#
April 24, 2025
Attendees
* Patrick Robb
* Luca Vizzarro
* Paul Szczepanek
#
Minutes
=
Gener
Hello all,
As has been discussed at the DPDK CI Meetings, Shai Brandes of AWS has been
working to set up the infrastructure to do per-patch testing for DPDK on
some AWS x86 and ARM environments. Currently they are doing a DPDK build +
dpdk-test fast tests on those environments, and reporting those
#
April 17, 2025
Attendees
1. Patrick Robb
2. Paul Szczepanek
3. Luca Vizzarro
4. Aaron Conole
5. Manit Mahajan
#
Minutes
===
I propose this patch for DPDK to enable coexistence between
DPDK and kernel drivers for regular NICs.This solution requires
adding a new pci_ops in rte_pci_driver, through which DPDK will
retrieve the required IOVA mode from the vendor driver.
This mechanism is necessary to handle different IOMMU
c
This is perfect for my TREX implementation!
Reviewed-by: Nicholas Pratte
On Fri, Mar 14, 2025 at 9:19 AM Luca Vizzarro wrote:
>
> Add BlockingDPDKApp class. Some non-interactive applications are
> blocking and run until the user interrupts them. As their main intended
> usage is to be kept runn
Rx hairpin queues support VLAN stripping,
but if port was started and application attempted
to configure stripping on hairpin queue,
segfault was triggered because of NULL dereference.
Underlying function, which was updating the RQ was passing
wrong object handle for hairpin queues.
This patch fixe
Reviewed-by: Nicholas Pratte
On Fri, Mar 14, 2025 at 9:19 AM Luca Vizzarro wrote:
>
> When making any command a privileged one in a LinuxSession, there
> currently is no consideration whether this command already includes single
> quotes. Therefore escape the existing single quotes before making
HW modify header commands generated for multiple modify field
flow actions, which modify/access the same packet fields
do not have to be separated by NOPs when used on group 0.
This is because:
- On group > 0, HW uses Modify Header Pattern objects
which require NOP explicitly.
- On group 0, modi
Add missing validation of GTP PSC QFI flow field width for
modify field flow action.
Fixes: 0f4aa72b99da ("net/mlx5: support flow modify field with HWS")
Cc: suanmi...@nvidia.com
Cc: sta...@dpdk.org
Signed-off-by: Dariusz Sosnowski
Acked-by: Bing Zhao
---
drivers/net/mlx5/mlx5_flow_dv.c | 2 ++
During counter pool initialization (mlx5_hws_cnt_pool_create()),
a background service thread is started by mlx5 PMD,
which refreshes counter values periodically.
During initialization it might happen that:
- background thread creation succeeds,
- pool creation fails (e.g. due to error in configur
In case of an error in mlx5_hws_cnt_pool_create(),
value stored in ret is used as "error code".
There are however a few cases inside this function,
when this variable is not set, leading to failed
assertions (e.g., when requested number of counters is bigger
than maximum supported).
This patch add
This patch series includes several fixes for flow counter pool
used with HW Steering flow engine.
Dariusz Sosnowski (2):
net/mlx5: fix counter pool init error propagation
net/mlx5: fix counter service thread init
drivers/net/mlx5/mlx5_hws_cnt.c | 29 +++--
1 file chan
Bruce,
rte_eth_stats_get() on Intel NICs seems slow to me.
E.g. getting the stats on a single port takes ~132 us (~451,000 CPU cycles)
using the igb driver, and ~50 us using the i40e driver.
Referring to the igb driver source code [1], it's 44 calls to E1000_READ_REG(),
so the math says that e
Hello all,
I tested v1 patch on CX-4 card and I can confirm my application boots now!
For traceability I am adding the original discussion thread:
https://mails.dpdk.org/archives/users/2025-April/008242.html
Probably a some other problem, but it still outputs these logs:
Config: dpdk: :b3:0
Enqueue generated counter IDs on a ring in bulk.
Generate them and store in an array before putting them
on a ring all at once. That bring better cache access
and speeds up the mlx5_hws_cnt_pool_create() function.
Signed-off-by: Alexander Kozyrev
---
drivers/net/mlx5/mlx5_hws_cnt.c | 19
Reviewed-by: Nicholas Pratte
On Fri, Mar 14, 2025 at 9:19 AM Luca Vizzarro wrote:
>
> Turn the `path` attribute of InteractiveShell into a method property.
> This allows path to both be defined statically by the class
> implementation and also to be defined dynamically as part of the class
> con
Hi,
I am having a VM on Azure where I have got two 'accelerated networking'
interfaces of Mellanox
# lspci -nn|grep -i ether
6561:00:02.0 Ethernet controller [0200]: Mellanox Technologies MT27710
Family [ConnectX-4 Lx Virtual Function] [15b3:1016] (rev 80)
f08c:00:02.0 Ethernet controller [0200]:
Reviewed-by: Nicholas Pratte
On Fri, Mar 14, 2025 at 9:19 AM Luca Vizzarro wrote:
>
> Previously InteractiveShell was split into two classes to differentiate
> a shell which execution must be controlled in a tight scope through a
> context manager, from a more looser approach. With the addition
Reviewed-by: Nicholas Pratte
On Fri, Mar 14, 2025 at 9:19 AM Luca Vizzarro wrote:
>
> Enable the ShellPool class to be used in the test run, and let
> InteractiveShells register themselves to the pool upon shell startup.
> Moreover, to avoid the ShellPool to call InteractiveShell.close more
> th
https://bugs.dpdk.org/show_bug.cgi?id=1700
Bug ID: 1700
Summary: BPF callback wait is not MP safe
Product: DPDK
Version: 25.03
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: minor
Priority:
Reviewed-by: Nicholas Pratte
On Fri, Mar 14, 2025 at 9:19 AM Luca Vizzarro wrote:
>
> Multi inheritance has proven to be flaky in Python, therefore revert
> back to a simpler approach where classes will only inherit one base
> class. As part of this change, instead of making the
> ScapyTrafficGe
Short answer Accelerated networking on Azure is not designed to support
bifurcated VF usage
On Fri, Apr 25, 2025, 10:47 Prashant Upadhyaya
wrote:
> Hi,
>
> I am having a VM on Azure where I have got two 'accelerated networking'
> interfaces of Mellanox
> # lspci -nn|grep -i ether
> 6561:00:02.0
22 matches
Mail list logo
