diff options
author | Miroslav Miklus <mmiklus@cisco.com> | 2016-07-20 17:57:57 +0200 |
---|---|---|
committer | Peter Mikus <pmikus@cisco.com> | 2016-07-28 05:22:19 +0000 |
commit | 8825ed229f1d1fbd12d0cec0a128f42f22ae11fe (patch) | |
tree | 221d3bbd2115607f639898db73eb224cdb79a094 /tests/perf | |
parent | 9d052c6e49788c6c8240d2599812f737e83469b4 (diff) |
CSIT-136 IMIX T-REX driver
Change-Id: Ife7c543d07bcac4bad82894eea0f4e0e5a8db4bb
Signed-off-by: Miroslav Miklus <mmiklus@cisco.com>
Diffstat (limited to 'tests/perf')
-rw-r--r-- | tests/perf/Long_Bridge_Domain_Intel-X520-DA2.robot | 47 | ||||
-rw-r--r-- | tests/perf/Long_IPv4_Intel-X520-DA2.robot | 49 | ||||
-rw-r--r-- | tests/perf/Long_Xconnect_Intel-X520-DA2.robot | 48 |
3 files changed, 144 insertions, 0 deletions
diff --git a/tests/perf/Long_Bridge_Domain_Intel-X520-DA2.robot b/tests/perf/Long_Bridge_Domain_Intel-X520-DA2.robot index 9d3843b5b5..549033fbca 100644 --- a/tests/perf/Long_Bridge_Domain_Intel-X520-DA2.robot +++ b/tests/perf/Long_Bridge_Domain_Intel-X520-DA2.robot @@ -453,3 +453,50 @@ | | ... | ${glob_loss_acceptance} | | ... | ${glob_loss_acceptance_type} +| TC19: IMIX_v4_1 NDR binary search - DUT L2BD - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame size +| | ... | using binary search start at 10GE linerate, step 100kpps. +| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B) +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | IMIX_v4_1 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_IMIX_v4_1} +| | ${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 all PCI devices to all DUTs +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And L2 bridge domain initialized in a 3-node circular topology +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} | 3-node-bridge +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC20: IMIX_v4_1 PDR binary search - DUT L2BD - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs L2BD switching config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame size +| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%. +| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B) +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR +| | ${framesize}= | Set Variable | IMIX_v4_1 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_IMIX_v4_1} +| | ${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 all PCI devices to all DUTs +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And L2 bridge domain initialized in a 3-node circular topology +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} | 3-node-bridge +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} diff --git a/tests/perf/Long_IPv4_Intel-X520-DA2.robot b/tests/perf/Long_IPv4_Intel-X520-DA2.robot index d287373cbf..3236f93c79 100644 --- a/tests/perf/Long_IPv4_Intel-X520-DA2.robot +++ b/tests/perf/Long_IPv4_Intel-X520-DA2.robot @@ -456,3 +456,52 @@ | | ... | ${threshold} | | ... | ${glob_loss_acceptance} | | ... | ${glob_loss_acceptance_type} + +*** Test Cases *** +| TC19: IMIX_v4_1 NDR binary search - DUT IPv4 - 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 IMIX_v4_1 frame size +| | ... | using binary search start at 10GE linerate, step 100kpps. +| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B) +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | IMIX_v4_1 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_IMIX_v4_1} +| | ${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 all PCI devices to all DUTs +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And IPv4 forwarding initialized in a 3-node circular topology +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} | 3-node-IPv4 +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC20: 64B PDR binary search - DUT IPv4 - 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 IMIX_v4_1 frame size +| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%. +| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B) +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR +| | ${framesize}= | Set Variable | IMIX_v4_1 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_IMIX_v4_1} +| | ${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 all PCI devices to all DUTs +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And IPv4 forwarding initialized in a 3-node circular topology +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} | 3-node-IPv4 +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} diff --git a/tests/perf/Long_Xconnect_Intel-X520-DA2.robot b/tests/perf/Long_Xconnect_Intel-X520-DA2.robot index 8c7a3a6f89..110ff4f25c 100644 --- a/tests/perf/Long_Xconnect_Intel-X520-DA2.robot +++ b/tests/perf/Long_Xconnect_Intel-X520-DA2.robot @@ -452,3 +452,51 @@ | | ... | ${threshold} | | ... | ${glob_loss_acceptance} | | ... | ${glob_loss_acceptance_type} + +| TC19: IMIX_v4_1 NDR binary search - DUT L2XC - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find NDR for IMIX_v4_1 frame size +| | ... | using binary search start at 10GE linerate, step 100kpps. +| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B) +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH +| | ${framesize}= | Set Variable | IMIX_v4_1 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_IMIX_v4_1} +| | ${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 all PCI devices to all DUTs +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And L2 xconnect initialized in a 3-node circular topology +| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} | 3-node-xconnect +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} + +| TC20: IMIX_v4_1 PDR binary search - DUT L2XC - 1thread 1core 1rxq +| | [Documentation] +| | ... | [Cfg] DUT runs L2XC switching config with 1 thread, 1 phy core, \ +| | ... | 1 receive queue per NIC port. [Ver] Find PDR for IMIX_v4_1 frame size +| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%. +| | ... | IMIX_v4_1 = (28x64B;16x570B;4x1518B) +| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR +| | ${framesize}= | Set Variable | IMIX_v4_1 +| | ${min_rate}= | Set Variable | 100000 +| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_IMIX_v4_1} +| | ${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 all PCI devices to all DUTs +| | And Add No Multi Seg to all DUTs +| | And Apply startup configuration on all VPP DUTs +| | And L2 xconnect initialized in a 3-node circular topology +| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min} +| | ... | ${binary_max} | 3-node-xconnect +| | ... | ${min_rate} | ${max_rate} +| | ... | ${threshold} +| | ... | ${glob_loss_acceptance} +| | ... | ${glob_loss_acceptance_type} |