diff options
Diffstat (limited to 'tests/perf')
| -rw-r--r-- | tests/perf/Long_IPv4_Fib_200k_Intel-X520-DA2.robot | 492 | ||||
| -rw-r--r-- | tests/perf/Long_IPv4_Fib_20k_Intel-X520-DA2.robot | 492 | ||||
| -rw-r--r-- | tests/perf/Long_IPv4_Fib_2M_Intel-X520-DA2.robot | 492 | ||||
| -rw-r--r-- | tests/perf/Short_IPv4_Fib_200k_Intel-X520-DA2.robot | 213 | ||||
| -rw-r--r-- | tests/perf/Short_IPv4_Fib_20k_Intel-X520-DA2.robot | 213 | ||||
| -rw-r--r-- | tests/perf/Short_IPv4_Fib_2M_Intel-X520-DA2.robot | 213 |
6 files changed, 2115 insertions, 0 deletions
diff --git a/tests/perf/Long_IPv4_Fib_200k_Intel-X520-DA2.robot b/tests/perf/Long_IPv4_Fib_200k_Intel-X520-DA2.robot new file mode 100644 index 0000000000..07634aa60f --- /dev/null +++ b/tests/perf/Long_IPv4_Fib_200k_Intel-X520-DA2.robot @@ -0,0 +1,492 @@ +# Copyright (c) 2016 Cisco and/or its affiliates. +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at: +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +*** Settings *** +| Resource | resources/libraries/robot/performance.robot +| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PERFTEST_LONG +| ... | FIB_200K | NIC_Intel-X520-DA2 +| Suite Setup | 3-node Performance Suite Setup with DUT's NIC model +| ... | L3 | Intel-X520-DA2 +| Suite Teardown | 3-node Performance Suite Teardown +| Test Setup | Setup all DUTs before test +| Test Teardown | Run Keyword | Remove startup configuration of VPP from all DUTs +| Documentation | *RFC2544: Pkt throughput IPv4 routing test cases* +| ... +| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology +| ... | with single links between nodes. +| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing. +| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4 +| ... | routing and 2x100k static IPv4 /32 route entries. DUT1 and DUT2 tested +| ... | with 2p10GE NIC X520 Niantic by Intel. +| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop +| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop +| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage +| ... | of packets transmitted. NDR and PDR are discovered for different +| ... | Ethernet L2 frame sizes using either binary search or linear search +| ... | algorithms with configured starting rate and final step that determines +| ... | throughput measurement resolution. Test packets are generated by TG on +| ... | links to DUTs. TG traffic profile contains two L3 flow-groups +| ... | (flow-group per direction, 100k flows per flow-group) with all packets +| ... | containing Ethernet header, IPv4 header with IP protocol=61 and +| ... | generated payload. MAC addresses are matching MAC addresses of the TG +| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field +| ... | is applied to both streams. +| ... | *[Ref] Applicable standard specifications:* RFC2544. + +*** Variables *** +| ${rts_per_flow}= | 100000 + +*** Test Cases *** +| TC01: 64B NDR binary search - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC02: 64B PDR binary search - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC03: 1518B NDR binary search - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC04: 1518B PDR binary search - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC05: 9000B NDR binary search - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC06: 9000B PDR binary search - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC07: 64B NDR binary search - DUT IPv4 Fib 2x100k - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC08: 64B PDR binary search - DUT IPv4 Fib 2x100k - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC09: 1518B NDR binary search - DUT IPv4 Fib 2x100k - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC10: 1518B PDR binary search - DUT IPv4 Fib 2x100k - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC11: 9000B NDR binary search - DUT IPv4 Fib 2x100k - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC12: 9000B PDR binary search - DUT IPv4 Fib 2x100k - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC13: 64B NDR binary search - DUT IPv4 Fib 2x100k - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC14: 64B PDR binary search - DUT IPv4 Fib 2x100k - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC15: 1518B NDR binary search - DUT IPv4 Fib 2x100k - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC16: 1518B PDR binary search - DUT IPv4 Fib 2x100k - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC17: 9000B NDR binary search - DUT IPv4 Fib 2x100k - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC18: 9000B PDR binary search - DUT IPv4 Fib 2x100k - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} diff --git a/tests/perf/Long_IPv4_Fib_20k_Intel-X520-DA2.robot b/tests/perf/Long_IPv4_Fib_20k_Intel-X520-DA2.robot new file mode 100644 index 0000000000..543f1d9123 --- /dev/null +++ b/tests/perf/Long_IPv4_Fib_20k_Intel-X520-DA2.robot @@ -0,0 +1,492 @@ +# Copyright (c) 2016 Cisco and/or its affiliates. +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at: +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +*** Settings *** +| Resource | resources/libraries/robot/performance.robot +| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PERFTEST_LONG +| ... | FIB_20K | NIC_Intel-X520-DA2 +| Suite Setup | 3-node Performance Suite Setup with DUT's NIC model +| ... | L3 | Intel-X520-DA2 +| Suite Teardown | 3-node Performance Suite Teardown +| Test Setup | Setup all DUTs before test +| Test Teardown | Run Keyword | Remove startup configuration of VPP from all DUTs +| Documentation | *RFC2544: Pkt throughput IPv4 routing test cases* +| ... +| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology +| ... | with single links between nodes. +| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing. +| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4 +| ... | routing and 2x10k static IPv4 /32 route entries. DUT1 and DUT2 tested +| ... | with 2p10GE NIC X520 Niantic by Intel. +| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop +| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop +| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage +| ... | of packets transmitted. NDR and PDR are discovered for different +| ... | Ethernet L2 frame sizes using either binary search or linear search +| ... | algorithms with configured starting rate and final step that determines +| ... | throughput measurement resolution. Test packets are generated by TG on +| ... | links to DUTs. TG traffic profile contains two L3 flow-groups +| ... | (flow-group per direction, 10k flows per flow-group) with all packets +| ... | containing Ethernet header, IPv4 header with IP protocol=61 and +| ... | generated payload. MAC addresses are matching MAC addresses of the TG +| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field +| ... | is applied to both streams. +| ... | *[Ref] Applicable standard specifications:* RFC2544. + +*** Variables *** +| ${rts_per_flow}= | 10000 + +*** Test Cases *** +| TC01: 64B NDR binary search - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC02: 64B PDR binary search - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC03: 1518B NDR binary search - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC04: 1518B PDR binary search - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC05: 9000B NDR binary search - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC06: 9000B PDR binary search - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC07: 64B NDR binary search - DUT IPv4 Fib 2x10k - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC08: 64B PDR binary search - DUT IPv4 Fib 2x10k - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC09: 1518B NDR binary search - DUT IPv4 Fib 2x10k - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC10: 1518B PDR binary search - DUT IPv4 Fib 2x10k - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC11: 9000B NDR binary search - DUT IPv4 Fib 2x10k - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC12: 9000B PDR binary search - DUT IPv4 Fib 2x10k - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC13: 64B NDR binary search - DUT IPv4 Fib 2x10k - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC14: 64B PDR binary search - DUT IPv4 Fib 2x10k - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC15: 1518B NDR binary search - DUT IPv4 Fib 2x10k - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC16: 1518B PDR binary search - DUT IPv4 Fib 2x10k - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC17: 9000B NDR binary search - DUT IPv4 Fib 2x10k - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC18: 9000B PDR binary search - DUT IPv4 Fib 2x10k - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} diff --git a/tests/perf/Long_IPv4_Fib_2M_Intel-X520-DA2.robot b/tests/perf/Long_IPv4_Fib_2M_Intel-X520-DA2.robot new file mode 100644 index 0000000000..1944ba4684 --- /dev/null +++ b/tests/perf/Long_IPv4_Fib_2M_Intel-X520-DA2.robot @@ -0,0 +1,492 @@ +# Copyright (c) 2016 Cisco and/or its affiliates. +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at: +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +*** Settings *** +| Resource | resources/libraries/robot/performance.robot +| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PERFTEST_LONG +| ... | FIB_2M | NIC_Intel-X520-DA2 +| Suite Setup | 3-node Performance Suite Setup with DUT's NIC model +| ... | L3 | Intel-X520-DA2 +| Suite Teardown | 3-node Performance Suite Teardown +| Test Setup | Setup all DUTs before test +| Test Teardown | Run Keyword | Remove startup configuration of VPP from all DUTs +| Documentation | *RFC2544: Pkt throughput IPv4 routing test cases* +| ... +| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology +| ... | with single links between nodes. +| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing. +| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4 +| ... | routing and 2x1M static IPv4 /32 route entries. DUT1 and DUT2 tested +| ... | with 2p10GE NIC X520 Niantic by Intel. +| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop +| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop +| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage +| ... | of packets transmitted. NDR and PDR are discovered for different +| ... | Ethernet L2 frame sizes using either binary search or linear search +| ... | algorithms with configured starting rate and final step that determines +| ... | throughput measurement resolution. Test packets are generated by TG on +| ... | links to DUTs. TG traffic profile contains two L3 flow-groups +| ... | (flow-group per direction, 1M flows per flow-group) with all packets +| ... | containing Ethernet header, IPv4 header with IP protocol=61 and +| ... | generated payload. MAC addresses are matching MAC addresses of the TG +| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field +| ... | is applied to both streams. +| ... | *[Ref] Applicable standard specifications:* RFC2544. + +*** Variables *** +| ${rts_per_flow}= | 1000000 + +*** Test Cases *** +| TC01: 64B NDR binary search - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC02: 64B PDR binary search - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC03: 1518B NDR binary search - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC04: 1518B PDR binary search - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC05: 9000B NDR binary search - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC06: 9000B PDR binary search - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC07: 64B NDR binary search - DUT IPv4 Fib 2x1M - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC08: 64B PDR binary search - DUT IPv4 Fib 2x1M - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC09: 1518B NDR binary search - DUT IPv4 Fib 2x1M - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC10: 1518B PDR binary search - DUT IPv4 Fib 2x1M - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC11: 9000B NDR binary search - DUT IPv4 Fib 2x1M - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC12: 9000B PDR binary search - DUT IPv4 Fib 2x1M - 2threads 2cores 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC13: 64B NDR binary search - DUT IPv4 Fib 2x1M - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC14: 64B PDR binary search - DUT IPv4 Fib 2x1M - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames +| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR +| | ${framesize}= | Set Variable | 64 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC15: 1518B NDR binary search - DUT IPv4 Fib 2x1M - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC16: 1518B PDR binary search - DUT IPv4 Fib 2x1M - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames +| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 1518 +| | ${min_rate}= | Set Variable | 10000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} + +| TC17: 9000B NDR binary search - DUT IPv4 Fib 2x1M - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC18: 9000B PDR binary search - DUT IPv4 Fib 2x1M - 4threads 4cores 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames +| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR | SKIP_PATCH +| | ${framesize}= | Set Variable | 9000 +| | ${min_rate}= | Set Variable | 5000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B} +| | ${binary_min}= | Set Variable | ${min_rate} +| | ${binary_max}= | Set Variable | ${max_rate} +| | ${threshold}= | Set Variable | ${min_rate} +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} +| | ... | 3-node-IPv4-dst-${rts_per_flow} +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} diff --git a/tests/perf/Short_IPv4_Fib_200k_Intel-X520-DA2.robot b/tests/perf/Short_IPv4_Fib_200k_Intel-X520-DA2.robot new file mode 100644 index 0000000000..845036bcfd --- /dev/null +++ b/tests/perf/Short_IPv4_Fib_200k_Intel-X520-DA2.robot @@ -0,0 +1,213 @@ +# Copyright (c) 2016 Cisco and/or its affiliates. +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at: +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +*** Settings *** +| Resource | resources/libraries/robot/performance.robot +| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PERFTEST_SHORT +| ... | FIB_200K | NIC_Intel-X520-DA2 +| Suite Setup | 3-node Performance Suite Setup with DUT's NIC model +| ... | L3 | Intel-X520-DA2 +| Suite Teardown | 3-node Performance Suite Teardown +| Test Setup | Setup all DUTs before test +| Test Teardown | Run Keyword | Remove startup configuration of VPP from all DUTs +| Documentation | *Reference NDR throughput IPv4 routing verify test cases* +| ... +| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology +| ... | with single links between nodes. +| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing. +| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4 +| ... | routing and 2x100k static IPv4 /32 route entries. DUT1 and DUT2 tested +| ... | with 2p10GE NIC X520 Niantic by Intel. +| ... | *[Ver] TG verification:* In short performance tests, TG verifies +| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet +| ... | loss tolerance. Ref-NDR value is periodically updated acording to +| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long +| ... | performance tests for the same DUT confiiguration. Test packets are +| ... | generated by TG on links to DUTs. TG traffic profile contains two L3 +| ... | flow-groups (flow-group per direction, 100k flows per flow-group) with +| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61 +| ... | and generated payload. Incrementing of IP.dst (IPv4 destination address) +| ... | field is applied to both streams. +| ... | *[Ref] Applicable standard specifications:* RFC2544. + +*** Variables *** +| ${rts_per_flow}= | 100000 + +*** Test Cases *** +| TC01: Verify 64B ref-NDR at 2x 3.5Mpps - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR +| | ${framesize}= | Set Variable | 64 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 3.5mpps +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC02: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR +| | ${framesize}= | Set Variable | 1518 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 812743pps +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC03: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR +| | ${framesize}= | Set Variable | 9000 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 138580pps +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC04: Verify 64B ref-NDR at 2x 7.5Mpps - DUT IPv4 Fib 2x100k - 2threads, 2cores, 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 64 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 7.5mpps +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC05: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x100k - 2threads, 2cores, 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 1518 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 812743pps +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC06: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x100k - 2threads, 2cores, 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 9000 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 138580pps +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC07: Verify 64B ref-NDR at 2x 10.0Mpps - DUT IPv4 Fib 2x100k - 4threads, 4cores, 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 64 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 64 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 10.0mpps +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC08: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x100k - 4threads, 4cores, 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 1518 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 812743pps +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC09: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x100k - 4threads, 4cores, 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 9000 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 138580pps +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} diff --git a/tests/perf/Short_IPv4_Fib_20k_Intel-X520-DA2.robot b/tests/perf/Short_IPv4_Fib_20k_Intel-X520-DA2.robot new file mode 100644 index 0000000000..9ac8c81536 --- /dev/null +++ b/tests/perf/Short_IPv4_Fib_20k_Intel-X520-DA2.robot @@ -0,0 +1,213 @@ +# Copyright (c) 2016 Cisco and/or its affiliates. +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at: +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +*** Settings *** +| Resource | resources/libraries/robot/performance.robot +| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PERFTEST_SHORT +| ... | FIB_20K | NIC_Intel-X520-DA2 +| Suite Setup | 3-node Performance Suite Setup with DUT's NIC model +| ... | L3 | Intel-X520-DA2 +| Suite Teardown | 3-node Performance Suite Teardown +| Test Setup | Setup all DUTs before test +| Test Teardown | Run Keyword | Remove startup configuration of VPP from all DUTs +| Documentation | *Reference NDR throughput IPv4 routing verify test cases* +| ... +| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology +| ... | with single links between nodes. +| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing. +| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4 +| ... | routing and 2x10k static IPv4 /32 route entries. DUT1 and DUT2 tested +| ... | with 2p10GE NIC X520 Niantic by Intel. +| ... | *[Ver] TG verification:* In short performance tests, TG verifies +| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet +| ... | loss tolerance. Ref-NDR value is periodically updated acording to +| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long +| ... | performance tests for the same DUT confiiguration. Test packets are +| ... | generated by TG on links to DUTs. TG traffic profile contains two L3 +| ... | flow-groups (flow-group per direction, 10k flows per flow-group) with +| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61 +| ... | and generated payload. Incrementing of IP.dst (IPv4 destination address) +| ... | field is applied to both streams. +| ... | *[Ref] Applicable standard specifications:* RFC2544. + +*** Variables *** +| ${rts_per_flow}= | 10000 + +*** Test Cases *** +| TC01: Verify 64B ref-NDR at 2x 3.5Mpps - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR +| | ${framesize}= | Set Variable | 64 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 3.5mpps +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC02: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR +| | ${framesize}= | Set Variable | 1518 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 812743pps +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC03: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR +| | ${framesize}= | Set Variable | 9000 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 138580pps +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC04: Verify 64B ref-NDR at 2x 7.5Mpps - DUT IPv4 Fib 2x10k - 2threads, 2cores, 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 64 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 7.5mpps +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC05: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x10k - 2threads, 2cores, 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 1518 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 812743pps +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC06: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x10k - 2threads, 2cores, 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 9000 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 138580pps +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC07: Verify 64B ref-NDR at 2x 10.0Mpps - DUT IPv4 Fib 2x10k - 4threads, 4cores, 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 64 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 64 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 10.0mpps +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC08: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x10k - 4threads, 4cores, 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 1518 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 812743pps +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC09: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x10k - 4threads, 4cores, 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 9000 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 138580pps +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} diff --git a/tests/perf/Short_IPv4_Fib_2M_Intel-X520-DA2.robot b/tests/perf/Short_IPv4_Fib_2M_Intel-X520-DA2.robot new file mode 100644 index 0000000000..7caa56a5b5 --- /dev/null +++ b/tests/perf/Short_IPv4_Fib_2M_Intel-X520-DA2.robot @@ -0,0 +1,213 @@ +# Copyright (c) 2016 Cisco and/or its affiliates. +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at: +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +*** Settings *** +| Resource | resources/libraries/robot/performance.robot +| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PERFTEST_SHORT +| ... | FIB_2M | NIC_Intel-X520-DA2 +| Suite Setup | 3-node Performance Suite Setup with DUT's NIC model +| ... | L3 | Intel-X520-DA2 +| Suite Teardown | 3-node Performance Suite Teardown +| Test Setup | Setup all DUTs before test +| Test Teardown | Run Keyword | Remove startup configuration of VPP from all DUTs +| Documentation | *Reference NDR throughput IPv4 routing verify test cases* +| ... +| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology +| ... | with single links between nodes. +| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing. +| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4 +| ... | routing and 2x1M static IPv4 /32 route entries. DUT1 and DUT2 tested +| ... | with 2p10GE NIC X520 Niantic by Intel. +| ... | *[Ver] TG verification:* In short performance tests, TG verifies +| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet +| ... | loss tolerance. Ref-NDR value is periodically updated acording to +| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long +| ... | performance tests for the same DUT confiiguration. Test packets are +| ... | generated by TG on links to DUTs. TG traffic profile contains two L3 +| ... | flow-groups (flow-group per direction, 1M flows per flow-group) with +| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61 +| ... | and generated payload. Incrementing of IP.dst (IPv4 destination address) +| ... | field is applied to both streams. +| ... | *[Ref] Applicable standard specifications:* RFC2544. + +*** Variables *** +| ${rts_per_flow}= | 1000000 + +*** Test Cases *** +| TC01: Verify 64B ref-NDR at 2x 3.5Mpps - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR +| | ${framesize}= | Set Variable | 64 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 3.5mpps +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC02: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq +| | [Documentation] +| | ... | Verify ref-NDR for 1518 Byte frames using single trial throughput +| | ... | test. DUT runs IPv4 routing config with 1 thread, 1 phy core, 1 +| | ... | receive queue per NIC port. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR +| | ${framesize}= | Set Variable | 1518 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 812743pps +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC03: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR +| | ${framesize}= | Set Variable | 9000 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 138580pps +| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC04: Verify 64B ref-NDR at 2x 7.5Mpps - DUT IPv4 Fib 2x1M - 2threads, 2cores, 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 64 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 7.5mpps +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC05: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x1M - 2threads, 2cores, 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 1518 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 812743pps +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC06: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x1M - 2threads, 2cores, 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \ +| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 9000 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 138580pps +| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC07: Verify 64B ref-NDR at 2x 10.0Mpps - DUT IPv4 Fib 2x1M - 4threads, 4cores, 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 64 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 64 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 10.0mpps +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC08: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x1M - 4threads, 4cores, 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 1518 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 812743pps +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} + +| TC09: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x1M - 4threads, 4cores, 2rxq +| | [Documentation] +| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \ +| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte +| | ... | frames using single trial throughput test. +| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR +| | ${framesize}= | Set Variable | 9000 +| | ${duration}= | Set Variable | 10 +| | ${rate}= | Set Variable | 138580pps +| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs +| | And Add PCI devices to DUTs from 3-node single link topology +| | And Apply startup configuration on all VPP DUTs +| | And Scale IPv4 forwarding initialized in a 3-node circular topology +| | ... | ${rts_per_flow} +| | Then Traffic should pass with no loss | ${duration} | ${rate} +| | ... | ${framesize} +| | ... | 3-node-IPv4-dst-${rts_per_flow} |