Age | Commit message (Collapse) | Author | Files | Lines |
|
Automatic selection of OVS and DPDK directories must be performed
before initialization of modules is called. Otherwise modules will
use wrong values with unpredictable results.
Default directory values have been changed to point to sources build
by make.
Clean&sweep of directory usage was performed.
Configuration file for pylint has been renamed to follow pylint
naming convention.
Change-Id: I42c89f8890c021f0062b478a856ae0113f1245fb
JIRA: VSPERF-121
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Tv Rao <tv.rao@freescale.com>
|
|
Change-Id: Ib852c81fc8e09593411984bfb6f9bc5d5629f0c4
|
|
Fixup the location of all rst documentation and any errors in the ltd.
JIRA:VSPERF-60
Change-Id: I1472d540b1b4eca4ab038d079bfb0ef2f0c5ebcd
Signed-off-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Tv Rao <tv.rao@freescale.com>
|
|
Update the rules for Make clean and distclean.
Update make not to clone the repositories if they are already
cloned, simply pull, checkout the tags and build.
Changes in this patch:
1. QEMU is built just once in .../src directory.
.../src_vanilla and .../src_cuse don't contain qemu anymore
2. 'make clean' deletes all built *.o files
3. 'make distclean' deletes all built and generated files
(distclean is used instead of cleanse now)
4. 'make' calls 'git pull' when repository is cloned already
JIRA: VSPERF-94
Change-Id: I80f9a135580b450fb4606cf947661049ff509548
Signed-off-by: Radek Zetik <radekx.zetik@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
|
|
Create bridge and set datapath_type in 1 transaction to prevent OVS from
generating misleading error messages which jenkins considers as
build errors.
Update vswitch interface to include optional parameters.
Change-Id: I668f474ea909b284f3726807aab42d52ae2fb504
JIRA: VSPERF-122
Signed-off-by: Dino Simeon Madarang <dino.simeonx.madarang@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
Reviewed-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Radek Zetik <radekx.zetik@intel.com>
|
|
Installation scripts for Centos7 have been updated with new packages.
Script will install also python3 packages and create python virtual
environment. As a last step DPDK, OVS and QEMU will be compiled. So
after script execution VSPERF should be ready for use.
It is recommended to execute installation script from user account
with password less sudo access. Otherwise python environment
must be copied over from root home directory.
After VSPERF generic Makefile will be created by VSPERF-114, it can
be enhanced to call build script too.
Change-Id: I6ff5d2660af29cf8be134a4cae3b1d7b559892fa
JIRA: VSPERF-93
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Signed-off-by: Abdul Halim <abdul.halim@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
|
|
Two sample contionous stream testcaes were introduced for PVP and PVVP
scenarios. Required frame rate percentage can be specified either by
tetcase configuration option "iLoad" or by command line option "iload".
Command line option take precendece to testcase option. By default 100%
linerate is used if not specified otherwise.
Change-Id: I95f7ae6081377a4c99ce70fc7504b6b542f9260b
JIRA: VSPERF-95
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
Reviewed-by: Al Morton <acmorton@att.com>
|
|
Initial pylint config file has been created to follow current
code structure and design. A few issues have been fixed, so currently
56 of 67 files have rating 10/10. There are still issues to be fixed,
mainly TODOs/FIXMEs and missing impelentation of abstract class members.
Change-Id: Icf5b12c2150258f8ac657f8d50b47a697d1660a0
JIRA: VSPERF-91
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Tv Rao <tv.rao@freescale.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
|
|
Flow definition can be changed by testcase setup. By default
only ingress ports are used to route traffic, but flow can be
extended by src and dst IP addresses used by the traffic
generator.
Change-Id: I694714537ad2d818192a39f3235d8e6bf0f59a49
JIRA: VSPERF-90
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Tv Rao <tv.rao@freescale.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
|
|
New configuration parameter EXCLUDE_MODULES defines module names,
which won't be automatically loaded by LoaderServant. It can be used
to exclude obsolete or abstract modules.
Change-Id: If98b50b1505465bcedcf28fe63421c73a4fe160a
JIRA: VSPERF-118
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Tv Rao <tv.rao@freescale.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
|
|
Module manager has been enhanced to support both 'insmod' and 'modprobe'
commands for kernel module insertion. In case, that .ko suffix is detected
then insmod will be used otherwise modprobe will be called. This allows
user to specify full path to each module. For example vanilla ovs module
can be defined as "OVS_DIR_VANILLA + 'datapath/linux/openvswitch.ko'"
to load kernel module matching OVS vanilla source tree version.
Change-Id: Ib8d16eca84449ad34d6b307ab836f58d2f0d341b
JIRA: VSPERF-116
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
|
|
The RFC 2889 Forwarding Rate test requires a Deployment Scenario
Section - was missing.
JIRA: VSPERF-102
Change-Id: Iaf661a6bbb9ad52f23147f9444f23d96a82ceb2d
Signed-off-by: Al Morton <acmorton@att.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Billy O'Mahony<billy.o.mahony@intel.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
|
|
LTD.Scalability.RFC2544.0PacketLoss needs to add the caution that
all flows must be learned before running Throughput testing,
and no flows can be allowed to time-out during the test.
JIRA: VSPERF-108
Change-Id: I95ae20a978b7af660b358450e67ce60939b0b7c1
Signed-off-by: Al Morton <acmorton@att.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Billy O'Mahony<billy.o.mahony@intel.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
|
|
Among the Updates needed are:
- Several New tests created since July, add to Matrix section.
- Re-named tests: fix in Matrix section.
- Link to Summaries on Wiki.
- Link to Test Setup Figures on Wiki.
- Add some general points on New Test Equipment Challenges/Metrics.
- Mention future test specs, and IFA003 as one source.
JIRA: VSPERF-120
Change-Id: I981b0b6301a208d20220f67784b4591aa3f51338
Signed-off-by: Al Morton <acmorton@att.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Billy O'Mahony<billy.o.mahony@intel.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
|
|
In the LTD, some of the tests with "Throughput" in the title are not actually
testing throughput in the data path.
Since the test names are used by TOIT/VSPERF to group similar tests for batch
execution, it's important that the names accurately represent the test methods
and metrics measured.
JIRA: VSPERF-98
Change-Id: I620ddc0779232f2dd325379005faba8692fea18c
Signed-off-by: Al Morton <acmorton@att.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Billy O'Mahony<billy.o.mahony@intel.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
|
|
A few of the tests should call out a prerequisite.
LTD.MemoryBandwidth.RFC2544.0PacketLoss.Scalability
and
LTD.Scalability.RFC2544.0PacketLoss
are examples.
JIRA: VSPERF-107
Change-Id: I1ca653ac765f9494b48759764103dfc9d42a42f1
Signed-off-by: Al Morton <acmorton@att.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Billy O'Mahony<billy.o.mahony@intel.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
|
|
Running a test with vanilla-ovs, vhost-user-ovs,
vhost-cuse-ovs,... does not require any building
or configuration changes - the vswitch needs only
to specified on the command line.
The config file contains new variables for particular
ovs build.
OvsVanilla:
OVS_DIR_VANILLA, QEMU_DIR_VANILLA
VhostCuse:
RTE_SDK_CUSE, OVS_DIR_CUSE, QEMU_DIR_CUSE
VhostUser:
RTE_SDK_USER, OVS_DIR_USER, QEMU_DIR_USER
The vspef chooses right paths according to command line
switches '--vswitch' (or 'VSWITCH') and '--vnf' (or 'VNF')
JIRA: VSPERF-86
Change-Id: I22e3a7f9e21d6ceb13830e444716b4274314560c
Signed-off-by: Radek Zetik <radekx.zetik@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Tv Rao <tv.rao@freescale.com>
|
|
LTD.Throughput.RFC2544.Profile refers to the metric "maximum forwarding rate"
but implies that this is the zero% loss rate, which would be the Throughput
level (MFR is usually higher than Throughput, because MFR allows losses).
It may be that MFR and Throughput are just interchanged in the last two
paragraphs of the description.
JIRA: VSPERF-110
Change-Id: I4f36fd4ce66f2f4ebee4f60a11ad03adfb2b3b7c
Signed-off-by: Al Morton <acmorton@att.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Billy O'Mahony<billy.o.mahony@intel.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
|
|
The following text in the Description needs to be formatted as a list in .RST:
Furthermore: - the ratio of reads to writes should be recorded.
JIRA: VSPERF-109
Change-Id: I3b6c819d1ac07cc466c747d3552cbcc4de5f34bf
Signed-off-by: Al Morton <acmorton@att.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Billy O'Mahony<billy.o.mahony@intel.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
|
|
In the RFC 2889 Broadcast rate testing, four test ports are required.
One of the ports is connected to the test device, so it can send
broadcast frames and listen for miss-routed frames.
JIRA: VSPERF-103
Change-Id: I5ee3d20c71d7debf6c7b09fc094f72b2d4c10684
Signed-off-by: Al Morton <acmorton@att.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Billy O'Mahony<billy.o.mahony@intel.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
|
|
Support for PVP and PVVP scenerios using Vanilla OVS was added.
VMs are configured to forward traffic between virtual interfaces
by standard linux capabilities. Traffic generator script was updated
to accept MAC and IP address configuration to allow VM forwarding
configuration.
VSPERF-66, VSPERF-70
Change-Id: Ia70ab6be547b39928a1eae319faba779d4e29284
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Signed-off-by: Dino Simeon Madarang <dino.simeonx.madarang@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
Reviewed-by: Gene Snider <eugene.snider@huawei.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Tv Rao <tv.rao@freescale.com>
|
|
Text & sequence diagrams as both source and png.
JIRA: VSPERF-106
Signed-off-by: Billy O'Mahony<billy.o.mahony@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Eugene Snider <Eugene.Snider@huawei.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Gurpreet Singh <gurpreet.singh@spirent.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
Change-Id: I33f57fba85710ab4cfc72d0a5d06db6f653545d6
|
|
The RFC 2889 Address Learning and Cache Capacity tests require
a listening port to check for flooded frames.
JIRA: VSPERF-74
Change-Id: I7b4d2461e819d01d642c9c4d95b938dfbe056f2c
Signed-off-by: Al Morton <acmorton@att.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
|
|
Technical Debt - Add appropriate build flags to the OVS makefiles to
build OVS for performance.
JIRA:VSPERF-92
Change-Id: I63d45ebe8ba41799158f97db892a94e16c7b3df5
Signed-off-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
|
|
Generic PVVP deployment support has been added. Two
new testcase scenarios for throughput and back2back
tests with PVVP deployment were added. Original
implementation of PVP has been refactored
Following files were affected:
modified: conf/01_testcases.conf
modified: conf/02_vswitch.conf
modified: conf/04_vnf.conf
modified: core/__init__.py
modified: core/component_factory.py
modified: core/loader/loader_servant.py
modified: core/vnf_controller.py
deleted: core/vnf_controller_p2p.py
deleted: core/vnf_controller_pvp.py
modified: core/vswitch_controller_p2p.py
modified: core/vswitch_controller_pvp.py
new file: core/vswitch_controller_pvvp.py
modified: docs/to-be-reorganized/NEWS.rst
modified: docs/to-be-reorganized/quickstart.rst
modified: src/dpdk/dpdk.py
modified: src/ovs/ofctl.py
modified: testcases/testcase.py
modified: tools/tasks.py
new file: vnfs/qemu/qemu.py
new file: vnfs/qemu/qemu_dpdk.py
modified: vnfs/qemu/qemu_dpdk_vhost_cuse.py
modified: vnfs/qemu/qemu_dpdk_vhost_user.py
modified: vnfs/vnf/vnf.py
modified: vswitches/ovs_dpdk_vhost.py
Change-Id: Ib6869a29337a184cb58c57fd96bba1183aba00ab
JIRA: VSPERF-68, VSPERF-69
Signed-off-by: Martin Klozik (martinx.klozik@intel.com)
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
|
|
Currently the VSPERF makefiles make just one variant
of vSwitch - OVS with DPDK-vhost-user. To make the other
variants the makefile has to be edited. And the make re-run.
The fix builds all required versions:
DPDK-vhost-user OVS (in src/)
Vanilla OVS (in src_vanilla/)
DPDK-vhost-cuse OVS (in src_cuse/)
HowTo:
- git clone vsperf
- cd .../src
- make
JIRA: VSPERF-85
Change-Id: I469baf8432f0b127cde785f35a6c57a2c778f23f
Signed-off-by: Radek Zetik <radekx.zetik@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
|
|
In case there is specified concrete value for "duration" in the
configuration there is raised TypeError exception
in the send_rfc2544_throughput function of dummy traffic generator.
Conversion to integer solved this issue.
Change-Id: Ib37ecabc1000369d807a997c88d7a3d7aca425d5
Signed-off-by: Robert Wojciechowicz <robertx.wojciechowicz@intel.com>
Reviewed-by: Dino Simeon Madarang <dino.simeonx.madarang@intel.com>
Reviewed-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
|
|
The Current list of fixed frame sizes need to be augmented with
IMIX Traffic Profiles in the test framework.
JIRA: VSPERF-100
Change-Id: I97425c09c1d692bae56f2518d479b83a282eb7d1
Signed-off-by: Al Morton <acmorton@att.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
|
|
The RFC 2889 tests require Full Mesh on multiple ports.
Modify the bunny-ears fig to show this (and address -99 in the process)
JIRA: VSPERF-99
JIRA: VSPERF-101
Change-Id: Id67cb5ef6748c83858a90033534f960874614c06
Signed-off-by: Al Morton <acmorton@att.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
|
|
Since vsperf (ixnet trafficgen) started using AggResults.csv
to pull out traffic results after bi-dir support was added.
Some of the derived results are wrong (derived meaning
these are figures vsperf calculates from the info available
in the raw ixnet results).
JIRA: VSPERF-88
Change-Id: I79d88bfc7718bda675d2ba7e5148053405a94211
Signed-off-by: Radek Zetik <radekx.zetik@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
|
|
New sysmetrics implementation is based on pidstat command line tool
from sysstat package. Old non-functional implementation was removed.
Reporting was refactored to generate report after each TC from values
already available in memory.
Following files were affected:
modified: conf/01_testcases.conf
modified: conf/02_vswitch.conf
modified: conf/05_collector.conf
deleted: core/collector_controller.py
modified: core/component_factory.py
modified: docs/NEWS.rst
modified: packages.txt
modified: requirements.txt
modified: testcases/testcase.py
modified: tools/collectors/collector/collector.py
modified: tools/collectors/sysmetrics/__init__.py
deleted: tools/collectors/sysmetrics/linuxmetrics.py
new file: tools/collectors/sysmetrics/pidstat.py
modified: tools/report/report.jinja
modified: tools/report/report.py
modified: tools/systeminfo.py
modified: vsperf
JIRA: VSPERF-67
Change-Id: I25a79f2afef405b9ac46ae85c18044af167a62a4
Signed-off-by: Martin Klozik (martinx.klozik@intel.com)
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Gurpreet Singh <gurpreet.singh@spirent.com>
Reviewed-by: Tv Rao <tv.rao@freescale.com>
|
|
Only works for P2P. Other deployment scenarios will need to implement
the same kind of multi-table flow as P2P to enable frame modification.
Signed-off-by: Billy O'Mahony<billy.o.mahony@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Radek Zetik <radekx.zetik@intel.com>
Reviewed-by: Dino Madarang <dino.simeonx.madarang@intel.com>
Change-Id: I2fa35972b90b165c00f0d0a7515fa275d0f17aa4
|
|
TODO: Reorganize documentation into properly named subfolders
TODO: Add logos to documentation (see index.rst for example)
Note you can see the results of this change via a link in the gerrit
comments below
NOTE: The merge job for your project has not been enabled, this code
will not go to the artifacts page upon a merge. you can merge it and
work on it in your own time, please contact me to enable the merge
(the merge job will make this documentation avaliable to the public on
artifacts.opnfv.org)
JIRA:RELENG-15
Change-Id: I3f3d8c9d7fceae90e0ed824b66bdaf7a8ea66328
Signed-off-by: Aric Gardner <agardner@linuxfoundation.org>
|
|
JIRA: VSPERF-72
This change adds optional IPv4 DNAT support to l2fwd module for level two
forwarding. With this change the l2fwd module can be used for testing
complex IP configuration routing with virtual switches.
Other L2 forwarding modules provide only Level 2 pass through. These can be
used for complex IP chain processing but rely on port forwarding in the switch.
With this module each packet which is forwarded optionally gets a new
destination IP and source masquerade. So a test packet can be routed to eth1
and forwarded on eth2 with a different target IP and source address.
To use this module compile it for your kernel and use the 'insmod' command
to insert it.
With no arguments this will forward eth1 to eth2 without modification.
With 'net1=ethX net2=ethX' Level 2 forwarding can be done between arbitrary
ports.
With 'netX=ethX,XX.XX.XX.XX,xx:xx:xx:xx:xx:xx' the packets which are forwarded
on the target port will be given the new destination IP and mac address. One
or both ports may enable DNAT.
Change-Id: If24281a6841930a7a85e4536da96b980ed48df1b
Signed-off-by: Gene Snider <eugene.snider@huawei.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
|
|
Expand description of Packet Delay Variation to capture
the additional details which will be important to measurement
design. Measurmeent devices may be implementing the RFC 5481
PDV metric for the firt time, so som general details will
be useful to include in the LTD. Addresses MT's comments.
JIRA: VSPERF-73
Change-Id: I60524969f72bb00519e8070ae51e219d857c554e
Signed-off-by: Al Morton <acmorton@att.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
|
|
Remove call to VNF start since it is already called in __enter__
of VnfControllerPVP.
VSPERF-66
Change-Id: I9c9174ffe7d03efacb6853c059b43c8a5ca66704
Signed-off-by: Dino Simeon Madarang <dino.simeonx.madarang@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
|
|
The current --tests option can make it hard to specify a single test to run.
This patch allows specification of tests using exact names.
JIRA: VSPERF-58
Signed-off-by: Billy O'Mahony<billy.o.mahony@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Radek Zetik <radekx.zetik@intel.com>
Reviewed-by: Dino Madarang <dino.simeonx.madarang@intel.com>
Reviewed-by: Brian Castelli <brian.castelli@spirent.com>
Change-Id: Ib820bdae96d6257d785f4e310a3b3d1fbfa1b575
|
|
Load eventfd only when vhost-cuse is specified as
the guest access method.
This commit also includes quickstart update for executing PVP tests.
JIRA: VSPERF-59
Change-Id: I41f116ab1c42f704026567a269922c07994b15a5
Signed-off-by: Dino Simeon Madarang <dino.simeonx.madarang@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Gurpreet Singh <gurpreet.singh@spirent.com>
Reviewed-by: Tv Rao <tv.rao@freescale.com>
Reviewed-by: Fatih Degirmenci <fatih.degirmenci@ericsson.com>
|
|
Bind DPDK ports back to the original driver they were bound to. This
will be useful when switching between the types of vswitches we are
testing.
Change-Id: I6818d3f8e3c2a5d3a5f083cf9e37e0819e0b39d1
Signed-off-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Dino Simeon Madarang <dino.simeonx.madarang@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
|
|
Refactor rfc25444_duration to duration and update the --test-params help
to show the type of parameters that can be set from the cli. Also
configure pkt_sizes so that it can be modified from the CLI. Also fixes
the tcl script to configure framesize properly (to the value selected
and not the default 64b).
JIRA: VSPERF-27
Change-Id: I76dcd824977a5f940cf7476090d2d2a3acca8160
Signed-off-by: Billy O'Mahony<billy.o.mahony@intel.com>
Signed-off-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Dino Madarang <dino.simeonx.madarang@intel.com>
|
|
LTD.CPU.RFC2544.0PacketLoss and LTD.Memory.RFC2544.0PacketLoss"
|
|
LTD.CPU.RFC2544.0PacketLoss and LTD.Memory.RFC2544.0PacketLoss
Every testcase can be configured to run background load
on the DUT. Load is generated by chosen command line tool,
which must be installed according to the installation.md.
Currently 'stress' and 'stress-ng' tools are supported,
but support for new tool can be easily added.
Load can be configured by testcase configuration option 'load'.
Details about supported load generator options can be found
in conf/01_testcases.conf.
JIRA: VSPERF-23
JIRA: VSPERF-24
Change-Id: I3e1aebc0d41357f5a2764015c41ffc7e50fdbf8b
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
|
|
|
|
JIRA: VSPERF-64
Change-Id: I7cefd05479da6fe17357da4da558f466175dd6fc
Signed-off-by: Dino Simeon Madarang <dino.simeonx.madarang@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
|
|
Enable PVP testing using vhost-cuse as guest access method
Recent changes:
* Move NEWS.md changes to NEWS.rst
* Update comment to vhost-cuse
* Restore config back to checkout state after make
* Merge OPNFV 1092 updates
* Add PVP-vhost-cuse to NEWS.md
* Add comment/example to GUEST_CORE_BINDING
* Move hardcoded values to conf/04_vnfs.conf
* Set default VNF to QemuDpdkVhostCuse
* Compile eventfd_link if VHOST_USER=n
* Use MAC and PCI address from conf
* Use vhost method in conf to properly set interface type
JIRA: VSPERF-59
Change-Id: Ib1159e216f3e25b9971c0935969680582683916b
Signed-off-by: Madarang, Dino Simeon <dino.simeonx.madarang@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
|
|
JIRA: VSPERF-60
Change-Id: I63695f8fa8a86cf1e64fd959658e11f3d827a3c4
Signed-off-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
|
|
Migrate all existing VSPERF documentation to a new directory called
docs/ and convert to ReStructuredText format. It's recommended that any
doc changes in the future are tested with: http://rst.ninjs.org/.
Change-Id: I18aa574b1259986502bde4ceef1fae7c6a5c1f33
JIRA: VSPERF-60
Signed-off-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Eugene Snider <Eugene.Snider@huawei.com>
Reviewed-by: Gurpreet Singh <gurpreet.singh@spirent.com>
Reviewed-by: Tv Rao <tv.rao@freescale.com
|
|
Update packages.txt with the required packages to build QEMU v2.2.1.
JIRA: VSPERF-56
Change-Id: Ifdf802de3c195b0dc5c48dcb8f098c31a1561012
Signed-off-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Madarang, Dino SimeonX <dino.simeonx.madarang@intel.com>
|
|
Reporting from TOIT was merged and improved. Default template was modified
to support any testcase and to show more details about system environment.
Affected files:
* docs/NEWS.md
* testcases/testcase.py
* tools/report/__init__.py
* tools/report/report.jinja
* tools/report/report.py
* tools/systeminfo.py
* vsperf
JIRA: VSPERF-71
Change-Id: I4dc84ca69e5c292eae1f8dede1411c06ae3ef8af
Signed-off-by: Martin Klozik (martinx.klozik@intel.com)
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
|
|
Enable booting of a VM with DPDK and run testpmd for
PVP testing.
* Added throughput and back2back tests with pvp deployment scenario in
01_testcases.conf
* PVP requires DPDK 2.0 with VHOST_USER enabled and QEMU 2.2.0
* Tested on CentOS7 and Fedora 20
* Fix conflict with change 1078
Recent Changes:
* Fix merge conflict (testcase.py and testcases.conf)
* Remove QEMU_DIR. User must set QEMU_BIN
* Set bidir traffic to True
* Add flow for bi-directional traffic
* Use working OVS_TAG ad2e649834be20dd01b1632799fe778106a96a2d
* Merge change 1096 (src: Add QEMU makefile)
* Set virtio-net-pci csum=off and other variables to off
* Move hardcoded values to conf/*
JIRA: VSPERF-56
Change-Id: I4ad184531064855493483d9833a7722c9f7d3576
Signed-off-by: Madarang, Dino Simeon <dino.simeonx.madarang@intel.com>
Signed-off-by: Meghan Halton <meghan.halton@intel.com>
Reviewed-by: Billy O Mahony <billy.o.mahony@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Martin Klozik <martinx.klozik@intel.com>
|