Age | Commit message (Collapse) | Author | Files | Lines |
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I56987d744d9143a95954d85f2557cda07220c681
|
|
Fix "Not enough availablehuge pages: 1483!".
Also update 3n-tsh docs.
Change-Id: I1d37a66af1e2363f77fdbd87d238e8ff5535b011
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech>
|
|
Update after switching one 1n-tx2 and the idle ThunderX2 servers between
racks. Update the idle ThunderX2 specs to a new perf testbed, 2n-tx2.
Add Server-Type-B12 which is a modified Server-Type-B2 with one extra
NIC (needed for 2n-tx2).
Change-Id: I51af358f1feb476652eddfe82b5af1d0d70ac259
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech>
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: Ie48a96d83d37d7292d261875371e09d4b9152c7b
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: Id56b87ab868f2897a6563914b0beca2acc25e706
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I145a4b5511141f1e2b4e387daa358e32dd2c8015
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: Id14c3f2f4f8689256172dc2b3ebc4fbaed5de8d3
|
|
Force check is doing its job but vt.handoff was deployed later.
vt.handoff (vt = virtualterminal) is a kernel boot parameter unique to
Ubuntu, and is not an upstream kernel boot parameter. Its purpose is to
allow the kernel to maintain the current contents of video memory on a
virtual terminal. So, when the operating system is booting up, when it
moves past the boot loader, vt.handoff allows showing of an aubergine
background, with Plymouth displaying a logo and progress indicator bar
on top of this. Once the display manager comes up, it smoothly replaces
this with a login prompt.
Useless
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I9d6db8833ccbef680fab2b643e9c3525bc709244
|
|
Change-Id: Ib668674a2a267d2ceed458288d21181b2a937778
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I8abfc47e5e816e2ab4b39d7ad7575e672ae19ca6
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I3954157e91aafd370c3ac0843708903d19b84936
|
|
One ThunderX2 9975 server (.69) was replaced with two ThunderX2 9980
(.70, .71) servers. Move the .69 server under ansible perf section in
anticipation of repurposing it for that purpose. Update the ansible
scripts with .70 and .71 config and rename port names in device.sh lib
to reflect the NIC differences between .69 and .70 (and .71).
Change-Id: I88b75648735243e5559175d3192ffcc8fc70071c
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech>
|
|
- From 2n-zn2 testing
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I9b1f0916f0f1d90a223918cfe48409d29f2ee773
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: Ic84b7cbcbcd269a5fe548b59d478cd6dfa57a952
|
|
+ DPDK 20.08
+ Migrate make -> meson
+ Fix all trending issues
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I31dcb22627c0f8d17ec63c5b138a2da958b006f4
|
|
The src field for ansible template module used in vpp_device.yaml of cleanup role should be jinja2 format and the mode for script to be transfered to the remote host should be executable for the file owner. Fix this error by replacing the ansible template module with ansible copy module and setting the file permission to be 744.
Change-Id: Ibf80b0c5bec77a13509122795a5b12b6faba2f8e
Signed-off-by: Jieqiang Wang <jieqiang.wang@arm.com>
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I99a33a15e7e87fe6f56deb1ab0e4ce3091508244
|
|
+ Bump T-Rex version. We need new features for ASTF test.
+ Apply core pining. Results in a more stable performance.
+ Tweak the number of T-Rex workers.
+ We need an even value to achieve ymmetric performance with pinning.
+ Value 8 was selected as a best compromise.
This is a combination of 3 commits.
This is the 1st commit message:
T-Rex: 2.82
This is the commit message #2:
Change Trex to CORE_MASK_PIN mode to improve performance
https://trex-tgn.cisco.com/trex/doc/trex_stateless.html#_core_masking_per_interface
Above link have below explaination,
"When the profile is symmetric, performance can be improved by pinning half of
the cores to port 0, and half of the cores to port 1, thus avoiding cache
trashing and bouncing."
The reason to change this is that to run CSIT with 100G NIC often failed with
"TRex stateless runtime error timeout", it caused by Trex can not send enough
traffic within the fixed duration.
by change to CORE_MASK_PIN mode fix the issue.
Not editing ASTF, as that supports different options.
This is the commit message #3:
Experiment: Vary number of TRex workers
With CORE_MASK_PIN, we can get more predictable time distribution.
Decided to use 8 workers, that gives good results
both for high end (RDMA-core l2patch) and low end (vhost) tests.
Change-Id: I5c61127799e0624464e960fcb980ad1c4058e744
Signed-off-by: pmikus <pmikus@cisco.com>
Signed-off-by: Yulong Pei <yulong.pei@intel.com>
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
+ random fixes
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: Icb320d1ba4bbe3b5f99956cf8e89c2a29e114e41
|
|
Signed-off-by: Xiaolong Jiang <xiaolongx.jiang@intel.com>
Change-Id: I275ad9dccc02e6b1f1c791fabf8896fb03acb3fb
|
|
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
Change-Id: I0549e9b5ac6769e3fc760ab2047d89a188baa36a
|
|
Change-Id: If61783fb717757c6189f06924412bd079e15a08f
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: Ie304118e6b1ecb06704e35bbbeccf7cc0f649e8b
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech>
|
|
Remove nomad installation from the ARM dev server and add a generic
vppdev user to it. Add ansible role dev for this.
Change-Id: Icb967ff18cbe939464ddb3d8dbce5101747e739f
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech>
|
|
- Add three new ThunderX servers
- Mark two existing ThunderX servers as out of order, since
they're inaccessible
- Fix microarchitecture of ARM build servers
Change-Id: I57f5f3f8436a9082a264eb4e0fa4faac77ad53be
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech>
|
|
Signed-off-by: Maros Mullner <maros.mullner@pantheon.tech>
Change-Id: Idc9317d3e37d9ed0a82adc273e0628a8defcfdf6
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I24a22b1492593df937af2a17cd0c8684af7cd500
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I2560e3e12b457413db81a7ecc52efd7e39f1aea2
|
|
+ Prometheus Telemetry stanza
- ARM Softirons
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: Ib7bf577ed0460ab5b79700ff0f8d52d5704cec59
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: Iffdb12cc7013a38341a09df2f863ee7c56196bb0
|
|
Signed-off-by: Maros Mullner <maros.mullner@pantheon.tech>
Change-Id: Ic78b5dae22b67763b99c50a5f15d7dd16152d5d5
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I67ab84c2c2b34ce5ccc3a2de629c41460bb52e8c
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I6c37287e79e48ba40a5dda497cc3614c367230aa
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: Iaa5c02d14223b578c316f48e31a0b3613303eecf
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: Icc03bf62d8639f603ca90aa6ab849a46dc911d56
|
|
+ As per last discoveries when running from_scratch vs. append_only
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I8b44314185bbc157a67136730dc6146d133c7cc4
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I69d9290d42c154f54e71e769a3a5d89a69c2e602
|
|
Signed-off-by: pmikus <pmikus@cisco.com>
Change-Id: I542401d77a777dfc2d930de626e19c1d8083941c
|
|
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: I745ceb519220c4d08951dff07350f2f4709f3533
|
|
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: I34446125071ee7157e1c0e29027a6d973ee03c7e
|
|
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: Iee65272169339ed0e5561cf3dee11567335b12be
|
|
- MLX, ENA
- Part I
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: I9f4b407770515fe6009ef5dfdba2b27fa2ff654f
|
|
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: I87020a6a26f54f9a1591a540f77f6469d75df007
|
|
By default irqbalance should be honoring isolcpu settings but as of
latest finding IRQs are distributed ad-hoc even if irqbalance is:
- enabled: with default settings
- disabled: by configuration
- disabled: as a service
- enabled: with incorrect mask
This patch brings explicit configuration of IRQBALANCE_BANNED_CPUS with
mask large enough to cover all cores and keeps IRQ to be processed by
core 0.
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: I0eb36d780feb284a5493e123acbf993a362ea9ec
|
|
Router advertisement from linux interfaces in VPP Device TGs could
result in unsolicited packets arriving in DUT which would produce false
negatives.
Change-Id: I121a92310a8fe47f36e253656e126123704efc0a
Signed-off-by: juraj.linkes <juraj.linkes@pantheon.tech>
|
|
Signed-off-by: Maros Mullner <maros.mullner@pantheon.tech>
Change-Id: Ifd0f4d09b7c518feab534186220d618a560f223c
|
|
Signed-off-by: Maros Mullner <maros.mullner@pantheon.tech>
Change-Id: I071c5184dcbc25a8bc6b53d65d9a3478194c72fc
|
|
Modified cpuutils numa to work with aws 18xlarge instances.
Added dmidecode package to calibration role as is not installed by default.
Added required constants
Signed-off-by: Maros Mullner <maros.mullner@pantheon.tech>
Change-Id: I99608d9df3cff0b085ad26c30e339216b17581eb
|
|
- There is currently no other way to support multiple version of
playbooks.
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: I2ade4a57d726b793e422996e83bf767ac4315488
|
|
Signed-off-by: Maros Mullner <maros.mullner@pantheon.tech>
Change-Id: I7eefdb5db39a0f5328c7f1a0a0a5d57c762cc21c
|