Age | Commit message (Collapse) | Author | Files | Lines |
|
In cases with VHost user and non-jumbo frame scenarios we can
disable merge buffers on the guest NICS to improve performance.
This patch adds this option to the 04_vnf conf file.
JIRA: VSPERF-388
Change-Id: Ic1f776e90fdccaa5d5a0c952c96811a8747fe936
Signed-off-by: Christian Trautman <ctrautma@redhat.com>
|
|
QemuPciPassthrough class responsible for SRIOV testing
used wrong device names for GUEST configuration. Details
of HOST interfaces connected to the traffic generator were
used to initialize forwarding applications inside the guest.
Code was fixed to use correct device configuration.
As part of this fix, old bug in VNF type detection was fixed.
JIRA: VSPERF-389
Change-Id: I269c9232012980c600c171571e90beb11e7b72cd
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Christian Trautman <ctrautma@redhat.com>
Reviewed-by: Bill Michalowski <bmichalo@redhat.com>
|
|
MoonGen code optimizations and general clean up. No
functionality changes.
JIRA: VSPERF-351
Change-Id: Ida16f16eb218eb6f0a1c90594c38511bfbb2cf97
Signed-off-by: bmichalo <bmichalo@redhat.com>
|
|
|
|
Documentation was updated as follows:
* section decorators were fixed in LTD
* links to testcase and deployment
descriptions were fixed
* duplicate labels removed from LTP<D
* fixed typos in formatting and text
* links to scalability and stress tests
fixed after their renaming
JIRA: VSPERF-349
Change-Id: I86c2b723231431388b69d3d696805de2a4e12435
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Christian Trautman <ctrautma@redhat.com>
Reviewed-by: Bill Michalowski <bmichalo@redhat.com>
|
|
|
|
Revert enforced vsperf failure in case, that VnfController
is used in scenario without VNFs. Instance of VnfController
is created for all scenarios including p2p, so it must
be prepared for it.
JIRA: VSPERF-361
Change-Id: Icf6e88e7ea62044c09bb194c260ea878d6a197c9
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Christian Trautman <ctrautma@redhat.com>
Reviewed-by: Bill Michalowski <bmichalo@redhat.com>
Reviewed-by: Antonio Fischetti <antonio.fischetti@intel.com>
|
|
|
|
|
|
Hyperlinks to the yardstick documentation were updated
to refer to files generated from the master branch.
JIRA: VSPERF-349
Change-Id: I248795b6a951377885492515fa9b96a694c9da5d
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Christian Trautman <ctrautma@redhat.com>
Reviewed-by: Bill Michalowski <bmichalo@redhat.com>
|
|
|
|
Remove files with IETF copyrights (Internet Draft .txt files).
I assume these files have to be removed in C-stable branch...
JIRA: VSPERF-387
Change-Id: I76709ab055f845cec0bec8d07b14c29dec349f54
Signed-off-by: Al Morton <acmorton@att.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Bill Michalowski <bmichalo@redhat.com>
Reviewed-by: Christian Trautman <ctrautma@redhat.com>
Reviewed-by: Antonio Fischetti <antonio.fischetti@intel.com>
|
|
Support for deployment scenarios with any number of VMs
in both serial and parallel configuration. Detailed
content of the patch:
* VswitchControllerPXP class for multi VM support
* pvvpxx and pvpvxx deployments for xx VMs in
serial respective parallel configuration
* special GUEST_ options expansion to requested
number of VMs;
* support of GUEST_ options specific macros
#VMINDEX, #MAC(), #IP() and #EVAL()
* all GUEST specific options are turned to lists
to be VM specific
* support for VM with 1 NIC
* support for VM with multiple NIC pairs; traffic
is routed in serial or parallel between NIC paris
based on deployment scenario
* support for PVVP and PVPV scenarios using VMs
with different numbers of NICs
JIRA: VSPERF-361
Change-Id: I05bedbdfa9a81ea0166d9b03d83ae49d6cb8b19b
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Christian Trautman <ctrautma@redhat.com>
Reviewed-by: Bill Michalowski <bmichalo@redhat.com>
Reviewed-by: Antonio Fischetti <antonio.fischetti@intel.com>
|
|
Change vhost affinity process lookup to use pgrep instead of
ugly error prone regex method as suggested.
Change-Id: I8cac00a7f2d26ba1ce1815535a198f578159222c
Signed-off-by: Christian Trautman <ctrautma@redhat.com>
|
|
Misplaced parenthesis regarding THROUGHPUT_RX_MBPS
JIRA: VSPERF-381
Change-Id: Ie33cf6ca88164f0686a819715b5f9f7409072063
Signed-off-by: bmichalo <bmichalo@redhat.com>
|
|
Documentation was updated to reflect recent changes
of vloop-vnf image.
JIRA: VSPERF-374
Change-Id: I65a34a7618593a7b77c38eb2fa4c87619b937346
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Christian Trautman <ctrautma@redhat.com>
Reviewed-by: Bill Michalowski <bmichalo@redhat.com>
|
|
Adds multi-queue for vanilla OVS using virtio-net.
TunTap ports will use multi_queue parameter when ports are
created/deleted if guest nic queues are enabled and
vswitch is ovs vanilla. Virtio net will now add guest nic
queues to qemu command line if appropriate.
Reworked multi-queue documentation to reflect these changes.
Adds vhost net thread affinitization which is recommended
when performing vanilla ovs multi-queue.
Guests will require ethtool if using l2fwd or linux bridge
as the loopback application when vanilla ovs multi-queue
is enabled.
Modified dpdk setting in vswitch conf to be better worded
for separation from vanilla ovs multi-queue.
Updated release and installation doc for new vloop image
that includes required ethtool utility.
JIRA: VSPERF-373
Change-Id: Idb550515190b1a93390308c11f54da368f962512
Signed-off-by: Christian Trautman <ctrautma@redhat.com>
|
|
Adds boot and mounted drive configuration options.
JIRA: VSPERF-379
Change-Id: Ic0673a9b9f7a3430d7ac4db5fb9bf4228872f501
signed-off-by: Christian Trautman <ctrautma@redhat.com>
|
|
Support of vHost Cuse interface is not supported
by OVS anymore, so it was removed also from VSPERF.
Cuse can be used with older VSPERF release if needed.
JIRA: VSPERF-376
Change-Id: Id786bddf5fa9dce5706a93de7b089707a60317cd
Signed-off-by: Martin Klozik <martinx.klozik@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Christian Trautman <ctrautma@redhat.com>
Reviewed-by: Bill Michalowski <bmichalo@redhat.com>
|
|
|
|
Implementation of 3 testcases with multi-VMs in parallel.
Flows are explicitly set on UDP destination ports so that
each packet is sent to a specific VM.
Topology is: Phy1-VM1-Phy2, Phy1-VM2-Phy2,..
This patch requires change #16855 "ovs/ofctl: Fix validation
method for complex flows."
JIRA: VSPERF-326
Change-Id: I59d23354bc9c6db470a59faba13bdcd6f2beb19d
Signed-off-by: Antonio Fischetti <antonio.fischetti@intel.com>
Reviewed-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Martin Klozik <martinx.klozik@intel.com>
|
|
The test case for verifying the vxlan tunneling feature without using any
ingress tunnel traffic. The virtual switch(OVS) is configured to mangle and
generate tunnel traffic in the deployment. The packet flow in the test case is
as follows
TRAFFIC-IN --> [ENCAP-PKT] --> [MOD-PKT] --> [DECAP-PKT] --> TRAFFIC-OUT
ENCAP-PKT - Encapsulate the ingress packet to a tunnel type.
MOD-PKT - Modify the tunnel header to match the following decap interface.
DECAP-PKT - Decapsulate the newly generated tunneled packet.
Change-Id: Ie24bacb3cb1c069bd60403e5a4ef8bcdf0e12e54
Signed-off-by: Sugesh Chandran <sugesh.chandran@intel.com>
|
|
Enables multi-queue option for OVS versions 2.5.0 or less with
dpdk. Instead of enabling MQ on a port by port basis this sets
the global MQ option after starting the ovs switch.
JIRA: VSPERF-372
Change-Id: I245c76779695518d0005122d8214bbf3f141bb96
Signed-off-by: Christian Trautman <ctrautma@redhat.com>
|
|
Change-Id: I8a89ee3eedd3b439f0243eaef2fafc50ee807739
Signed-off-by: Maryam Tahhan <maryam.tahhan@intel.com>
|
|
|
|
Fixes minor grammar and link issues.
JIRA: VSPERF-349
Change-Id: Id8708aee5b099287b5c99e79ca6da9ab8778897f
Signed-off-by: Christian Trautman <ctrautma@redhat.com>
|
|
Added changes to make use of the tests parameter.
The change is minor.
Prior to this the value was taken from configuration file.
With this change, the tests value is used as passed.
JIRA: VSPERF-367
Change-Id: I9d1553d3b53dd19a79428ddaaacadb55fc27324f
Signed-off-by: Sridhar Rao <sridhar.rao@spirent.com>
|
|
Fix build issues and add the LTP to index.rst.
JIRA: DOCS-95
Change-Id: I2d22646af935c544645c01600b92a5f51ede39ec
Signed-off-by: Maryam Tahhan <maryam.tahhan@intel.com>
|
|
Updating MoonGen configuration guide
JIRA: VSPERF-370
Change-Id: I577491a6ece249a6162e1a3595a8682b826e5734
Signed-off-by: bmichalo <bmichalo@redhat.com>
|
|
Fixes hugepage free calculation when hugepage size is not
1GB. Calculation previously assumed 1GB in size. This change
allows for all sizes to be calculated properly.
Change-Id: Ib05800f51992e9757c412a66296a47ad29e88a3b
Signed-off-by: Christian Trautman <ctrautma@redhat.com>
|
|
|
|
|
|
|
|
|
|
Change-Id: Id0b2ce7460e76ffb3d478bd255a7e5b8e75ba001
Signed-off-by: Maryam Tahhan <maryam.tahhan@intel.com>
Reviewed-by: Al Morton <acmorton@att.com>
Reviewed-by: Christian Trautman <ctrautma@redhat.com>
Reviewed-by: <bmichalo@redhat.com>
Reviewed-by: <sridhar.rao@spirent.com>
|
|
During testcase development on the integration side, if
namespaces are used or veth ports and an exception is raised.
The namespaces and veth ports are not properly cleanup up
because the run_finalize does not get executed.
This patch wraps the testing block in a try block where
a finally will ALWAYS execute the run_finalize thus
preventing the possiblity of leftover testing artifacts.
JIRA: VSPERF-328
Change-Id: I4ea5f29a48b9171f330847f34b5b30ef3ab22f27
Signed-off-by: Christian Trautman <ctrautma@redhat.com>
|
|
Adds options to Xena for testing options when running RFC 2544
throughput testing.
JIRA: VSPERF-369
Change-Id: I3145bd1a4d697d7b06cafda4f9d77cce859d1839
Signed-off-by: Christian Trautman <ctrautma@redhat.com>
|
|
|
|
|
|
Instead of hard coding line speed numbers for rate calcuation,
MoonGen should use its class object initialized at start time.
JIRA: VSPERF-368
Change-Id: I68c688f8c244df16d98cfd15efabf491a317256b
Signed-off-by: bmichalo <bmichalo@redhat.com>
|
|
One of the new tests we agreed by working on Etherpad
was MatchAction. Maryam led the development, and many
provided comments there. This is another chance for
comments before we merge.
JIRA: VSPERF-242
Change-Id: I9da74efb703d5f221280567159e4ed6d830eb92c
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: Christian Trautman <ctrautma@redhat.com>
|
|
|
|
Change the default number of allocated pages to the min 10GB. Deallocate
hugepages if we allocated them on the fly.
Change-Id: If9946cec423dd095c1ca1ab3bee3698e2f025047
Signed-off-by: Maryam Tahhan <maryam.tahhan@intel.com>
|
|
Change hugepage detection to check if hugepages are configured. Also
verifies enough free hugepages are available based on test
configuration. Will result in an exception if not enough hugepages
are found. This performs the check based on the guest memory configs
as well as the dpdk socket configs.
Adds some new methods to check for free hugepages based on all
hugepages or for socket free hugepages.
Adds new logging indicitaing needed and free hugepages as well as info
level event if hugepages are not configured at recommended 1GB size.
JIRA: VSPerf-255
Change-Id: I80d5605308861070c412568d8d40339f01b9718b
Signed-off-by: Christian Trautman <ctrautma@redhat.com>
|
|
|
|
For MoonGen, the final results parameters: throughput_rx_percent,
throughput_tx_percent were incorrectly calculated when converting
MoonGen to VSPERF format. In particular, this would occur when more
than one data stream (single channel unidirectional traffic) was used in
the test. When calculating these percentages, only aggregate throughput
values equaling unidirectional traffic were used (for example, for 64
byte frames bidirectional traffic should use 29.76 Mpps, not 14.88 Mpps)
JIRA: VSPERF-341
Change-Id: Ie62bded7aea023c287697f18c4172b58c320e050
Signed-off-by: bmichalo <bmichalo@redhat.com>
|
|
|
|
|
|
|
|
|