diff options
author | Peter Mikus <pmikus@cisco.com> | 2018-07-10 06:10:41 +0000 |
---|---|---|
committer | Peter Mikus <pmikus@cisco.com> | 2018-07-13 09:11:04 +0000 |
commit | ad71afc5baa8008bd5d4ea26f51a71eefe3fbe62 (patch) | |
tree | 3b1c24ab540b5f57ee4b3592368191bc1f227cc7 /tests/vpp/perf/ip4/10ge2p1x710-ethip4-ip4scale20k-mrr.robot | |
parent | 338b195c0d7274b921eb18944d27ed1b4cb3739f (diff) |
Refactor perf test cases
- Refactor Performance TCs due to changes in Hyperthreading and RXQ allocations.
- Simplify the structure of suite to avoid large refactors in future.
Change-Id: Ifb1b9719391f26745f075a900bd9fb39b6cb809c
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Diffstat (limited to 'tests/vpp/perf/ip4/10ge2p1x710-ethip4-ip4scale20k-mrr.robot')
-rw-r--r-- | tests/vpp/perf/ip4/10ge2p1x710-ethip4-ip4scale20k-mrr.robot | 144 |
1 files changed, 26 insertions, 118 deletions
diff --git a/tests/vpp/perf/ip4/10ge2p1x710-ethip4-ip4scale20k-mrr.robot b/tests/vpp/perf/ip4/10ge2p1x710-ethip4-ip4scale20k-mrr.robot index 3436d0ab3d..95767a4b00 100644 --- a/tests/vpp/perf/ip4/10ge2p1x710-ethip4-ip4scale20k-mrr.robot +++ b/tests/vpp/perf/ip4/10ge2p1x710-ethip4-ip4scale20k-mrr.robot @@ -25,6 +25,8 @@ | ... | Test Teardown | Tear down performance mrr test | ... +| Test Template | Local template +| ... | Documentation | *Raw results IPv4 routing test cases* | ... | ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology @@ -50,22 +52,27 @@ | ${traffic_profile}= | trex-sl-3n-ethip4-ip4dst${rts_per_flow} *** Keywords *** -| Check RR for ethip4-ip4scale20k +| Local template | | ... | | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config with ${wt} thread(s), ${wt}\ -| | ... | phy core(s), ${rxq} receive queue(s) per NIC port. -| | ... | [Ver] Measure MaxReceivedRate for ${framesize} frames using single\ +| | ... | [Cfg] DUT runs IPv4 routing config. +| | ... | Each DUT uses ${phy_cores} physical core(s) for worker threads. +| | ... | [Ver] Measure MaxReceivedRate for ${framesize}B frames using single\ | | ... | trial throughput test. | | ... -| | [Arguments] | ${wt} | ${rxq} | ${framesize} +| | ... | *Arguments:* +| | ... | - framesize - Framesize in Bytes in integer or string (IMIX_v4_1). +| | ... | Type: integer, string +| | ... | - phy_cores - Number of physical cores. Type: integer +| | ... | - rxq - Number of RX queues, default value: ${None}. Type: integer +| | ... +| | [Arguments] | ${phy_cores} | ${framesize} | ${rxq}=${None} | | ... -| | # Test Variables required for test execution and test teardown | | Set Test Variable | ${framesize} | | ${get_framesize}= | Get Frame Size | ${framesize} | | ${max_rate}= | Calculate pps | ${s_limit} | ${framesize} | | ... -| | Given Add '${wt}' worker threads and '${rxq}' rxqueues in 3-node single-link circular topology +| | Given Add worker threads and rxqueues to all DUTs | ${phy_cores} | ${rxq} | | And Add PCI devices to all DUTs | | And Run Keyword If | ${get_framesize} < ${1522} | | ... | Add no multi seg to all DUTs @@ -77,148 +84,49 @@ *** Test Cases *** | tc01-64B-1t1c-ethip4-ip4scale20k-mrr -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config with 1 phy core, \ -| | ... | 1 receive queue per NIC port. -| | ... | [Ver] Measure MaxReceivedRate for 64B frames using single trial\ -| | ... | throughput test. -| | ... | | [Tags] | 64B | 1C -| | ... -| | [Template] | Check RR for ethip4-ip4scale20k -| | wt=1 | rxq=1 | framesize=${64} +| | phy_cores=${1} | framesize=${64} | tc02-1518B-1t1c-ethip4-ip4scale20k-mrr -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config with 1 phy core, \ -| | ... | 1 receive queue per NIC port. -| | ... | [Ver] Measure MaxReceivedRate for 1518B frames using single\ -| | ... | trial throughput test. -| | ... | | [Tags] | 1518B | 1C -| | ... -| | [Template] | Check RR for ethip4-ip4scale20k -| | wt=1 | rxq=1 | framesize=${1518} +| | phy_cores=${1} | framesize=${1518} | tc03-9000B-1t1c-ethip4-ip4scale20k-mrr -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config with 1 phy core, \ -| | ... | 1 receive queue per NIC port. -| | ... | [Ver] Measure MaxReceivedRate for 9000B frames using single\ -| | ... | trial throughput test. -| | ... | | [Tags] | 9000B | 1C -| | ... -| | [Template] | Check RR for ethip4-ip4scale20k -| | wt=1 | rxq=1 | framesize=${9000} +| | phy_cores=${1} | framesize=${9000} | tc04-IMIX-1t1c-ethip4-ip4scale20k-mrr -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config with \ -| | ... | 1 phy core, 1 receive queue per NIC port. -| | ... | [Ver] Measure MaxReceivedRate for IMIX_v4_1 frames using single trial\ -| | ... | throughput test. -| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B) -| | ... | | [Tags] | IMIX | 1C -| | ... -| | [Template] | Check RR for ethip4-ip4scale20k -| | wt=1 | rxq=1 | framesize=IMIX_v4_1 +| | phy_cores=${1} | framesize=IMIX_v4_1 | tc05-64B-2t2c-ethip4-ip4scale20k-mrr -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config with 2 phy cores, \ -| | ... | 1 receive queue per NIC port. -| | ... | [Ver] Measure MaxReceivedRate for 64B frames using single trial\ -| | ... | throughput test. -| | ... | | [Tags] | 64B | 2C -| | ... -| | [Template] | Check RR for ethip4-ip4scale20k -| | wt=2 | rxq=1 | framesize=${64} +| | phy_cores=${2} | framesize=${64} | tc06-1518B-2t2c-ethip4-ip4scale20k-mrr -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config with 2 phy cores, \ -| | ... | 1 receive queue per NIC port. -| | ... | [Ver] Measure MaxReceivedRate for 1518B frames using single\ -| | ... | trial throughput test. -| | ... | | [Tags] | 1518B | 2C -| | ... -| | [Template] | Check RR for ethip4-ip4scale20k -| | wt=2 | rxq=1 | framesize=${1518} +| | phy_cores=${2} | framesize=${1518} | tc07-9000B-2t2c-ethip4-ip4scale20k-mrr -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config with 2 phy cores, \ -| | ... | 1 receive queue per NIC port. -| | ... | [Ver] Measure MaxReceivedRate for 9000B frames using single\ -| | ... | trial throughput test. -| | ... | | [Tags] | 9000B | 2C -| | ... -| | [Template] | Check RR for ethip4-ip4scale20k -| | wt=2 | rxq=1 | framesize=${9000} +| | phy_cores=${2} | framesize=${9000} | tc08-IMIX-2t2c-ethip4-ip4scale20k-mrr -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config with \ -| | ... | 2 phy cores, 1 receive queue per NIC port. -| | ... | [Ver] Measure MaxReceivedRate for IMIX_v4_1 frames using single trial\ -| | ... | throughput test. -| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B) -| | ... | | [Tags] | IMIX | 2C -| | ... -| | [Template] | Check RR for ethip4-ip4scale20k -| | wt=2 | rxq=1 | framesize=IMIX_v4_1 +| | phy_cores=${2} | framesize=IMIX_v4_1 | tc09-64B-4t4c-ethip4-ip4scale20k-mrr -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config with 4 phy cores, \ -| | ... | 2 receive queues per NIC port. -| | ... | [Ver] Measure MaxReceivedRate for 64B frames using single trial\ -| | ... | throughput test. -| | ... | | [Tags] | 64B | 4C -| | ... -| | [Template] | Check RR for ethip4-ip4scale20k -| | wt=4 | rxq=2 | framesize=${64} +| | phy_cores=${4} | framesize=${64} | tc10-1518B-4t4c-ethip4-ip4scale20k-mrr -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config with 4 phy cores, \ -| | ... | 2 receive queues per NIC port. -| | ... | [Ver] Measure MaxReceivedRate for 1518B frames using single\ -| | ... | trial throughput test. -| | ... | | [Tags] | 1518B | 4C -| | ... -| | [Template] | Check RR for ethip4-ip4scale20k -| | wt=4 | rxq=2 | framesize=${1518} +| | phy_cores=${4} | framesize=${1518} | tc11-9000B-4t4c-ethip4-ip4scale20k-mrr -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config with 4 phy cores, \ -| | ... | 2 receive queues per NIC port. -| | ... | [Ver] Measure MaxReceivedRate for 9000B frames using single\ -| | ... | trial throughput test. -| | ... | | [Tags] | 9000B | 4C -| | ... -| | [Template] | Check RR for ethip4-ip4scale20k -| | wt=4 | rxq=2 | framesize=${9000} +| | phy_cores=${4} | framesize=${9000} | tc12-IMIX-4t4c-ethip4-ip4scale20k-mrr -| | [Documentation] -| | ... | [Cfg] DUT runs IPv4 routing config with \ -| | ... | 4 phy cores, 2 receive queue per NIC port. -| | ... | [Ver] Measure MaxReceivedRate for IMIX_v4_1 frames using single trial\ -| | ... | throughput test. -| | ... | IMIX_v4_1 = (28x64B; 16x570B; 4x1518B) -| | ... | | [Tags] | IMIX | 4C -| | ... -| | [Template] | Check RR for ethip4-ip4scale20k -| | wt=4 | rxq=2 | framesize=IMIX_v4_1 +| | phy_cores=${4} | framesize=IMIX_v4_1 |