diff options
Diffstat (limited to 'docs/report/dpdk_performance_tests')
| -rw-r--r-- | docs/report/dpdk_performance_tests/csit_release_notes.rst | 52 | ||||
| -rw-r--r-- | docs/report/dpdk_performance_tests/overview.rst | 9 | ||||
| -rw-r--r-- | docs/report/dpdk_performance_tests/test_environment.rst | 17 |
3 files changed, 20 insertions, 58 deletions
diff --git a/docs/report/dpdk_performance_tests/csit_release_notes.rst b/docs/report/dpdk_performance_tests/csit_release_notes.rst index 0c207e5a52..ceacd89eb3 100644 --- a/docs/report/dpdk_performance_tests/csit_release_notes.rst +++ b/docs/report/dpdk_performance_tests/csit_release_notes.rst @@ -6,54 +6,19 @@ Changes in |csit-release| #. DPDK PERFORMANCE TESTS - - **Intel Xeon 2n-skx, 3n-skx testbeds**: Testpmd and L3fwd - performance test data is not included in this report version. - This is due to the lower performance and behaviour inconsistency - of these systems following the upgrade of processor microcode - packages (skx ucode 0x2000064) as part of updating Ubuntu 18.04 - LTS kernel version. Tested VPP and DPDK applications (L3fwd) are - affected. Skx test data will be added in subsequent maintenance - report version(s) once the issue is resolved. See - :ref:`dpdk_known_issues`. + - Refactor of CSIT initialization helper scripts for DPDK testpmd + and l3fwd. Removing obsolete parameters from DPDK testpmd and l3fwd + command line. - - **Intel Xeon 2n-clx testbeds**: DPDK performance test data is now - included in this report, after resolving the issue of lower - performance and behaviour inconsistency of these systems due to - the Linux kernel driven upgrade of processor microcode packages - to 0x500002c. The resolution is to use latest SuperMicro BIOS 3.2 - (for X11DPG-QT motherboards used) that upgrades processor - microcode to 0x500002c, AND NOT kernel provided ucode package as - it does put THE system into sub-optimal state. + - Fixed 9000B L2 packet size tests not passing for all NICs. + - Fixed DPDK compilation and test initialization of Mellanox NICs using + CONFIG_RTE_LIBRTE_MLX5_PMD=y in compile configuration. #. DPDK RELEASE VERSION CHANGE - |csit-release| tested |dpdk-release|, as used by |vpp-release|. -#. TEST ENVIRONMENT - - - **TRex Fortville NIC Performance**: Received FVL fix from Intel - resolving TRex low throughput issue. TRex per FVL NIC throughput - increased from ~27 Mpps to the nominal ~37 Mpps. For detail see - `CSIT-1503 <https://jira.fd.io/browse/CSIT-1503>`_ and `TRex-519 - <https://trex-tgn.cisco.com/youtrack/issue/trex-519>`_]. - - - **New Intel Xeon Cascadelake Testbeds**: Added performance tests - for 2-Node-Cascadelake (2n-clx) testbeds with x710, xxv710 and - cx556a-edat NIC cards. - -.. - // Alternative Note for 1st Bullet when bad microcode Skx, Clx results are published - - **Intel Xeon 2n-skx, 3n-skx and 2n-clx testbeds**: Testpmd and - L3fwd performance test data is included in this report version, - but it shows lower performance and behaviour inconsistency of - these systems following the upgrade of processor microcode - packages (skx ucode 0x2000064, clx ucode 0x500002c) as part of - updating Ubuntu 18.04 LTS kernel version. Tested VPP and DPDK - applications (L3fwd) are affected. Skx and Clx test data will be - corrected in subsequent maintenance report version(s) once the - issue is resolved. See :ref:`vpp_known_issues`. - .. _dpdk_known_issues: Known Issues @@ -64,7 +29,6 @@ List of known issues in |csit-release| for DPDK performance tests: +----+------------------------------------------+----------------------------------------------------------------------------------------------------------+ | # | JiraID | Issue Description | +====+==========================================+==========================================================================================================+ -| 1 | `CSIT-1675 | Intel Xeon 2n-skx, 3n-skx and 2n-clx testbeds behaviour and performance became inconsistent following | -| | <https://jira.fd.io/browse/CSIT-1675>`_ | the upgrade to the latest Ubuntu 18.04 LTS kernel version (4.15.0-72-generic) and associated microcode | -| | | packages (skx ucode 0x2000064, clx ucode 0x500002c). VPP as well as DPDK L3fwd tests are affected. | +| 1 | `CSIT-1701 | DPDK L3fwd tests with 9000B L2 packet size are not passing with Mellanox NICs. | +| | <https://jira.fd.io/browse/CSIT-1701>`_ | L3fwd application does not accept parameter for increasing -mbuf-size in same way DPDK testpmd does. | +----+------------------------------------------+----------------------------------------------------------------------------------------------------------+ diff --git a/docs/report/dpdk_performance_tests/overview.rst b/docs/report/dpdk_performance_tests/overview.rst index 212e202c13..7f3e930324 100644 --- a/docs/report/dpdk_performance_tests/overview.rst +++ b/docs/report/dpdk_performance_tests/overview.rst @@ -1,11 +1,10 @@ Overview ======== -DPDK performance test results are reported for all three physical -testbed types present in FD.io labs: 3-Node Xeon Haswell (3n-hsw), -3-Node Xeon Skylake (3n-skx), 2-Node Xeon Skylake (2n-skx) and installed -NIC models. For description of physical testbeds used for DPDK -performance tests please refer to :ref:`tested_physical_topologies`. +DPDK performance test results are reported for all physical testbed types +present in FD.io labs and installed NIC models. For description of physical +testbeds used for DPDK performance tests please refer to +:ref:`tested_physical_topologies`. Logical Topologies ------------------ diff --git a/docs/report/dpdk_performance_tests/test_environment.rst b/docs/report/dpdk_performance_tests/test_environment.rst index 03013a4981..3ea7bdc2a8 100644 --- a/docs/report/dpdk_performance_tests/test_environment.rst +++ b/docs/report/dpdk_performance_tests/test_environment.rst @@ -33,29 +33,28 @@ DPDK Compile Parameters .. code-block:: bash - make install T=<arch>-native-linuxapp-gcc -j + make install T=<arch>-<machine>-linuxapp-gcc -j Testpmd Startup Configuration ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -Testpmd startup configuration changes per test case with different -settings for `$$CORES`, `$$RXQ` and max-pkt-len parameter if test is +Testpmd startup configuration changes per test case with different settings for +`$$INT`, `$$CORES`, `$$RXQ`, `$$RXD` and max-pkt-len parameter if test is sending jumbo frames. Startup command template: .. code-block:: bash - testpmd -c $$CORE_MASK -n 4 -- --numa --nb-ports=2 --portmask=0x3 --nb-cores=$$CORES [--max-pkt-len=9000] --txqflags=0 --forward-mode=io --rxq=$$RXQ --txq=$$TXQ --burst=64 --rxd=1024 --txd=1024 --disable-link-check --auto-start + testpmd -v -l $$CORE_LIST -w $$INT1 -w $$INT2 --master-lcore 0 --in-memory -- --forward-mode=io --burst=64 --txd=$$TXD --rxd=$$RXD --txq=$$TXQ --rxq=$$RXQ --tx-offloads=0x0 --numa --auto-start --total-num-mbufs=16384 --nb-ports=2 --portmask=0x3 --disable-link-check --max-pkt-len=$$PKT_LEN [--mbuf-size=16384] --nb-cores=$$CORES L3FWD Startup Configuration ~~~~~~~~~~~~~~~~~~~~~~~~~~~ -L3FWD startup configuration changes per test case with different -settings for `$$CORES` and enable-jumbo parameter if test is sending -jumbo frames. Startup command template: +L3FWD startup configuration changes per test case with different settings for +`$$INT`, `$$CORES` and enable-jumbo parameter if test is sending jumbo frames. +Startup command template: .. code-block:: bash - l3fwd -l $$CORE_LIST -n 4 -- -P -L -p 0x3 --config='${port_config}' [--enable-jumbo --max-pkt-len=9000] --eth-dest=0,${adj_mac0} --eth-dest=1,${adj_mac1} --parse-ptype - + l3fwd -v -l $$CORE_LIST -w $$INT1 -w $$INT2 --master-lcore 0 --in-memory -- --parse-ptype --eth-dest="0,${adj_mac0}" --eth-dest="1,${adj_mac1}" --config="${port_config}" [--enable-jumbo] -P -L -p 0x3 .. include:: ../introduction/test_environment_tg.rst |