
And Initialize layer interface
And Start containers for test
And Initialize L2 xconnect with memif pairs
Then Find NDR and PDR intervals using optimized search
7.1.3.2. 64B-1c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.1.3.3. 64B-2c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.1.3.4. 64B-4c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.1.3.5. 1518B-1c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.1.3.6. 1518B-2c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.1.3.7. 1518B-4c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.1.3.8. 9000B-1c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.1.3.9. 9000B-2c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.1.3.10. 9000B-4c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.1.3.11. IMIX-1c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.1.3.12. IMIX-2c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.1.3.13. IMIX-4c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.1.4. 10ge2p1x710-eth-l2xcbase-eth-2memif-1lxc-ndrpdr suite¶
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 ${nic_name}.LXC is connected to VPP via Memif interface. LXC is running same VPP version as running on DUT. LXC is limited via cgroup to use 3 cores allocated from pool of isolated CPUs. There are no memory contraints. [Ver] TG verification: TG finds and reports throughput NDR (Non DropRate) with zero packet loss tolerance and throughput PDR (Partial DropRate) with non-zero packet loss tolerance (LT) expressed in percentageof packets transmitted. NDR and PDR are discovered for differentEthernet L2 frame sizes using MLRsearch library.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.
# Copyright (c) 2021 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
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | ETH | L2XCFWD | BASE | MEMIF | LXC
| ... | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | eth-l2xcbase-eth-2memif-1lxc
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | container
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | memif_plugin.so
| ${crypto_type}= | ${None}
| ${nic_name}= | Intel-X710
| ${nic_driver}= | vfio-pci
| ${nic_rxq_size}= | 0
| ${nic_txq_size}= | 0
| ${nic_pfs}= | 2
| ${nic_vfs}= | 0
| ${osi_layer}= | L2
| ${overhead}= | ${0}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-ethip4-ip4src254
# Container
| ${container_engine}= | LXC
| ${container_chain_topology}= | chain7.1.4.1. Local Template¶
[Cfg] DUT runs L2XC switching config. Each DUT uses ${phy_cores} physical core(s) for worker threads. [Ver] Measure NDR and PDR values using MLRsearch algorithm.Arguments: - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). Type: integer, string - phy_cores - Number of physical cores. Type: integer - rxq - Number of RX queues, default value: ${None}. Type: integer
Set Test Variable \${frame_size}
Given Set Max Rate And Jumbo
And Add worker threads to all DUTs ${phy_cores} ${rxq}
And Pre-initialize layer driver ${nic_driver}
And Apply startup configuration on all VPP DUTs
When Initialize layer driver ${nic_driver}
And Initialize layer interface
And Start containers for test
And Initialize L2 xconnect with memif pairs
Then Find NDR and PDR intervals using optimized search
7.1.4.2. 64B-1c-eth-l2xcbase-eth-2memif-1lxc-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.1.4.3. 64B-2c-eth-l2xcbase-eth-2memif-1lxc-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.1.4.4. 64B-4c-eth-l2xcbase-eth-2memif-1lxc-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.1.4.5. 1518B-1c-eth-l2xcbase-eth-2memif-1lxc-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.1.4.6. 1518B-2c-eth-l2xcbase-eth-2memif-1lxc-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.1.4.7. 1518B-4c-eth-l2xcbase-eth-2memif-1lxc-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.1.4.8. 9000B-1c-eth-l2xcbase-eth-2memif-1lxc-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.1.4.9. 9000B-2c-eth-l2xcbase-eth-2memif-1lxc-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.1.4.10. 9000B-4c-eth-l2xcbase-eth-2memif-1lxc-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.1.4.11. IMIX-1c-eth-l2xcbase-eth-2memif-1lxc-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.1.4.12. IMIX-2c-eth-l2xcbase-eth-2memif-1lxc-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.1.4.13. IMIX-4c-eth-l2xcbase-eth-2memif-1lxc-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.1.5. 10ge2p1x710-ethip4-ip4base-eth-2memif-1dcr-ndrpdr suite¶
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 with two FIB tables and two static IPv4 /24 route entries per FIB table. Container is connected to VPP via Memif interface. Container is running same VPP version as running on DUT. Container is limited via cgroup to use cores allocated from pool of isolated CPUs. There are no memory contraints. DUTs are tested with ${nic_name}.[Ver] TG verification: TG finds and reports throughput NDR (Non DropRate) with zero packet loss tolerance and throughput PDR (Partial DropRate) with non-zero packet loss tolerance (LT) expressed in percentageof packets transmitted. NDR and PDR are discovered for differentEthernet L2 frame sizes using MLRsearch library.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.
# Copyright (c) 2021 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
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | ETH | IP4FWD | BASE | MEMIF | DOCKER | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4-ip4base-eth-2memif-1dcr
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | container
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | memif_plugin.so
| ${crypto_type}= | ${None}
| ${nic_name}= | Intel-X710
| ${nic_driver}= | vfio-pci
| ${nic_rxq_size}= | 0
| ${nic_txq_size}= | 0
| ${nic_pfs}= | 2
| ${nic_vfs}= | 0
| ${osi_layer}= | L3
| ${overhead}= | ${0}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-ethip4-ip4src254
# Container
| ${container_engine}= | Docker
| ${container_chain_topology}= | chain7.1.5.1. Local Template¶
[Cfg] DUT runs IPv4 routing config. Each DUT uses ${phy_cores} physical core(s) for worker threads. [Ver] Measure NDR and PDR values using MLRsearch algorithm.Arguments: - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). Type: integer, string - phy_cores - Number of physical cores. Type: integer - rxq - Number of RX queues, default value: ${None}. Type: integer
Set Test Variable \${frame_size}
Given Set Max Rate And Jumbo
And Add worker threads to all DUTs ${phy_cores} ${rxq}
And Pre-initialize layer driver ${nic_driver}
And Apply startup configuration on all VPP DUTs
When Initialize layer driver ${nic_driver}
And Initialize layer interface
And Start containers for test
And Initialize IPv4 routing with memif pairs
Then Find NDR and PDR intervals using optimized search
7.1.5.2. 64B-1c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.1.5.3. 64B-2c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.1.5.4. 64B-4c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.1.5.5. 1518B-1c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.1.5.6. 1518B-2c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.1.5.7. 1518B-4c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.1.5.8. 9000B-1c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.1.5.9. 9000B-2c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.1.5.10. 9000B-4c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.1.5.11. IMIX-1c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.1.5.12. IMIX-2c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.1.5.13. IMIX-4c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.1.6. 2n1l-10ge2p1x710-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr suite¶
RFC2544: Pkt throughput L2BD with memif and IEEE 802.1Q test cases [Top] Network Topologies: TG-DUT1-TG 2-node circular topology withsingle links between nodes. [Enc] Packet Encapsulations: Eth-IPv4 for L2 bridge domain. IEEE802.1Q tagging is applied on link between DUT1-if2 and TG-if2. [Cfg] DUT configuration: DUT1 is configured with two L2 bridgedomains and MAC learning enabled. Container is connected to VPP viaMemif interface. Container is running same VPP version as runningon DUT. Container is limited via cgroup to use 3 cores allocated frompool of isolated CPUs. There are no memory constraints. DUT1 is testedwith ${nic_name}.[Ver] TG verification: TG finds and reports throughput NDR (Non DropRate) with zero packet loss tolerance and throughput PDR (Partial DropRate) with non-zero packet loss tolerance (LT) expressed in percentageof packets transmitted. NDR and PDR are discovered for differentEthernet L2 frame sizes using MLRsearch library.Test packets aregenerated by TG on links to DUT1. TG traffic profile contains two L3flow-groups (flow-group per direction, 254 flows per flow-group) withall packets containing Ethernet header, IPv4 header with IPprotocol=61 and static payload. MAC addresses are matching MACaddresses of the TG node interfaces. [Ref] Applicable standard specifications: RFC2544.
# Copyright (c) 2021 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
|
| Force Tags | 2_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | DOT1Q | L2BDMACLRN | BASE | MEMIF | DOCKER
| ... | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | dot1q-l2bdbasemaclrn-eth-2memif-1dcr
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | container
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | memif_plugin.so
| ${crypto_type}= | ${None}
| ${nic_name}= | Intel-X710
| ${nic_driver}= | vfio-pci
| ${nic_rxq_size}= | 0
| ${nic_txq_size}= | 0
| ${nic_pfs}= | 2
| ${nic_vfs}= | 0
| ${osi_layer}= | L2
| ${overhead}= | ${4}
| ${subid}= | 10
| ${tag_rewrite}= | pop-1
| ${bd_id1}= | 1
| ${bd_id2}= | 2
# Traffic profile:
| ${traffic_profile}= | trex-stl-2n-dot1qip4asym-ip4src254
# Container
| ${container_engine}= | Docker
| ${container_chain_topology}= | chain7.1.6.1. Local Template¶
[Cfg] Each DUT runs L2BD switching with VLAN and uses ${phy_cores}physical core(s) for worker threads. [Ver] Measure NDR and PDR values using MLRsearch algorithm.Arguments: - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). Type: integer, string - phy_cores - Number of physical cores. Type: integer - rxq - Number of RX queues, default value: ${None}. Type: integer
Set Test Variable \${frame_size}
Given Set Max Rate And Jumbo
And Add worker threads to all DUTs ${phy_cores} ${rxq}
And Pre-initialize layer driver ${nic_driver}
And Apply startup configuration on all VPP DUTs
When Initialize layer driver ${nic_driver}
And Initialize layer interface
And Start containers for test
And Initialize L2 Bridge Domain with memif pairs and VLAN in circular topology ${bd_id1} ${bd_id2} ${subid} ${tag_rewrite}
Then Find NDR and PDR intervals using optimized search
7.1.6.2. 64B-1c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.1.6.3. 64B-2c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.1.6.4. 64B-4c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.1.6.5. 1518B-1c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.1.6.6. 1518B-2c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.1.6.7. 1518B-4c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.1.6.8. 9000B-1c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.1.6.9. 9000B-2c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.1.6.10. 9000B-4c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.1.6.11. IMIX-1c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.1.6.12. IMIX-2c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.1.6.13. IMIX-4c-dot1q-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.1.7. 2n1l-10ge2p1x710-eth-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr suite¶
RFC2544: Pkt throughput L2BD test cases [Top] Network Topologies: TG-DUT1-TG 2-node circular topology with single links between nodes. [Enc] Packet Encapsulations: Eth-IPv4 for L2 bridge domain. [Cfg] DUT configuration: DUT1 is configured with two L2 bridge domains and MAC learning enabled. DUT1 tested with ${nic_name}.Container is connected to VPP via Memif interface. Container is running same VPP version as running on DUT. Container is limited via cgroup to use cores allocated from pool of isolated CPUs. There are no memory contraints. [Ver] TG verification: TG finds and reports throughput NDR (Non DropRate) with zero packet loss tolerance and throughput PDR (Partial DropRate) with non-zero packet loss tolerance (LT) expressed in percentageof packets transmitted. NDR and PDR are discovered for differentEthernet L2 frame sizes using MLRsearch library.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.
# Copyright (c) 2021 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
|
| Force Tags | 2_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | ETH | L2BDMACLRN | BASE | MEMIF | DOCKER
| ... | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | eth-l2bdbasemaclrn-eth-2memif-1dcr
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | container
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | memif_plugin.so
| ${crypto_type}= | ${None}
| ${nic_name}= | Intel-X710
| ${nic_driver}= | vfio-pci
| ${nic_rxq_size}= | 0
| ${nic_txq_size}= | 0
| ${nic_pfs}= | 2
| ${nic_vfs}= | 0
| ${osi_layer}= | L2
| ${overhead}= | ${0}
# Traffic profile:
| ${traffic_profile}= | trex-stl-2n-ethip4-ip4src254
# Container
| ${container_engine}= | Docker
| ${container_chain_topology}= | chain7.1.7.1. Local Template¶
[Cfg] DUT runs L2BD switching config. Each DUT uses ${phy_cores} physical core(s) for worker threads. [Ver] Measure NDR and PDR values using MLRsearch algorithm.Arguments: - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). Type: integer, string - phy_cores - Number of physical cores. Type: integer - rxq - Number of RX queues, default value: ${None}. Type: integer
Set Test Variable \${frame_size}
Given Set Max Rate And Jumbo
And Add worker threads to all DUTs ${phy_cores} ${rxq}
And Pre-initialize layer driver ${nic_driver}
And Apply startup configuration on all VPP DUTs
When Initialize layer driver ${nic_driver}
And Initialize layer interface
And Start containers for test
And Initialize L2 Bridge Domain for multiple chains with memif pairs
Then Find NDR and PDR intervals using optimized search
7.1.7.2. 64B-1c-eth-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.1.7.3. 64B-2c-eth-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.1.7.4. 64B-4c-eth-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.1.7.5. 1518B-1c-eth-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.1.7.6. 1518B-2c-eth-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.1.7.7. 1518B-4c-eth-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.1.7.8. 9000B-1c-eth-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.1.7.9. 9000B-2c-eth-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.1.7.10. 9000B-4c-eth-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.1.7.11. IMIX-1c-eth-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.1.7.12. IMIX-2c-eth-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.1.7.13. IMIX-4c-eth-l2bdbasemaclrn-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.1.8. 2n1l-10ge2p1x710-eth-l2xcbase-eth-2memif-1dcr-ndrpdr suite¶
RFC2544: Pkt throughput L2XC test cases [Top] Network Topologies: TG-DUT1-TG 2-node circular topology with single links between nodes. [Enc] Packet Encapsulations: Eth-IPv4 for L2 cross connect. [Cfg] DUT configuration: DUT1 is configured with L2 cross-connect. DUT1 tested with ${nic_name}.Container is connected to VPP via Memif interface. Container is running same VPP version as running on DUT. Container is limited via cgroup to use cores allocated from pool of isolated CPUs. There are no memory contraints. [Ver] TG verification: TG finds and reports throughput NDR (Non DropRate) with zero packet loss tolerance and throughput PDR (Partial DropRate) with non-zero packet loss tolerance (LT) expressed in percentageof packets transmitted. NDR and PDR are discovered for differentEthernet L2 frame sizes using MLRsearch library.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.
# Copyright (c) 2021 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
|
| Force Tags | 2_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | ETH | L2XCFWD | BASE | MEMIF | DOCKER
| ... | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | eth-l2xcbase-eth-2memif-1dcr
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | container
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | memif_plugin.so
| ${crypto_type}= | ${None}
| ${nic_name}= | Intel-X710
| ${nic_driver}= | vfio-pci
| ${nic_rxq_size}= | 0
| ${nic_txq_size}= | 0
| ${nic_pfs}= | 2
| ${nic_vfs}= | 0
| ${osi_layer}= | L2
| ${overhead}= | ${0}
# Traffic profile:
| ${traffic_profile}= | trex-stl-2n-ethip4-ip4src254
# Container
| ${container_engine}= | Docker
| ${container_chain_topology}= | chain7.1.8.1. Local Template¶
[Cfg] DUT runs L2XC switching config. Each DUT uses ${phy_cores} physical core(s) for worker threads. [Ver] Measure NDR and PDR values using MLRsearch algorithm.Arguments: - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). Type: integer, string - phy_cores - Number of physical cores. Type: integer - rxq - Number of RX queues, default value: ${None}. Type: integer
Set Test Variable \${frame_size}
Given Set Max Rate And Jumbo
And Add worker threads to all DUTs ${phy_cores} ${rxq}
And Pre-initialize layer driver ${nic_driver}
And Apply startup configuration on all VPP DUTs
When Initialize layer driver ${nic_driver}
And Initialize layer interface
And Start containers for test
And Initialize L2 xconnect with memif pairs
Then Find NDR and PDR intervals using optimized search
7.1.8.2. 64B-1c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.1.8.3. 64B-2c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.1.8.4. 64B-4c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.1.8.5. 1518B-1c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.1.8.6. 1518B-2c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.1.8.7. 1518B-4c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.1.8.8. 9000B-1c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.1.8.9. 9000B-2c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.1.8.10. 9000B-4c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.1.8.11. IMIX-1c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.1.8.12. IMIX-2c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.1.8.13. IMIX-4c-eth-l2xcbase-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.1.9. 2n1l-10ge2p1x710-ethip4-ip4base-eth-2memif-1dcr-ndrpdr suite¶
RFC2544: Pkt throughput IPv4 routing test cases [Top] Network Topologies: TG-DUT1-TG 2-node circular topology with single links between nodes. [Enc] Packet Encapsulations: Eth-IPv4 for IPv4 routing. [Cfg] DUT configuration: DUT1 is configured with IPv4 routing with two FIB tables and two static IPv4 /24 route entries per FIB table. Container is connected to VPP via Memif interface. Container is running same VPP version as running on DUT. Container is limited via cgroup to use cores allocated from pool of isolated CPUs. There are no memory contraints. DUT1 is tested with ${nic_name}.[Ver] TG verification: TG finds and reports throughput NDR (Non DropRate) with zero packet loss tolerance and throughput PDR (Partial DropRate) with non-zero packet loss tolerance (LT) expressed in percentageof packets transmitted. NDR and PDR are discovered for differentEthernet L2 frame sizes using MLRsearch library.Test packets are generated by TG on links to DUT1. 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.
# Copyright (c) 2021 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
|
| Force Tags | 2_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | ETH | IP4FWD | BASE | MEMIF | DOCKER
| ... | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4-ip4base-eth-2memif-1dcr
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | container
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | memif_plugin.so
| ${crypto_type}= | ${None}
| ${nic_name}= | Intel-X710
| ${nic_driver}= | vfio-pci
| ${nic_rxq_size}= | 0
| ${nic_txq_size}= | 0
| ${nic_pfs}= | 2
| ${nic_vfs}= | 0
| ${osi_layer}= | L3
| ${overhead}= | ${0}
# Traffic profile:
| ${traffic_profile}= | trex-stl-2n-ethip4-ip4src254
# Container
| ${container_engine}= | Docker
| ${container_chain_topology}= | chain7.1.9.1. Local Template¶
[Cfg] DUT runs IPv4 routing config. Each DUT uses ${phy_cores} physical core(s) for worker threads. [Ver] Measure NDR and PDR values using MLRsearch algorithm.Arguments: - frame_size - Framesize in Bytes in integer or string (IMIX_v4_1). Type: integer, string - phy_cores - Number of physical cores. Type: integer - rxq - Number of RX queues, default value: ${None}. Type: integer
Set Test Variable \${frame_size}
Given Set Max Rate And Jumbo
And Add worker threads to all DUTs ${phy_cores} ${rxq}
And Pre-initialize layer driver ${nic_driver}
And Apply startup configuration on all VPP DUTs
When Initialize layer driver ${nic_driver}
And Initialize layer interface
And Start containers for test
And Initialize IPv4 routing with memif pairs
Then Find NDR and PDR intervals using optimized search
7.1.9.2. 64B-1c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.1.9.3. 64B-2c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.1.9.4. 64B-4c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.1.9.5. 1518B-1c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.1.9.6. 1518B-2c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.1.9.7. 1518B-4c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.1.9.8. 9000B-1c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.1.9.9. 9000B-2c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.1.9.10. 9000B-4c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.1.9.11. IMIX-1c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.1.9.12. IMIX-2c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.1.9.13. IMIX-4c-ethip4-ip4base-eth-2memif-1dcr-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}