7.6. ip4_tunnels¶
7.6.1. 10ge2p1x710-dot1q–ethip4vxlan-l2bdscale100l2bd100vlan100vxlan-ndrpdr suite¶
RFC2544: Pkt throughput L2BD with IEEE 802.1Q and VXLANoIPv4 test cases
[Top] Network Topologies: TG-DUT1-DUT2-TG 3-node circular topology with single links between nodes.
[Enc] Packet Encapsulations: Eth-Dot1Q-IPv4 for L2 switching of IPv4. on TG-DUTn. Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUTs.
[Cfg] DUT configuration: DUT1 and DUT2 are configured with 100 L2 bridge domains. VXLAN tunnels are configured between L2BDs on DUT1 and DUT2. DUT1 and DUT2 are tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
Test packets are generated by TG on links to DUTs. TG traffic profile contains two L3 flow-groups (flow-group per direction, up to 64,5k flows per flow-group) with all packets containing Ethernet header, IEEE 802.1Q 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.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | L2BDMACLRN | SCALE | L2BD_100 | DOT1Q | VLAN_100
| ... | ENCAP | VXLAN | L2OVRLAY | IP4UNRLAY | VXLAN_100 | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | dot1q--ethip4vxlan-l2bdscale100l2bd100vlan100vxlan
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_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}= | ${50}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-dot1qip4-vlan100ip4src254ip4dst254
# Number of VXLAN tunnels
| ${vxlan_count}= | ${100}
7.6.1.1. Local Template¶
[Cfg] Each DUT runs L2BD forwarding config with VLAN and VXLAN 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 ${nic_driver}
And Initialize layer interface
And Initialize L2 bridge domain with VLAN and VXLANoIPv4 in 3-node circular topology vxlan_count=${vxlan_count}
Then Find NDR and PDR intervals using optimized search
7.6.1.2. 64B-1c-dot1q–ethip4vxlan-l2bdscale100l2bd100vlan100vxlan-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.1.3. 64B-2c-dot1q–ethip4vxlan-l2bdscale100l2bd100vlan100vxlan-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.1.4. 64B-4c-dot1q–ethip4vxlan-l2bdscale100l2bd100vlan100vxlan-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.1.5. 1518B-1c-dot1q–ethip4vxlan-l2bdscale100l2bd100vlan100vxlan-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.1.6. 1518B-2c-dot1q–ethip4vxlan-l2bdscale100l2bd100vlan100vxlan-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.1.7. 1518B-4c-dot1q–ethip4vxlan-l2bdscale100l2bd100vlan100vxlan-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.1.8. 9000B-1c-dot1q–ethip4vxlan-l2bdscale100l2bd100vlan100vxlan-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.1.9. 9000B-2c-dot1q–ethip4vxlan-l2bdscale100l2bd100vlan100vxlan-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.1.10. 9000B-4c-dot1q–ethip4vxlan-l2bdscale100l2bd100vlan100vxlan-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.1.11. IMIX-1c-dot1q–ethip4vxlan-l2bdscale100l2bd100vlan100vxlan-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.1.12. IMIX-2c-dot1q–ethip4vxlan-l2bdscale100l2bd100vlan100vxlan-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.1.13. IMIX-4c-dot1q–ethip4vxlan-l2bdscale100l2bd100vlan100vxlan-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.2. 10ge2p1x710-dot1q–ethip4vxlan-l2bdscale10l2bd10vlan10vxlan-ndrpdr suite¶
RFC2544: Pkt throughput L2BD with IEEE 802.1Q and VXLANoIPv4 test cases
[Top] Network Topologies: TG-DUT1-DUT2-TG 3-node circular topology with single links between nodes.
[Enc] Packet Encapsulations: Eth-Dot1Q-IPv4 for L2 switching of IPv4. on TG-DUTn. Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUTs.
[Cfg] DUT configuration: DUT1 and DUT2 are configured with 10 L2 bridge domains. VXLAN tunnels are configured between L2BDs on DUT1 and DUT2. DUT1 and DUT2 are tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
Test packets are generated by TG on links to DUTs. TG traffic profile contains two L3 flow-groups (flow-group per direction, up to 64,5k flows per flow-group) with all packets containing Ethernet header, IEEE 802.1Q 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.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | L2BDMACLRN | SCALE | L2BD_10 | DOT1Q | VLAN_10
| ... | ENCAP | VXLAN | L2OVRLAY | IP4UNRLAY | VXLAN_10 | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | dot1q--ethip4vxlan-l2bdscale10l2bd10vlan10vxlan
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_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}= | ${50}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-dot1qip4-vlan10ip4src254ip4dst254
# Number of VXLAN tunnels
| ${vxlan_count}= | ${10}
7.6.2.1. Local Template¶
[Cfg] Each DUT runs L2BD forwarding config with VLAN and VXLAN 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 ${nic_driver}
And Initialize layer interface
And Initialize L2 bridge domain with VLAN and VXLANoIPv4 in 3-node circular topology vxlan_count=${vxlan_count}
Then Find NDR and PDR intervals using optimized search
7.6.2.2. 64B-1c-dot1q–ethip4vxlan-l2bdscale10l2bd10vlan10vxlan-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.2.3. 64B-2c-dot1q–ethip4vxlan-l2bdscale10l2bd10vlan10vxlan-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.2.4. 64B-4c-dot1q–ethip4vxlan-l2bdscale10l2bd10vlan10vxlan-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.2.5. 1518B-1c-dot1q–ethip4vxlan-l2bdscale10l2bd10vlan10vxlan-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.2.6. 1518B-2c-dot1q–ethip4vxlan-l2bdscale10l2bd10vlan10vxlan-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.2.7. 1518B-4c-dot1q–ethip4vxlan-l2bdscale10l2bd10vlan10vxlan-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.2.8. 9000B-1c-dot1q–ethip4vxlan-l2bdscale10l2bd10vlan10vxlan-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.2.9. 9000B-2c-dot1q–ethip4vxlan-l2bdscale10l2bd10vlan10vxlan-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.2.10. 9000B-4c-dot1q–ethip4vxlan-l2bdscale10l2bd10vlan10vxlan-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.2.11. IMIX-1c-dot1q–ethip4vxlan-l2bdscale10l2bd10vlan10vxlan-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.2.12. IMIX-2c-dot1q–ethip4vxlan-l2bdscale10l2bd10vlan10vxlan-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.2.13. IMIX-4c-dot1q–ethip4vxlan-l2bdscale10l2bd10vlan10vxlan-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.3. 10ge2p1x710-dot1q–ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan-ndrpdr suite¶
RFC2544: Pkt throughput L2BD with IEEE 802.1Q and VXLANoIPv4 test cases
[Top] Network Topologies: TG-DUT1-DUT2-TG 3-node circular topology with single links between nodes.
[Enc] Packet Encapsulations: Eth-Dot1Q-IPv4 for L2 switching of IPv4. on TG-DUTn. Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUTs.
[Cfg] DUT configuration: DUT1 and DUT2 are configured with 1k L2 bridge domains. VXLAN tunnels are configured between L2BDs on DUT1 and DUT2. DUT1 and DUT2 are tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
Test packets are generated by TG on links to DUTs. TG traffic profile contains two L3 flow-groups (flow-group per direction, up to 64,5k flows per flow-group) with all packets containing Ethernet header, IEEE 802.1Q 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.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | L2BDMACLRN | SCALE | L2BD_1K | DOT1Q | VLAN_1K
| ... | ENCAP | VXLAN | L2OVRLAY | IP4UNRLAY | VXLAN_1K | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | dot1q--ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_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}= | ${50}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-dot1qip4-vlan1kip4src254ip4dst254
# Number of VXLAN tunnels
| ${vxlan_count}= | ${1000}
7.6.3.1. Local Template¶
[Cfg] Each DUT runs L2BD forwarding config with VLAN and VXLAN 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 ${nic_driver}
And Initialize layer interface
And Initialize L2 bridge domain with VLAN and VXLANoIPv4 in 3-node circular topology vxlan_count=${vxlan_count}
Then Find NDR and PDR intervals using optimized search
7.6.3.2. 64B-1c-dot1q–ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.3.3. 64B-2c-dot1q–ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.3.4. 64B-4c-dot1q–ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.3.5. 1518B-1c-dot1q–ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.3.6. 1518B-2c-dot1q–ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.3.7. 1518B-4c-dot1q–ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.3.8. 9000B-1c-dot1q–ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.3.9. 9000B-2c-dot1q–ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.3.10. 9000B-4c-dot1q–ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.3.11. IMIX-1c-dot1q–ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.3.12. IMIX-2c-dot1q–ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.3.13. IMIX-4c-dot1q–ethip4vxlan-l2bdscale1kl2bd1kvlan1kvxlan-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.4. 10ge2p1x710-dot1q–ethip4vxlan-l2bdscale1l2bd1vlan1vxlan-ndrpdr suite¶
RFC2544: Pkt throughput L2BD with IEEE 802.1Q and VXLANoIPv4 test cases
[Top] Network Topologies: TG-DUT1-DUT2-TG 3-node circular topology with single links between nodes.
[Enc] Packet Encapsulations: Eth-Dot1Q-IPv4 for L2 switching of IPv4. on TG-DUTn. Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUTs.
[Cfg] DUT configuration: DUT1 and DUT2 are configured with 1 L2 bridge domains. VXLAN tunnels are configured between L2BDs on DUT1 and DUT2. DUT1 and DUT2 are tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
Test packets are generated by TG on links to DUTs. TG traffic profile contains two L3 flow-groups (flow-group per direction, up to 64,5k flows per flow-group) with all packets containing Ethernet header, IEEE 802.1Q 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.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | L2BDMACLRN | SCALE | L2BD_1 | DOT1Q | VLAN_1
| ... | ENCAP | VXLAN | L2OVRLAY | IP4UNRLAY | VXLAN_1 | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | dot1q--ethip4vxlan-l2bdscale1l2bd1vlan1vxlan
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_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}= | ${50}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-dot1qip4-vlan1ip4src254ip4dst254
# Number of VXLAN tunnels
| ${vxlan_count}= | ${1}
7.6.4.1. Local Template¶
[Cfg] Each DUT runs L2BD forwarding config with VLAN and VXLAN 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 ${nic_driver}
And Initialize layer interface
And Initialize L2 bridge domain with VLAN and VXLANoIPv4 in 3-node circular topology vxlan_count=${vxlan_count}
Then Find NDR and PDR intervals using optimized search
7.6.4.2. 64B-1c-dot1q–ethip4vxlan-l2bdscale1l2bd1vlan1vxlan-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.4.3. 64B-2c-dot1q–ethip4vxlan-l2bdscale1l2bd1vlan1vxlan-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.4.4. 64B-4c-dot1q–ethip4vxlan-l2bdscale1l2bd1vlan1vxlan-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.4.5. 1518B-1c-dot1q–ethip4vxlan-l2bdscale1l2bd1vlan1vxlan-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.4.6. 1518B-2c-dot1q–ethip4vxlan-l2bdscale1l2bd1vlan1vxlan-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.4.7. 1518B-4c-dot1q–ethip4vxlan-l2bdscale1l2bd1vlan1vxlan-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.4.8. 9000B-1c-dot1q–ethip4vxlan-l2bdscale1l2bd1vlan1vxlan-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.4.9. 9000B-2c-dot1q–ethip4vxlan-l2bdscale1l2bd1vlan1vxlan-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.4.10. 9000B-4c-dot1q–ethip4vxlan-l2bdscale1l2bd1vlan1vxlan-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.4.11. IMIX-1c-dot1q–ethip4vxlan-l2bdscale1l2bd1vlan1vxlan-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.4.12. IMIX-2c-dot1q–ethip4vxlan-l2bdscale1l2bd1vlan1vxlan-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.4.13. IMIX-4c-dot1q–ethip4vxlan-l2bdscale1l2bd1vlan1vxlan-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.5. 10ge2p1x710-ethip4gtpuhw-ip4base-ndrpdr suite¶
RFC2544: Pkt throughput GTPU test cases
[Top] Network Topologies: TG-DUT1-DUT2-TG 3-node circular topology with single links between nodes.
[Enc] Packet Encapsulations: Eth-IPv4-GTPU on DUT1-DUT2, Eth-IPv4 on TG-DUTn for IPv4 routing over GTPU tunnel.
[Cfg] DUT configuration: DUT1 and DUT2 are configured with IPv4. routing and static routes. GTPU tunnel is configured on each DUT and enabled GTPU offload RX. DUT1 and DUT2 are tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
Test packets are generated by TG on links to DUTs. TG traffic profile contains two L3 flow-groups 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.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
| Resource | resources/libraries/robot/overlay/gtpu.robot
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | IP4FWD | IP4BASE | ENCAP | GTPU | GTPU_HWACCEL
| ... | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4gtpuhw-ip4base
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | gtpu_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}= | ${36}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-ethip4-ip4src2537.6.5.1. Local Template¶
[Cfg] DUT runs GTPU routing config, and Enable GTPU offload RX 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 ${nic_driver}
And Initialize layer interface
And Initialize IP4 forwarding with GTPU tunnel in 3-node circular topology offload=${True}
Then Find NDR and PDR intervals using optimized search
7.6.5.2. 64B-1c-ethip4gtpuhw-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.5.3. 64B-2c-ethip4gtpuhw-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.5.4. 64B-4c-ethip4gtpuhw-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.5.5. 1518B-1c-ethip4gtpuhw-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.5.6. 1518B-2c-ethip4gtpuhw-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.5.7. 1518B-4c-ethip4gtpuhw-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.5.8. 9000B-1c-ethip4gtpuhw-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.5.9. 9000B-2c-ethip4gtpuhw-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.5.10. 9000B-4c-ethip4gtpuhw-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.5.11. IMIX-1c-ethip4gtpuhw-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.5.12. IMIX-2c-ethip4gtpuhw-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.5.13. IMIX-4c-ethip4gtpuhw-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.6. 10ge2p1x710-ethip4gtpusw-ip4base-ndrpdr suite¶
RFC2544: Pkt throughput GTPU test cases
[Top] Network Topologies: TG-DUT1-DUT2-TG 3-node circular topology with single links between nodes.
[Enc] Packet Encapsulations: Eth-IPv4-GTPU on DUT1-DUT2, Eth-IPv4 on TG-DUTn for IPv4 routing over GTPU tunnel.
[Cfg] DUT configuration: DUT1 and DUT2 are configured with IPv4. routing and static routes. GTPU tunnel is configured between DUT1 and DUT2. DUT1 and DUT2 are tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
Test packets are generated by TG on links to DUTs. TG traffic profile contains two L3 flow-groups 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.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
| Resource | resources/libraries/robot/overlay/gtpu.robot
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | IP4FWD | IP4BASE | ENCAP | GTPU | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4gtpusw-ip4base
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | gtpu_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}= | ${36}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-ethip4-ip4src2537.6.6.1. Local Template¶
[Cfg] DUT runs GTPU 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 ${nic_driver}
And Initialize layer interface
And Initialize IP4 forwarding with GTPU tunnel in 3-node circular topology
Then Find NDR and PDR intervals using optimized search
7.6.6.2. 64B-1c-ethip4gtpusw-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.6.3. 64B-2c-ethip4gtpusw-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.6.4. 64B-4c-ethip4gtpusw-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.6.5. 1518B-1c-ethip4gtpusw-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.6.6. 1518B-2c-ethip4gtpusw-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.6.7. 1518B-4c-ethip4gtpusw-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.6.8. 9000B-1c-ethip4gtpusw-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.6.9. 9000B-2c-ethip4gtpusw-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.6.10. 9000B-4c-ethip4gtpusw-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.6.11. IMIX-1c-ethip4gtpusw-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.6.12. IMIX-2c-ethip4gtpusw-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.6.13. IMIX-4c-ethip4gtpusw-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.7. 10ge2p1x710-ethip4lispip4-ip4base-ndrpdr suite¶
RFC2544: 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 are tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
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: RFC6830.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
| Resource | resources/libraries/robot/overlay/lisp.robot
|
| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | IP4FWD | ENCAP | LISP | IP4UNRLAY | IP4OVRLAY
| ... | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4lispip4-ip4base
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | lisp_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}= | ${8}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-ethip4-ip4src2537.6.7.1. Local Template¶
[Cfg] DUT runs IPv4 LISP remote static mappings and whitelist filters 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 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} ${DUT1_${int}2}[0] ${NONE} ${dut2} ${DUT2_${int}1}[0] ${NONE} ${duts_locator_set} ${dut1_ip4_eid} ${dut2_ip4_eid} ${dut1_ip4_static_adjacency} ${dut2_ip4_static_adjacency}
Then Find NDR and PDR intervals using optimized search
7.6.7.2. 64B-1c-ethip4lispip4-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.7.3. 64B-2c-ethip4lispip4-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.7.4. 64B-4c-ethip4lispip4-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.7.5. 1518B-1c-ethip4lispip4-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.7.6. 1518B-2c-ethip4lispip4-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.7.7. 1518B-4c-ethip4lispip4-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.7.8. 9000B-1c-ethip4lispip4-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.7.9. 9000B-2c-ethip4lispip4-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.7.10. 9000B-4c-ethip4lispip4-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.7.11. IMIX-1c-ethip4lispip4-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.7.12. IMIX-2c-ethip4lispip4-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.7.13. IMIX-4c-ethip4lispip4-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.8. 10ge2p1x710-ethip4lispip6-ip4base-ndrpdr suite¶
RFC2544: 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, 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 are tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
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: RFC6830.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
| Resource | resources/libraries/robot/overlay/lisp.robot
|
| Variables | resources/test_data/lisp/performance/lisp_static_adjacency.py
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | IP4FWD | ENCAP | LISP | IP6UNRLAY | IP4OVRLAY
| ... | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4lispip6-ip4base
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | lisp_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}= | ${48}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-ethip4-ip4src2537.6.8.1. Local Template¶
[Cfg] DUT runs IPv6 LISP remote static mappings and whitelist filters 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 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} ${DUT1_${int}2}[0] ${NONE} ${dut2} ${DUT2_${int}1}[0] ${NONE} ${duts_locator_set} ${dut1_ip4o6_eid} ${dut2_ip4o6_eid} ${dut1_ip4o6_static_adjacency} ${dut2_ip4o6_static_adjacency}
Then Find NDR and PDR intervals using optimized search
7.6.8.2. 64B-1c-ethip4lispip6-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.8.3. 64B-2c-ethip4lispip6-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.8.4. 64B-4c-ethip4lispip6-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.8.5. 1518B-1c-ethip4lispip6-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.8.6. 1518B-2c-ethip4lispip6-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.8.7. 1518B-4c-ethip4lispip6-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.8.8. 9000B-1c-ethip4lispip6-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.8.9. 9000B-2c-ethip4lispip6-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.8.10. 9000B-4c-ethip4lispip6-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.8.11. IMIX-1c-ethip4lispip6-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.8.12. IMIX-2c-ethip4lispip6-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.8.13. IMIX-4c-ethip4lispip6-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.9. 10ge2p1x710-ethip4udpwireguard1tnlsw-ip4base-ndrpdr suite¶
RFC2544: Pkt throughput IPv4 WireGuard tunnel mode.
[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-UDP-WireGuard on DUT1-DUT2.
[Cfg] DUT configuration: DUT1 and DUT2 are configured with single WireGuard tunnels between them. DUTs get IPv4 traffic from TG, 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 and 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 MLRsearch library.
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 WireGuard 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) is applied to both streams.
[Ref] Applicable standard specifications: RFC4303 and RFC2544.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
| Resource | resources/libraries/robot/wireguard/wireguard.robot
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR | TNL_1
| ... | IP4FWD | NIC_Intel-X710 | WIREGUARD | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4udpwireguard1tnlsw-ip4base
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so
| ... | crypto_native_plugin.so | crypto_ipsecmb_plugin.so
| ... | wireguard_plugin.so | crypto_openssl_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}= | ${60}
| ${tg_if1_ip4}= | 192.168.10.2
| ${dut1_if1_ip4}= | 192.168.10.1
| ${dut1_if2_ip4}= | 200.0.0.1
| ${dut2_if1_ip4}= | 200.0.0.2
| ${dut2_if2_ip4}= | 192.168.20.1
| ${tg_if2_ip4}= | 192.168.20.2
| ${wg_if1_ip4}= | 192.168.110.1
| ${wg_if2_ip4}= | 192.168.120.1
| ${raddr_ip4}= | 20.0.0.0
| ${laddr_ip4}= | 10.0.0.0
| ${n_tunnels}= | ${1}
| ${listen_port}= | ${51820}
| ${keepalive_time}= | ${256}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-ethip4-ip4dst${n_tunnels}
7.6.9.1. Local Template¶
[Cfg] DUT runs wireguard TUNNEL. 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 Initialize WireGuard in 3-node circular topology
And VPP WireGuard Create Tunnel Interface On DUTs ${nodes} ${DUT1_${int}2}[0] ${DUT2_${int}1}[0] ${dut1_if2_ip4} ${dut2_if1_ip4} ${DUT1_${int}2_mac}[0] ${DUT2_${int}1_mac}[0] ${wg_if1_ip4} ${wg_if2_ip4} ${n_tunnels} ${listen_port} ${keepalive_time} ${laddr_ip4} ${raddr_ip4}
Then Find NDR and PDR intervals using optimized search
7.6.9.2. 64B-1c-ethip4udpwireguard1tnlsw-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.9.3. 64B-2c-ethip4udpwireguard1tnlsw-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.9.4. 64B-4c-ethip4udpwireguard1tnlsw-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.9.5. 1518B-1c-ethip4udpwireguard1tnlsw-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.9.6. 1518B-2c-ethip4udpwireguard1tnlsw-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.9.7. 1518B-4c-ethip4udpwireguard1tnlsw-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.9.8. 9000B-1c-ethip4udpwireguard1tnlsw-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.9.9. 9000B-2c-ethip4udpwireguard1tnlsw-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.9.10. 9000B-4c-ethip4udpwireguard1tnlsw-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.9.11. IMIX-1c-ethip4udpwireguard1tnlsw-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.9.12. IMIX-2c-ethip4udpwireguard1tnlsw-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.9.13. IMIX-4c-ethip4udpwireguard1tnlsw-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.10. 10ge2p1x710-ethip4vxlan-l2bdbasemaclrn-ndrpdr suite¶
RFC2544: Pkt 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 switching of IPv4 on TG-DUTn. Eth-IPv4-VXLAN-Eth-IPv4 is applied on link between DUTs.
[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 are tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 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.
[Ref] Applicable standard specifications: RFC2544, RFC7348.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | L2BDMACLRN | ENCAP | VXLAN | L2OVRLAY | IP4UNRLAY
| ... | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4vxlan-l2bdbasemaclrn
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_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}= | ${50}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-ethip4-ip4src2547.6.10.1. Local Template¶
[Cfg] Each DUT runs L2BD forwarding config with VXLAN 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 ${nic_driver}
And Initialize layer interface
And Initialize L2 bridge domain with VXLANoIPv4 in 3-node circular topology
Then Find NDR and PDR intervals using optimized search
7.6.10.2. 64B-1c-ethip4vxlan-l2bdbasemaclrn-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.10.3. 64B-2c-ethip4vxlan-l2bdbasemaclrn-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.10.4. 64B-4c-ethip4vxlan-l2bdbasemaclrn-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.10.5. 1518B-1c-ethip4vxlan-l2bdbasemaclrn-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.10.6. 1518B-2c-ethip4vxlan-l2bdbasemaclrn-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.10.7. 1518B-4c-ethip4vxlan-l2bdbasemaclrn-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.10.8. 9000B-1c-ethip4vxlan-l2bdbasemaclrn-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.10.9. 9000B-2c-ethip4vxlan-l2bdbasemaclrn-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.10.10. 9000B-4c-ethip4vxlan-l2bdbasemaclrn-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.10.11. IMIX-1c-ethip4vxlan-l2bdbasemaclrn-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.10.12. IMIX-2c-ethip4vxlan-l2bdbasemaclrn-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.10.13. IMIX-4c-ethip4vxlan-l2bdbasemaclrn-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.11. 10ge2p1x710-ethip4vxlan-l2xcbase-ndrpdr suite¶
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 are tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
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.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
|
| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | L2XCFWD | ENCAP | VXLAN | L2OVRLAY | IP4UNRLAY
| ... | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4vxlan-l2xcbase
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_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}= | ${50}
# Traffic profile:
| ${traffic_profile}= | trex-stl-3n-ethip4-ip4src2547.6.11.1. Local Template¶
[Cfg] DUT runs L2XC forwarding 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 ${nic_driver}
And Initialize layer interface
And Initialize L2 xconnect with VXLANoIPv4 in 3-node circular topology
Then Find NDR and PDR intervals using optimized search
7.6.11.2. 64B-1c-ethip4vxlan-l2xcbase-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.11.3. 64B-2c-ethip4vxlan-l2xcbase-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.11.4. 64B-4c-ethip4vxlan-l2xcbase-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.11.5. 1518B-1c-ethip4vxlan-l2xcbase-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.11.6. 1518B-2c-ethip4vxlan-l2xcbase-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.11.7. 1518B-4c-ethip4vxlan-l2xcbase-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.11.8. 9000B-1c-ethip4vxlan-l2xcbase-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.11.9. 9000B-2c-ethip4vxlan-l2xcbase-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.11.10. 9000B-4c-ethip4vxlan-l2xcbase-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.11.11. IMIX-1c-ethip4vxlan-l2xcbase-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.11.12. IMIX-2c-ethip4vxlan-l2xcbase-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.11.13. IMIX-4c-ethip4vxlan-l2xcbase-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.12. 2n1l-10ge2p1x710-ethip4–ethip4udpgeneve-1024tun-ip4base-ndrpdr suite¶
RFC2544: Pkt throughput GENEVE tunnel L3 mode performance test cases
[Top] Network Topologies: TG-DUT1-TG 2-node circular topology with single links between nodes.
[Enc] Packet Encapsulations: Eth-IPv4 between TG-if1 and DUT1-if1 and Eth-IPv4-UDP-GENEVE-Eth-IPv4 between DUT1-if2 and TG-if2 for IPv4 routing over GENEVE tunnels.
[Cfg] DUT configuration: DUT1 is configured with IPv4 routing over ${n_tunnels} GENEVE tunnels and ${${4} * ${n_tunnels}} static IPv4 /24 route entries. DUT1 is tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
Test packets are generated by TG on links to DUT1. TG traffic profile contains two L3 flow-groups (flow-group per direction, 1 flow 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, RFC791, RFC768 and RFC8926.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
| Resource | resources/libraries/robot/ip/geneve.robot
|
| Force Tags | 2_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | ETH | IP4FWD | IP4BASE | UDP | ENCAP | GENEVE_L3MODE
| ... | SCALE | GENEVE4_1024TUN | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4--ethip4udpgeneve-1024tun-ip4base
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | geneve4
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | geneve_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}= | ${50}
# IP settings
| ${dut1_if1_ip4}= | 20.0.0.1
| ${dut1_if2_ip4}= | 30.0.0.1
| ${tg_if1_ip4}= | 20.0.0.2
| ${tg_if2_ip4}= | 30.0.0.2
# GENEVE settings
| ${gen_mode}= | ${osi_layer}
| ${n_tunnels}= | ${1024}
| &{gen_tunnel}=
| ... | local=1.1.1.2 | remote=1.1.1.1 | vni=${1}
| ... | src_ip=10.128.1.0 | dst_ip=10.0.1.0 | ip_mask=${24} | if_ip=11.0.1.2
# Traffic profile
| ${traffic_profile}= | trex-stl-2n-ethip4-geneve-${n_tunnels}t7.6.12.1. Local Template¶
[Cfg] DUT runs GENEVE ${gen_mode} mode configuration. 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 ${nic_driver}
And Initialize layer interface
And Initialize GENEVE L3 mode in circular topology
Then Find NDR and PDR intervals using optimized search
7.6.12.2. 64B-1c-ethip4–ethip4udpgeneve-1024tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.12.3. 64B-2c-ethip4–ethip4udpgeneve-1024tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.12.4. 64B-4c-ethip4–ethip4udpgeneve-1024tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.12.5. 1518B-1c-ethip4–ethip4udpgeneve-1024tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.12.6. 1518B-2c-ethip4–ethip4udpgeneve-1024tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.12.7. 1518B-4c-ethip4–ethip4udpgeneve-1024tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.12.8. 9000B-1c-ethip4–ethip4udpgeneve-1024tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.12.9. 9000B-2c-ethip4–ethip4udpgeneve-1024tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.12.10. 9000B-4c-ethip4–ethip4udpgeneve-1024tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.12.11. IMIX-1c-ethip4–ethip4udpgeneve-1024tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.12.12. IMIX-2c-ethip4–ethip4udpgeneve-1024tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.12.13. IMIX-4c-ethip4–ethip4udpgeneve-1024tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.13. 2n1l-10ge2p1x710-ethip4–ethip4udpgeneve-16tun-ip4base-ndrpdr suite¶
RFC2544: Pkt throughput GENEVE tunnel L3 mode performance test cases
[Top] Network Topologies: TG-DUT1-TG 2-node circular topology with single links between nodes.
[Enc] Packet Encapsulations: Eth-IPv4 between TG-if1 and DUT1-if1 and Eth-IPv4-UDP-GENEVE-Eth-IPv4 between DUT1-if2 and TG-if2 for IPv4 routing over GENEVE tunnels.
[Cfg] DUT configuration: DUT1 is configured with IPv4 routing over ${n_tunnels} GENEVE tunnels and ${${4} * ${n_tunnels}} static IPv4 /24 route entries. DUT1 is tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
Test packets are generated by TG on links to DUT1. TG traffic profile contains two L3 flow-groups (flow-group per direction, 1 flow 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, RFC791, RFC768 and RFC8926.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
| Resource | resources/libraries/robot/ip/geneve.robot
|
| Force Tags | 2_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | ETH | IP4FWD | IP4BASE | UDP | ENCAP | GENEVE_L3MODE
| ... | SCALE | GENEVE4_16TUN | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4--ethip4udpgeneve-16tun-ip4base
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | geneve4
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | geneve_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}= | ${50}
# IP settings
| ${dut1_if1_ip4}= | 20.0.0.1
| ${dut1_if2_ip4}= | 30.0.0.1
| ${tg_if1_ip4}= | 20.0.0.2
| ${tg_if2_ip4}= | 30.0.0.2
# GENEVE settings
| ${gen_mode}= | ${osi_layer}
| ${n_tunnels}= | ${16}
| &{gen_tunnel}=
| ... | local=1.1.1.2 | remote=1.1.1.1 | vni=${1}
| ... | src_ip=10.128.1.0 | dst_ip=10.0.1.0 | ip_mask=${24} | if_ip=11.0.1.2
# Traffic profile
| ${traffic_profile}= | trex-stl-2n-ethip4-geneve-${n_tunnels}t7.6.13.1. Local Template¶
[Cfg] DUT runs GENEVE ${gen_mode} mode configuration. 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 ${nic_driver}
And Initialize layer interface
And Initialize GENEVE L3 mode in circular topology
Then Find NDR and PDR intervals using optimized search
7.6.13.2. 64B-1c-ethip4–ethip4udpgeneve-16tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.13.3. 64B-2c-ethip4–ethip4udpgeneve-16tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.13.4. 64B-4c-ethip4–ethip4udpgeneve-16tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.13.5. 1518B-1c-ethip4–ethip4udpgeneve-16tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.13.6. 1518B-2c-ethip4–ethip4udpgeneve-16tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.13.7. 1518B-4c-ethip4–ethip4udpgeneve-16tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.13.8. 9000B-1c-ethip4–ethip4udpgeneve-16tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.13.9. 9000B-2c-ethip4–ethip4udpgeneve-16tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.13.10. 9000B-4c-ethip4–ethip4udpgeneve-16tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.13.11. IMIX-1c-ethip4–ethip4udpgeneve-16tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.13.12. IMIX-2c-ethip4–ethip4udpgeneve-16tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.13.13. IMIX-4c-ethip4–ethip4udpgeneve-16tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.14. 2n1l-10ge2p1x710-ethip4–ethip4udpgeneve-1tun-ip4base-ndrpdr suite¶
RFC2544: Pkt throughput GENEVE tunnel L3 mode performance test cases
[Top] Network Topologies: TG-DUT1-TG 2-node circular topology with single links between nodes.
[Enc] Packet Encapsulations: Eth-IPv4 between TG-if1 and DUT1-if1 and Eth-IPv4-UDP-GENEVE-Eth-IPv4 between DUT1-if2 and TG-if2 for IPv4 routing over GENEVE tunnel.
[Cfg] DUT configuration: DUT1 is configured with IPv4 routing over GENEVE tunnel and 4 static IPv4 /24 route entries. DUT1 is tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
Test packets are generated by TG on links to DUT1. TG traffic profile contains two L3 flow-groups (flow-group per direction, 1 flow 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, RFC791, RFC768 and RFC8926.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
| Resource | resources/libraries/robot/ip/geneve.robot
|
| Force Tags | 2_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | ETH | IP4FWD | IP4BASE | UDP | ENCAP | GENEVE_L3MODE
| ... | BASE | GENEVE4_1TUN | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4--ethip4udpgeneve-1tun-ip4base
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | geneve4
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | geneve_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}= | ${50}
# IP settings
| ${dut1_if1_ip4}= | 20.0.0.1
| ${dut1_if2_ip4}= | 30.0.0.1
| ${tg_if1_ip4}= | 20.0.0.2
| ${tg_if2_ip4}= | 30.0.0.2
# GENEVE settings
| ${gen_mode}= | ${osi_layer}
| ${n_tunnels}= | ${1}
| &{gen_tunnel}=
| ... | local=1.1.1.2 | remote=1.1.1.1 | vni=${1}
| ... | src_ip=10.128.1.0 | dst_ip=10.0.1.0 | ip_mask=${24} | if_ip=11.0.1.2
# Traffic profile
| ${traffic_profile}= | trex-stl-2n-ethip4-geneve-${n_tunnels}t7.6.14.1. Local Template¶
[Cfg] DUT runs GENEVE ${gen_mode} mode configuration. 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 ${nic_driver}
And Initialize layer interface
And Initialize GENEVE L3 mode in circular topology
Then Find NDR and PDR intervals using optimized search
7.6.14.2. 64B-1c-ethip4–ethip4udpgeneve-1tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.14.3. 64B-2c-ethip4–ethip4udpgeneve-1tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.14.4. 64B-4c-ethip4–ethip4udpgeneve-1tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.14.5. 1518B-1c-ethip4–ethip4udpgeneve-1tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.14.6. 1518B-2c-ethip4–ethip4udpgeneve-1tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.14.7. 1518B-4c-ethip4–ethip4udpgeneve-1tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.14.8. 9000B-1c-ethip4–ethip4udpgeneve-1tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.14.9. 9000B-2c-ethip4–ethip4udpgeneve-1tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.14.10. 9000B-4c-ethip4–ethip4udpgeneve-1tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.14.11. IMIX-1c-ethip4–ethip4udpgeneve-1tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.14.12. IMIX-2c-ethip4–ethip4udpgeneve-1tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.14.13. IMIX-4c-ethip4–ethip4udpgeneve-1tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.15. 2n1l-10ge2p1x710-ethip4–ethip4udpgeneve-256tun-ip4base-ndrpdr suite¶
RFC2544: Pkt throughput GENEVE tunnel L3 mode performance test cases
[Top] Network Topologies: TG-DUT1-TG 2-node circular topology with single links between nodes.
[Enc] Packet Encapsulations: Eth-IPv4 between TG-if1 and DUT1-if1 and Eth-IPv4-UDP-GENEVE-Eth-IPv4 between DUT1-if2 and TG-if2 for IPv4 routing over GENEVE tunnels.
[Cfg] DUT configuration: DUT1 is configured with IPv4 routing over ${n_tunnels} GENEVE tunnels and ${${4} * ${n_tunnels}} static IPv4 /24 route entries. DUT1 is tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
Test packets are generated by TG on links to DUT1. TG traffic profile contains two L3 flow-groups (flow-group per direction, 1 flow 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, RFC791, RFC768 and RFC8926.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
| Resource | resources/libraries/robot/ip/geneve.robot
|
| Force Tags | 2_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | ETH | IP4FWD | IP4BASE | UDP | ENCAP | GENEVE_L3MODE
| ... | SCALE | GENEVE4_256TUN | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4--ethip4udpgeneve-256tun-ip4base
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | geneve4
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | geneve_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}= | ${50}
# IP settings
| ${dut1_if1_ip4}= | 20.0.0.1
| ${dut1_if2_ip4}= | 30.0.0.1
| ${tg_if1_ip4}= | 20.0.0.2
| ${tg_if2_ip4}= | 30.0.0.2
# GENEVE settings
| ${gen_mode}= | ${osi_layer}
| ${n_tunnels}= | ${256}
| &{gen_tunnel}=
| ... | local=1.1.1.2 | remote=1.1.1.1 | vni=${1}
| ... | src_ip=10.128.1.0 | dst_ip=10.0.1.0 | ip_mask=${24} | if_ip=11.0.1.2
# Traffic profile
| ${traffic_profile}= | trex-stl-2n-ethip4-geneve-${n_tunnels}t7.6.15.1. Local Template¶
[Cfg] DUT runs GENEVE ${gen_mode} mode configuration. 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 ${nic_driver}
And Initialize layer interface
And Initialize GENEVE L3 mode in circular topology
Then Find NDR and PDR intervals using optimized search
7.6.15.2. 64B-1c-ethip4–ethip4udpgeneve-256tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.15.3. 64B-2c-ethip4–ethip4udpgeneve-256tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.15.4. 64B-4c-ethip4–ethip4udpgeneve-256tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.15.5. 1518B-1c-ethip4–ethip4udpgeneve-256tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.15.6. 1518B-2c-ethip4–ethip4udpgeneve-256tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.15.7. 1518B-4c-ethip4–ethip4udpgeneve-256tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.15.8. 9000B-1c-ethip4–ethip4udpgeneve-256tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.15.9. 9000B-2c-ethip4–ethip4udpgeneve-256tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.15.10. 9000B-4c-ethip4–ethip4udpgeneve-256tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.15.11. IMIX-1c-ethip4–ethip4udpgeneve-256tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.15.12. IMIX-2c-ethip4–ethip4udpgeneve-256tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.15.13. IMIX-4c-ethip4–ethip4udpgeneve-256tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.16. 2n1l-10ge2p1x710-ethip4–ethip4udpgeneve-4tun-ip4base-ndrpdr suite¶
RFC2544: Pkt throughput GENEVE tunnel L3 mode performance test cases
[Top] Network Topologies: TG-DUT1-TG 2-node circular topology with single links between nodes.
[Enc] Packet Encapsulations: Eth-IPv4 between TG-if1 and DUT1-if1 and Eth-IPv4-UDP-GENEVE-Eth-IPv4 between DUT1-if2 and TG-if2 for IPv4 routing over GENEVE tunnels.
[Cfg] DUT configuration: DUT1 is configured with IPv4 routing over ${n_tunnels} GENEVE tunnels and ${${4} * ${n_tunnels}} static IPv4 /24 route entries. DUT1 is tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
Test packets are generated by TG on links to DUT1. TG traffic profile contains two L3 flow-groups (flow-group per direction, 1 flow 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, RFC791, RFC768 and RFC8926.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
| Resource | resources/libraries/robot/ip/geneve.robot
|
| Force Tags | 2_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | ETH | IP4FWD | IP4BASE | UDP | ENCAP | GENEVE_L3MODE
| ... | SCALE | GENEVE4_4TUN | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4--ethip4udpgeneve-4tun-ip4base
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | geneve4
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | geneve_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}= | ${50}
# IP settings
| ${dut1_if1_ip4}= | 20.0.0.1
| ${dut1_if2_ip4}= | 30.0.0.1
| ${tg_if1_ip4}= | 20.0.0.2
| ${tg_if2_ip4}= | 30.0.0.2
# GENEVE settings
| ${gen_mode}= | ${osi_layer}
| ${n_tunnels}= | ${4}
| &{gen_tunnel}=
| ... | local=1.1.1.2 | remote=1.1.1.1 | vni=${1}
| ... | src_ip=10.128.1.0 | dst_ip=10.0.1.0 | ip_mask=${24} | if_ip=11.0.1.2
# Traffic profile
| ${traffic_profile}= | trex-stl-2n-ethip4-geneve-${n_tunnels}t7.6.16.1. Local Template¶
[Cfg] DUT runs GENEVE ${gen_mode} mode configuration. 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 ${nic_driver}
And Initialize layer interface
And Initialize GENEVE L3 mode in circular topology
Then Find NDR and PDR intervals using optimized search
7.6.16.2. 64B-1c-ethip4–ethip4udpgeneve-4tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.16.3. 64B-2c-ethip4–ethip4udpgeneve-4tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.16.4. 64B-4c-ethip4–ethip4udpgeneve-4tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.16.5. 1518B-1c-ethip4–ethip4udpgeneve-4tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.16.6. 1518B-2c-ethip4–ethip4udpgeneve-4tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.16.7. 1518B-4c-ethip4–ethip4udpgeneve-4tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.16.8. 9000B-1c-ethip4–ethip4udpgeneve-4tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.16.9. 9000B-2c-ethip4–ethip4udpgeneve-4tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.16.10. 9000B-4c-ethip4–ethip4udpgeneve-4tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.16.11. IMIX-1c-ethip4–ethip4udpgeneve-4tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.16.12. IMIX-2c-ethip4–ethip4udpgeneve-4tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.16.13. IMIX-4c-ethip4–ethip4udpgeneve-4tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}
7.6.17. 2n1l-10ge2p1x710-ethip4–ethip4udpgeneve-64tun-ip4base-ndrpdr suite¶
RFC2544: Pkt throughput GENEVE tunnel L3 mode performance test cases
[Top] Network Topologies: TG-DUT1-TG 2-node circular topology with single links between nodes.
[Enc] Packet Encapsulations: Eth-IPv4 between TG-if1 and DUT1-if1 and Eth-IPv4-UDP-GENEVE-Eth-IPv4 between DUT1-if2 and TG-if2 for IPv4 routing over GENEVE tunnels.
[Cfg] DUT configuration: DUT1 is configured with IPv4 routing over ${n_tunnels} GENEVE tunnels and ${${4} * ${n_tunnels}} static IPv4 /24 route entries. DUT1 is tested with ${nic_name}.
[Ver] TG verification: TG finds and reports throughput NDR (Non Drop Rate) with zero packet loss tolerance and 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 MLRsearch library.
Test packets are generated by TG on links to DUT1. TG traffic profile contains two L3 flow-groups (flow-group per direction, 1 flow 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, RFC791, RFC768 and RFC8926.
*** Settings ***
| Resource | resources/libraries/robot/shared/default.robot
| Resource | resources/libraries/robot/ip/geneve.robot
|
| Force Tags | 2_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | NDRPDR
| ... | NIC_Intel-X710 | ETH | IP4FWD | IP4BASE | UDP | ENCAP | GENEVE_L3MODE
| ... | SCALE | GENEVE4_64TUN | DRV_VFIO_PCI
| ... | RXQ_SIZE_0 | TXQ_SIZE_0
| ... | ethip4--ethip4udpgeneve-64tun-ip4base
|
| Suite Setup | Setup suite topology interfaces | performance
| Suite Teardown | Tear down suite | performance
| Test Setup | Setup test | performance
| Test Teardown | Tear down test | performance | geneve4
|
| Test Template | Local Template
|
| *** Variables ***
| @{plugins_to_enable}= | dpdk_plugin.so | perfmon_plugin.so | geneve_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}= | ${50}
# IP settings
| ${dut1_if1_ip4}= | 20.0.0.1
| ${dut1_if2_ip4}= | 30.0.0.1
| ${tg_if1_ip4}= | 20.0.0.2
| ${tg_if2_ip4}= | 30.0.0.2
# GENEVE settings
| ${gen_mode}= | ${osi_layer}
| ${n_tunnels}= | ${64}
| &{gen_tunnel}=
| ... | local=1.1.1.2 | remote=1.1.1.1 | vni=${1}
| ... | src_ip=10.128.1.0 | dst_ip=10.0.1.0 | ip_mask=${24} | if_ip=11.0.1.2
# Traffic profile
| ${traffic_profile}= | trex-stl-2n-ethip4-geneve-${n_tunnels}t7.6.17.1. Local Template¶
[Cfg] DUT runs GENEVE ${gen_mode} mode configuration. 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 ${nic_driver}
And Initialize layer interface
And Initialize GENEVE L3 mode in circular topology
Then Find NDR and PDR intervals using optimized search
7.6.17.2. 64B-1c-ethip4–ethip4udpgeneve-64tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${1}
7.6.17.3. 64B-2c-ethip4–ethip4udpgeneve-64tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${2}
7.6.17.4. 64B-4c-ethip4–ethip4udpgeneve-64tun-ip4base-ndrpdr¶
frame_size=${64} phy_cores=${4}
7.6.17.5. 1518B-1c-ethip4–ethip4udpgeneve-64tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${1}
7.6.17.6. 1518B-2c-ethip4–ethip4udpgeneve-64tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${2}
7.6.17.7. 1518B-4c-ethip4–ethip4udpgeneve-64tun-ip4base-ndrpdr¶
frame_size=${1518} phy_cores=${4}
7.6.17.8. 9000B-1c-ethip4–ethip4udpgeneve-64tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${1}
7.6.17.9. 9000B-2c-ethip4–ethip4udpgeneve-64tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${2}
7.6.17.10. 9000B-4c-ethip4–ethip4udpgeneve-64tun-ip4base-ndrpdr¶
frame_size=${9000} phy_cores=${4}
7.6.17.11. IMIX-1c-ethip4–ethip4udpgeneve-64tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${1}
7.6.17.12. IMIX-2c-ethip4–ethip4udpgeneve-64tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${2}
7.6.17.13. IMIX-4c-ethip4–ethip4udpgeneve-64tun-ip4base-ndrpdr¶
frame_size=IMIX_v4_1 phy_cores=${4}