Age | Commit message (Collapse) | Author | Files | Lines |
|
Change-Id: I008b9e0fd62cdc8e29136930762bd7412bd50181
Signed-off-by: Jan Gelety <jgelety@cisco.com>
(cherry picked from commit 2230a8ab108fa114752decfc69321ec5a47f36a6)
|
|
Use parent system (Host, Container) installation of VPP. This will save
the internet bandwith by skip installing of prerequisites packages.
It will also skip dpkg install and simplify the process of initializing
VPP inside container.
Previosly initialization of VPP in container takes about 55s. With this
patch it is reduced to 2-3s.
This patch removes the bloated VOLUME creation between container
sidecars (a.k.a nested container) and fixes the hugepage allocation.
Change-Id: Ifa2be532edb77354657e1b84568bdc34993b00d0
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: Ie3d2b1f40a607ce5190ccfea6a372bc072d0a3b9
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: Id63fc8670c7816cc71d8d6d8704ff7f1adf3db65
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: I7463d47a9d65069332eb2a288ff44f4499a8a467
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Missing bits:
- Add up-to-date .rst document (in child Change).
- Prepare for releasing to PyPI.org (in child Change):
-- Either copy dependencies from MLRsearch, or list in requirements.
-- Perhaps move common dependencies to separate package
for both search to depend on.
-- All the other metadata stuff.
Jira: CSIT-1276
Change-Id: I277efdb63dbb90b30e11f4e30a82e2130ac8efc3
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
- Fix detecting total allocable CPUs for NFs
Change-Id: If3345518befb887655d3e162551c49df3163ed92
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: Ia87156b2ba826eb0a9a93d756b9765c7373ca6ce
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: I9148f6de75c3100f827eedec48c4c0ccb343bc54
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: I5a369f4b9f9b478375fda4a634bf57cc9623e972
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
CSIT-1372
Add following baseline tests to VPP_Device:
- ip4
eth2p-ethip4-ip4base-eth-2memif-1dcr-dev
- ip6
eth2p-ethip6-ip6base-eth-2memif-1dcr-dev
- l2bd
eth2p-eth-l2bdbasemaclrn-eth-2memif-1dcr-dev
- l2xc
eth2p-eth-l2xcbase-eth-2memif-1dcr-dev
Change-Id: Ic4a3a01b62d800c528a9c9371891dcc26b875220
Signed-off-by: Jan Gelety <jgelety@cisco.com>
|
|
Change-Id: Idf7cbcf285ff130f771eb3878065e6daa869ac71
Signed-off-by: Michal Cmarada <mcmarada@cisco.com>
|
|
CSIT-1372
Add following baseline tests to VPP_Device:
- ip4
eth2p-ethip4-ip4base-eth-2vhost-1vm-dev
- ip6
eth2p-ethip6-ip6base-eth-2vhost-1vm-dev
- l2bd
eth2p-eth-l2bdbasemaclrn-eth-2vhost-1vm-dev
- l2xc
eth2p-eth-l2xcbase-eth-2vhost-1vm-dev
+ fix doc strings in all vpp_device tests
Change-Id: Icc5ed801f3e8f6dedaa611779bd86fc20102fa74
Signed-off-by: Jan Gelety <jgelety@cisco.com>
|
|
- Add ability to create matrix of containers (chains x nodeness)
- Change default Docker image to csit-dut
- Use new resource mapping function
Change-Id: Iaac0f2efc99146fbaef0b33dc71b38770610023d
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Extend TG in CSIT to support to send unidirectional traffic, and then
it can be used by unidirectional test cases e.g. Load balancer.
Change-Id: Ie25bcdf7d4525a6d88d8ecd11a1b6d98275fa4d6
Signed-off-by: Yulong Pei <yulong.pei@intel.com>
|
|
Implement universal mapping function for placing the NF
threads to cpu cores/threads.
- Must be SMT aware.
- Must work with N-core NF.
- Must be DUT (vswtich) aware.
- Must be deterministic.
- Must use DTCR and MTCR parameters.
Change-Id: I98b61861f85c5af7ecd856719b4adf2ba95c262f
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: I5e17c5ab5cb99af0ac3d063ffc911bf95457d3e4
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: Ia5153d8e716d506c85ec8b723331438efd66ab6a
Signed-off-by: sharath <sharathkumarboyanapally@gmail.com>
|
|
Change-Id: I3b5f0568b5928cf12aaf2a555f52dd9809348579
Signed-off-by: Jan Gelety <jgelety@cisco.com>
|
|
Jira: CSIT-1336
Change-Id: I96d2b0221c5a7466484a82339fc132c5921532d1
Signed-off-by: Jan Gelety <jgelety@cisco.com>
|
|
Change-Id: I8c4090b508e28ae8f289e9dbbd5c72bbc3af4aca
Signed-off-by: Michal Cmarada <mcmarada@cisco.com>
|
|
- L2XC ping ICMPv4/ICMPv6 test
Change-Id: I1c1708e8c1050f140a02d56012f63b683fc3eec1
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: Ibaa7652b24385805b5cdd1540ed98b09e68c69bc
Signed-off-by: sharath <sharathkumarboyanapally@gmail.com>
|
|
VPP was not initialized (shm), after service restart, quickly enough.
This results into vpp_api_test console timeouts when communicating over
uninitialized socket.
Patch is introducing more stable test to detect service (VPP) is fully
ready and initialized.
Change-Id: I300a763814feba28cb8163dcd4996b53b2994fec
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: I2ccca8955949840305350bec1f60bc6d1cddcf82
Signed-off-by: Michal Cmarada <mcmarada@cisco.com>
|
|
Jira: CSIT-1273
Change-Id: Ic2a41661c158384a5aaa7d4e73f30ffca13ddf82
Signed-off-by: Jan Gelety <jgelety@cisco.com>
|
|
- API changed.
Change-Id: If70e1a930aa14b129ad38ef544cff32ad5a08cdd
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Currently there is and issue when consecutive AVF suites are
trying to initialize AVF and bind VF to vfio-pci driver, there
seems to be race condition if VPP has already VFs allocated.
This patch should introduce safer contstrain to disable VPP prior
AVF init.
Change-Id: Id8839200f19d9e7a18d161b660cfb1e43a8ca379
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: I0f734c2f11cc89716371554880f1d7404ea16090
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
This patch suppose to fix the single memif TCs failing on SKX nodes.
By stoping Host VPP during startup of Guest VPP. This way Guest VPP
can take interface on single pass.o
Change-Id: I88e6932bf007ce0f6a367927de5f017ffb8fdaca
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: I960c2b4170efa06597851853cfc15bda1c60fe94
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: Ifb1a0702353e71c61a694dea5164df41f4e23389
Signed-off-by: sharath <sharathkumarboyanapally@gmail.com>
|
|
Change-Id: I8aa557fe75133d20eef9832e314357af71f1d19a
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: Ia8da73cbf7ab213b61a330b757419330667be24b
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Add ability to see Vat history for file sourced commands.
Change-Id: I49b19236bec7253ea412659368cb7ccb5f5e3274
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
+ "elif" instead of "else if".
+ Some suites had test cases reordered.
+ Some suites contained 9000B cases for avf.
Change-Id: I3bf8241c317f20dbf66b497a2c8be35ef71307f5
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
Change-Id: Ia74a87346fe2c0a170b881b510f1b0558a538f3c
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: Iaeac10676f46e0267041afee927bc6e3201146b2
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Single Socket platform e.g. Intel Atom cpu based SOC platform with
Ubuntu 16.04.4(kernel 4.13.0-36-generic) or Centos 7.5 (kernel 3.10.0-862.el7),
value of /sys/bus/pci/devices/<pci_device_id>/numa_node is -1, this will
break CSIT performance test running, but for this kind of SOC platform,
it can consider that is not NUMA based platform, numa_node=-1 is reasonable,
so fix it at CSIT side, when numa_node=1 and the system's NUMA node count
is 1, set numa_node=0. DPDK also did it as this way.
Change-Id: I9ac23d3cece2f1489e38f05b50a462bb2ad9f661
Signed-off-by: Yulong Pei <yulong.pei@intel.com>
|
|
- Add L1 KWs for SR-IOV handling (init Vfs, remove Vfs, ...)
- Cleanup L1 KWs for SR-IOV bind/unbind/pci_get/...
- Add L2 KWs for Test Setup/Teardown, L2patch, Create AVF interface
- Add sample L2patch test fox x710, xxv710
Change-Id: If17077877455a14043617d8ea0d06cbe47b469e3
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
+ SchedUtils.py: Do not use `len(SEQUENCE)` to determine if a sequence is empty
+ VatHistory.py: Do not use `len(SEQUENCE)` to determine if a sequence is empty
+ VppCounters.py: Do not use `len(SEQUENCE)` to determine if a sequence is empty
+ Memif.py:
++ Do not use `len(SEQUENCE)` to determine if a sequence is empty
++ Either all return statements in a function should return an expression,
or none of them should.
++ Update :return: on possible None.
+ Classify.py: Unnecessary "else" after "return"
+ ContainerUtils.py: Useless super delegation in method '__init__'
+ CpuUtils.py: Do not use `len(SEQUENCE)` to determine if a sequence is empty
+ DropRateSearch.py: Either all return statements in a function
should return an expression, or none of them should.
+ IPv4NodeAddress.py: Do not use `len(SEQUENCE)` to determine if a sequence is empty
++ Also improve docstrings.
+ IPv4Setup.py: Useless super delegation in method '__init__'
+ IPv6Setup.py: Do not use `len(SEQUENCE)` to determine if a sequence is empty
++ Also improve docstrings.
+ IPv6Setup.py: standard import "from ipaddress import IPv6Network"
should be placed before "from robot.api import logger"
+ MacSwap.py: Trailing newlines
+ NATUtil.py: Do not use `len(SEQUENCE)` to determine if a sequence is empty
+ NodePath.py: Unnecessary "else" after "return"
+ Tap.py: Do not use `len(SEQUENCE)` to determine if a sequence is empty
+ topology.py: Either all return statements in a function
should return an expression, or none of them should.
+ topology.py: Unnecessary "else" after "return"
++ Do not use `len(SEQUENCE)` to determine if a sequence is empty
++ Improve docstrings
+ DUTSetup.py: Do not use `len(SEQUENCE)` to determine if a sequence is empty
++ Also do not compare int(ret_code) just to access zero-ness.
+ ssh.py: Do not use `len(SEQUENCE)` to determine if a sequence is empty
+ InterfaceUtil.py: Unnecessary "else" after "return"
Change-Id: Iba4244aa79661ee7df15fed5c7c6dbf04dfa88b2
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
- added also 3-node test cases for comparison
Jira: CSIT-1261
Change-Id: I1bfcc184e925e969d4a610920c4e49af23047507
Signed-off-by: Jan Gelety <jgelety@cisco.com>
|
|
- Decrease pylint errors. Part: easy
Change-Id: I452e5b5a11e9b78c03cd173a3848babe21b93c73
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
if ssh use root account to login ubuntu os, its prompt is ":~# ",
so add it to let csit test can run with root account.
Change-Id: I21f5380245324c8e48099433e3d374188e9001d9
Signed-off-by: Yulong Pei <yulong.pei@intel.com>
|
|
Make number of doublings configurable,
keep Python default at 1, set Robot default to 2.
Also make docstring types unique
(pylint was complaining about classes and modules having the same name).
Increase MLRsearch version to 0.2.0.
Change-Id: Ib846032e79ff52994503c0cfef2f86655502c275
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
CSIT-1164 Create VM vhost tests for 2-node topology
CSIT-1173 Implement VM vhost KWs for 2-node topology
- Cleanup QemuUtils.py Library (PEP8, some TODOs, readability, multi queue,
ring size, qemu version detection)
- Cleanup VHOST KWs perf (Unify running QEMU on N-node topology)
- Add ability to configure Queues, Hyperthreading, automatically compute
number of CPU for VM/Testpmd and RXQ for Testpmd workload.
- Partial cleanup of configuration KWs (l2xc, l2bd) for 2-node
- Create IPv4 2-node KW
Change-Id: I8aae1355bafa651de715cd438fe706c443ea1d88
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
CSIT-1155 Implement container memif KWs for 2-node topology
Change-Id: I4fc7bcab499a0b8a8594114d3f331b0d9610bf9d
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
- DPDK and VPP bootstrap merged into one (ligato will follow).
- Added more error detections
- Added topo-cleanup.py to cleanup TB before and after test.
- Remove installation of VPP
Change-Id: I7a86117eb03cbe4ee8fde47baeed58bc86c0dfb2
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
For now this solution does not work on Haswells, need to find permanent
fix.
Change-Id: Id5323a078db3a8f14c009d6f66ee393c89d851ef
Signed-off-by: Peter Mikus <pmikus@cisco.com>
(cherry picked from commit 5f634b8c486f7a404c9cd33b60cc92a6907e7c64)
|
|
- Increase timeout when trying to check interface (this is useful if
there are multiple interfaces and dump command take more than
TIMEOUT/NR_IF which is happening quite often).
- Reverse the order of applying MTU and interface UP to try to prevent
API overload during interface initialization phase.
Change-Id: I4e8623963d15b3c0d202172b4c58c43a972bb9cb
Signed-off-by: Peter Mikus <pmikus@cisco.com>
(cherry picked from commit 584345da4f7330f7517e5defd894b4995c2b57b0)
|