Age | Commit message (Collapse) | Author | Files | Lines |
|
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: I3d9074175e1aaf5afba4e351790644df4ad33914
|
|
Change-Id: I7e3ed2a7ad75617be1ef5a80e345e7e62a6da3df
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: Ib451636d46517cd85839d3e3b9aca9b6958b322e
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: Idd6bf8b89ce0fb5ebf9734b39a8f6058861b8ec5
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Last argument to Vpp Add Ip Neighbor should be a MAC address.
Change-Id: Ia93ddf68aa2b8647c66b6a72210db3a8a01b81fb
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: I84b3c07e22a313e96ac59fc7818960c502507651
|
|
+ Rework BASH scripts (more code to python).
+ Move BASH into libraries.
+ Allows RDMA usage.
+ Fix 9000B tests.
+ Rename confusing l2fwd -> testpmd.
+ Fix suite setup.
+ Fix PCI whitelist to not accidentally pickup wrong interface.
+ Fix deprecated DPDK cli arguments.
- MLX5 jumbo are disabled on NIC (i will increase separately).
https://jenkins.fd.io/job/csit-dpdk-perf-verify-master-2n-clx/6/console (l3fwd still broken)
- MLX5 IMIX seems to be some TRex issue with IMIX for mlx5 (i will handle separately)
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: I31d1b67305fa247cb5e1f57e739d3ef30dc1a04b
|
|
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: I0e773060c344fda7b06937a81a53863a96a4373a
|
|
Change-Id: I91677a0f6e1c62633977c80fc8dec8019e580d0c
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: Ib965b768f725a4d4bc09cb1f995a452ed6e64653
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: Ic305d135b14697332436d6afd76173eff0c3874b
|
|
Change-Id: Ia1e658347051ea0476cbea76361783ab7f6d6d1d
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: I0446263333812a57a8d0f7f288044482bee04a14
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
Change-Id: I57fcb06c72caf517231f09478d9d6089d2f46af4
Signed-off-by: Jan Gelety <jgelety@cisco.com>
|
|
Change-Id: I809ee89a5798c36d883fd49de75b6f67f1d336f8
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: I18df883d66932fdafba83fd1b831736c4cb9ffa9
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
+ FIX for Vratko's FIX
Change-Id: Ia6ff137120d84bdd2b9e02e3e935c62d1c732b7b
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
On 3n-hsw, xl-710 is numa 1.
The old code returned 0, causing performance regression.
Fixes: 14a71b74b414df7616ccb5ada3d50ecb90d96bae
Change-Id: I81d3f711b6a37438f6411e29baafae9dec1c00a4
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
Change-Id: I41867bdb395a72733c1f616b981e952087b017ef
Signed-off-by: Jan Gelety <jgelety@cisco.com>
|
|
+ some pylint fixies
Change-Id: I650ce16282ae953a1a5ee96e810702c01f71efd6
Signed-off-by: Jan Gelety <jgelety@cisco.com>
|
|
From trending it seems like TRex threads use thread-local PRNGs,
resulting in multiple packets (one from each thread) for the same flow.
Hopefully, without a seed set, each thread will pick a different seed.
We are risking some threads will still pick the same seed occasionally,
but I doubt TRex API has a better way to prevent this issue.
Change-Id: Ie6ed42669d6e2463627760f89d2426d4e6bf0155
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
Change-Id: Ic6cdabf65846a0d7d1e38025a1a7459947aed385
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: Id7267f5ab13a362894c5fe5a2ae40d51d4b5a36d
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
Change-Id: I4bd56d17b8f28ca40424067068384bae0447db4b
|
|
- use new vpp ref build - ubuntu 18.04:
20.05-rc0~532-g4fde4ae03
- use new vpp ref build - centos7:
20.05-rc0~532_g4fde4ae~b9253
Change-Id: If6a5478ca9b89050beae8f2dc4ce4e91b328aa21
Signed-off-by: Jan Gelety <jgelety@cisco.com>
|
|
Change-Id: Ie2f11cd5144daf249200fe07430b0809185bccce
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: I909845026c0495db74e0ef82f17d7d14db2ac9d5
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: I276f9bb6aaf03c897ef255d124a3204dee45abf6
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: Ibadf4fabb69bdd9d2412ed8f33c97ab12303800f
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
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
|
|
- hoststack
- replace old 20001 with new 2001
Change-Id: Id1913527a0878b56fbb3c1372da6251cebd5e30b
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: I6bdce8156ba214be210c195a9f3effd718301a86
|
|
Change-Id: Ib5d8f9f0c8e8920a2d19e89d4d5518f4882ff32d
Signed-off-by: Jan Gelety <jgelety@cisco.com>
|
|
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: I3ad5675cda017b3dcf8cba88d56d5d6c11184e20
|
|
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: I87020a6a26f54f9a1591a540f77f6469d75df007
|
|
+ Merge single/double link
+ Introduce _pf{n}[0] variables so we can access physical function
same way as virtual function
+ Cleanup code by moving complex logic to python
+ Prepare code for multiple vf functions
Signed-off-by: Peter Mikus <pmikus@cisco.com>
Change-Id: Ic2e74a38bfa146441357de8f0916aeb638941c49
|
|
Change-Id: I4926a15e0b89507aad89845e12385ba34adfd68d
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: I3aba33e90fd99fd2277358f184a34d071ed4e805
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: I9b8377e336651f850d50d050471ebefd5647993d
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: Iff28b68e4fb33b304433ada64ea6ef2d751e4c3f
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Signed-off-by: Jan Gelety <jgelety@cisco.com>
Change-Id: I2e72b7e136d3907a038cb7729c11d155a12c486f
(cherry picked from commit 9e2e4eaa2c6854eee47f7057d16b75339f453f89)
|
|
Change-Id: I05a61b075c467681f2e05b17f0245a5e77a911b5
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: I144d1ee15505a9d08acdacb50f53df02420d60a8
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: Iafff8f88e1d54ede712d8fc3b4afa88172f599ab
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
The previous values lead to duration stretching,
which affects MRR results (making receive rate bigger than it really is).
+ Move timestamp for starting of traffic,
the previous position was too conservative.
Change-Id: Ie983f0d3eacf56d57caa551d1b7fbb87ece6c158
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
|
|
Change-Id: I901e20ec6bc7214bf3376a57aee3eba4149ec8a1
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
Change-Id: I9bea40d22f9a4aa4f5161f203ba5cc04685a8b19
Signed-off-by: Tibor Frank <tifrank@cisco.com>
|
|
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>
|