RE: [PATCH] net/intel: enable ethertype filter for E610

> -Original Message- > From: Kaiwen Deng > Sent: Wednesday, April 23, 2025 1:54 PM > To: dev@dpdk.org > Cc: sta...@dpdk.org; Deng, KaiwenX ; Burakov, > Anatoly ; Medvedkin, Vladimir > ; Wenzhuo Lu ; Wei > Zhao > Subject: [PATCH] net/intel: enable ethertype filter for E610 > > This commit

[PATCH v3 4/4] app/testpmd: support disable DCB command

After the "port config 0 dcb ..." command is invoked, no command is available to disable DCB. This commit introduces disable DCB when num_tcs is 1, so user could disable the DCB by command: port config 0 dcb vt off 1 pfc off Signed-off-by: Chengwen Feng --- app/test-pmd/cmdline.c

[PATCH v3 3/4] app/testpmd: support don't adjust queue num in DCB command

In some test scenarios, users want to test DCB by specifying the number of Rx/Tx queues. But the "port config 0 dcb ..." command will auto adjust Rx/Tx queue number. This patch introduces an optional parameter "keep-qnum" which make sure the "port config 0 dcb ..." command don't adjust Rx/Tx queue

[PATCH v3 1/4] app/testpmd: remove restrict of number of TCs in DCB command

Currently, the "port config 0 dcb ..." command only supports 4 or 8 TCs. Other number of TCs may be used in actual applications. This commit removes this restriction. Cc: sta...@dpdk.org Signed-off-by: Chengwen Feng --- app/test-pmd/cmdline.c | 4 ++-- doc/guides/testpmd_a

[PATCH v3 2/4] app/testpmd: support config prio-tc map in DCB command

Currently, the "port config 0 dcb ..." command config the prio-tc map by remainder operation, which means the prio-tc = prio % nb_tcs. This commit introduces an optional parameter "prio-tc" which is the same as kernel dcb ets tool. The new command: port config 0 dcb vt off 4 pfc off prio-tc 0:1

[PATCH v3 0/4] enhance testpmd DCB command

This patchset enhance the testpmd DCB command features: 1) remove restrict of number of TCs (currently its 4 or 8) 2) support config prio-tc map 3) support don't adjust queue num 4) support disable DCB Chengwen Feng (4): app/testpmd: remove restrict of number of TCs in DCB command app/testpmd:

[PATCH 2/2] dts: add port control testing suite

From: Jeremy Spewock This patch ports over the port_control test suite from the legacy framework and adapts the functionality to fit with the current testing framework. The test suite provides validation of basic port control functions such as starting, stopping, and closing ports. It should be n

[PATCH 1/2] dts: add method for closing ports to testpmd

From: Jeremy Spewock Closing ports is a standard configuration feature that is available in testpmd but the framework lacks the ability to access this command through the testpmd API. This patch adds a method that performs this action and verifies the results of sending the command to allow devel

[PATCH 0/2] dts: port over port_control testing suite

This series ports over most of the test coverage provided from the port_control testing suite in the Old DTS framework. The only functionality that is missing is testing port functions in a VM through QEMU and testing the support of resetting ports. Since we have no method of handling virtual machi

[PATCH v2 13/45] drivers: enable support for AMD Solarflare X4 adapter family

From: Denis Pryazhennikov Enable support for Medford4 (X4) adapters X4522 and X4542. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton --- drivers/common/sfc_efx/efsys.h | 4 ++-- drivers/net/sfc/sfc_ethdev.c | 3 +++ 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a

Re: [PATCH v9 4/5] app/graph: add custom feature nodes for ip4 output arc

Hi Nitin, At the DPDK Community Lab, we inadvertently let CI testing on your patch run concurrently with some system maintenance we were doing on testbeds yesterday, and I think the testing on your patch was affected (so you got a couple fails when you should not have). I'm going to put in some re

[RFC Patch v1 3/5] dts: add asychronous support to ssh sessions.

Execution of the TREX server process requires an SSH session rework to support asynchronous process management. Allowing access to asynchronous functionality allows developers to execute processes without hijacking the SSH session being used. In doing so, both timeout and runtime errors may be avoi

[PATCH v2 07/45] common/sfc_efx/base: add Medford4 support to MAC module

From: Denis Pryazhennikov Use common EF10 method table for that. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton --- drivers/common/sfc_efx/base/efx_impl.h | 1 + drivers/common/sfc_efx/base/efx_mac.c | 7 +++ 2 files changed, 8 insertions(+) diff --git a/drivers/common/sfc

Re: [PATCH] ethdev: support RSS based on RoCEv2 header

On Tue, 22 Apr 2025 10:44:59 +0530 wrote: > From: > To: Aman Singh , Thomas Monjalon > , Ferruh Yigit , Andrew Rybchenko > > CC: , Kiran Kumar K > Subject: [PATCH] ethdev: support RSS based on RoCEv2 header > Date: Tue, 22 Apr 2025 10:44:59 +0530 > X-Mailer: git-send-email 2.48.1 > > From:

Re: [PATCH v2 08/45] common/sfc_efx/base: add Medford4 support to PHY module

On 4/23/25 18:59, Ivan Malov wrote: From: Denis Pryazhennikov Use common EF10 method table for that. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton --- drivers/common/sfc_efx/base/efx_phy.c | 6 ++ 1 file changed, 6 insertions(+) diff --git a/drivers/common/sfc_efx/base

[RFC Patch v1 5/5] dts: add performance test functions to test suite api

Provide functional performance method to run performance tests using a user-supplied performance traffic generator. The single core performance test is included, with some basic statistics checks verifying TG packet transmission rates. Bugzilla ID: 1697 Signed-off-by: Nicholas Pratte --- dts/fra

[RFC Patch v1 4/5] dts: add trex traffic generator to dts framework

Implement the TREX traffic generator for use in the DTS framework. The provided implementation leverages TREX's stateless API automation library, via use of a Python shell. As such, version control of TREX may be needed. The DTS context has been modified to include a performance traffic generator i

[RFC Patch v1 1/5] dts: rework config module to support perf TGs

Rework test run configuration file for TGs to support both application directory location and any necessary configuration files; an example TREX configuration file is provided. Configuration files have been moved to a configurations directory, requiring a slight modification to the settings module.

[RFC Patch v1 2/5] dts: rework traffic generator inheritance structure.

Rework TG class hierarchy to include performance traffic generators, in addition to capturing traffic generators. As such, methods garnered to capturing traffic have been moved to the CapturingTrafficGenerator subclass. Bugzilla ID: 1697 Signed-off-by: Nicholas Pratte --- .../capturing_traffic_g

[RFC Patch v1 0/5] Add TREX Traffic Generator to DTS Framework

Included in a semi-complete, RFC implementation for a would-be implementation of the TREX traffic generator, in addition to a mock implementation of a single core performance test suite, leveraging newly added performance test API functionality. Code is incomplete with a handful of typing issues,

[PATCH v2 17/45] common/sfc_efx/base: extend list of supported X4 port modes

From: Denis Pryazhennikov Add X4 port mode that allocates two 10G/25G/50G ports to separate cages. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton Reviewed-by: Richard Houldsworth Reviewed-by: Ivan Malov --- drivers/common/sfc_efx/base/ef10_nic.c | 1 + 1 file changed, 1 insert

Re: [PATCH v2 00/45] Support AMD Solarflare X45xx adaptors

On 4/23/25 18:59, Ivan Malov wrote: New X4522 (dual port SFP56) and X4542 (dual port QSFP56) adaptors are Medford4 (X4) chips that are based on EF10 architecture. An X4 NIC supports multiple network engine types. This series provides support only for the Medford2-alike, 'full-feature' (FF) networ

[DPDK/DTS Bug 1697] Add TREX Traffic Generator Functionality & Single Core Performance Test.

https://bugs.dpdk.org/show_bug.cgi?id=1697 Bug ID: 1697 Summary: Add TREX Traffic Generator Functionality & Single Core Performance Test. Product: DPDK Version: unspecified Hardware: All OS: All

Re: [PATCH v2 45/45] doc: advertise support for AMD Solarflare X45xx adapters

On 4/23/25 19:00, Ivan Malov wrote: The two NICs, X4522 and X4542, are based on EF10 architecture. Signed-off-by: Ivan Malov --- doc/guides/nics/sfc_efx.rst| 9 - doc/guides/rel_notes/release_25_07.rst | 4 2 files changed, 12 insertions(+), 1 deletion(-) diff --gi

[PATCH v2] net/netvsc: add stats counters from VF

From: Long Li The netvsc driver should add per-queue and rx_nombuf counters from VF. Fixes: 4e9c73e96e83 ("net/netvsc: add Hyper-V network device") Cc: sta...@dpdk.org Signed-off-by: Long Li --- Change log: v2 is a resend, there is no change to v1 drivers/net/netvsc/hn_ethdev.c | 10 +

[PATCH v2 23/45] common/sfc_efx/base: get netport fixed capabilities on probe

These make the basic subset of PHY capabilities known to EFX Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/sfc_efx/base/efx.h| 21 ++- drivers/common/sfc_efx/base/efx_np.c | 209 +++ 2 files changed, 22

[PATCH v2 15/45] common/sfc_efx/base: add port mode for 8 port hardware

From: Andy Moreton Add support for 8 port mode and adjust the bus bandwidth computation and external port mapping table. Signed-off-by: Andy Moreton Reviewed-by: Ivan Malov --- drivers/common/sfc_efx/base/ef10_nic.c| 20 +++ drivers/common/sfc_efx/base/ef10_tlv_layout.

[PATCH v2 11/45] common/sfc_efx/base: add Medford4 support to Rx module

From: Denis Pryazhennikov Use common EF10 method table for that. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton --- drivers/common/sfc_efx/base/efx_rx.c | 6 ++ 1 file changed, 6 insertions(+) diff --git a/drivers/common/sfc_efx/base/efx_rx.c b/drivers/common/sfc_efx/base/

[PATCH v2 41/45] common/sfc_efx/base: subscribe to netport link change events

Subscribe to the new link events on netport MCDI attach path. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/sfc_efx/base/efx_np.c | 56 1 file changed, 56 insertions(+) diff --git a/drivers/common/sfc_

[PATCH v2 28/45] common/sfc_efx/base: provide PHY link get method on Medford4

To do so, make use of new netport MCDI supported by Medford4. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/sfc_efx/base/ef10_phy.c | 8 ++ drivers/common/sfc_efx/base/efx.h | 4 +- drivers/common/sfc_efx/base/efx_

[PATCH v2 36/45] common/sfc_efx/base: correct MAC PDU calculation on Medford4

For managing MAC PDU (max. frame size), client drivers apply EFX macros to switch between PDU and SDU forms. These macros include a workaround for a bug that dates back to Siena NICs. Starting with Medford4, the bug is no longer there and it is wrong to use the macros, so provide users with a repl

[PATCH v2 10/45] common/sfc_efx/base: add Medford4 support to MCDI module

From: Denis Pryazhennikov Use common EF10 method table for that. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton --- drivers/common/sfc_efx/base/efx_mcdi.c | 6 ++ 1 file changed, 6 insertions(+) diff --git a/drivers/common/sfc_efx/base/efx_mcdi.c b/drivers/common/sfc_efx/b

[PATCH v2 34/45] common/sfc_efx/base: support MAC statistics on Medford4 NICs

Supply Medford4-specific methods to clear, upload and update MAC statistics, as well as the method to toggle periodic DMA updates. All of these leverage the same netport MCDI command. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/s

[PATCH v2 43/45] common/sfc_efx/base: support controls for netport lane count

On netport MCDI capable adaptors, link modes exported by libefx can be backed by different technologies with different lane counts. Allow the client drivers to get and set the lane count and query possible values. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van

[PATCH v2 44/45] net/sfc: add support for control of physical port lane count

Since 24.11, DPDK has supported APIs to control lane count of the physical link. Provide driver-level support for that on adaptors that are netport MCDI capabale (Medford4, for instance). Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/net/

[PATCH v2 45/45] doc: advertise support for AMD Solarflare X45xx adapters

The two NICs, X4522 and X4542, are based on EF10 architecture. Signed-off-by: Ivan Malov --- doc/guides/nics/sfc_efx.rst| 9 - doc/guides/rel_notes/release_25_07.rst | 4 2 files changed, 12 insertions(+), 1 deletion(-) diff --git a/doc/guides/nics/sfc_efx.rst b/doc/gui

[PATCH v2 42/45] net/sfc: offer support for 200G link ability on new adaptors

New adaptors (for instance, Medford4) can support 200G speed. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/net/sfc/sfc.c| 5 - drivers/net/sfc/sfc_ethdev.c | 15 +++ drivers/net/sfc/sfc_port.c | 6 ++ 3 fi

[PATCH v2 39/45] common/sfc_efx/base: add link event processing on new boards

Link change events on netport MCDI capable NICs do not carry any specifics of the new link state. Such need to be queried separately. Add processing of the events, with an indication to the client driver telling it to collect the status itself. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton

[PATCH v2 38/45] common/sfc_efx/base: ignore legacy link events on new boards

New adaptors (for instance, Medford4) come with netport MCDI and also support new link change events, however, older ones can also be generated. For consistency, ignore legacy events. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/s

[PATCH v2 40/45] net/sfc: query link status on link change events on new NICs

Link events signaled on new adaptors (Medford4 and later) do not carry any specifics, so query the link status separately. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/net/sfc/sfc.h| 2 ++ drivers/net/sfc/sfc_ethdev.c | 2 ++ d

[PATCH v2 37/45] net/sfc: make use of generic EFX MAC PDU calculation helpers

To make sure that MAC PDU values do not come with legacy bug workaround baked in when running on newer Medford4 NICs, use generic replacement APIs from EFX in place of MAC PDU macros. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/net/sfc/

[PATCH v2 35/45] common/sfc_efx/base: implement MAC PDU controls for Medford4

Leverage new netport MCDI to implement support for such APIs. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/sfc_efx/base/efx_impl.h | 4 ++ drivers/common/sfc_efx/base/efx_mac.c | 4 +- drivers/common/sfc_efx/base/efx_

[PATCH v2 31/45] common/sfc_efx/base: add MAC reconfigure method for Medford4

That leverages MAC control functionality of new netport MCDI. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/sfc_efx/base/efx_impl.h | 13 drivers/common/sfc_efx/base/efx_mac.c | 2 +- drivers/common/sfc_efx/base/ef

[PATCH v2 33/45] common/sfc_efx/base: fill in MAC statistics mask on Medford4

Let client drivers know which MAC statistics can be accessed. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/sfc_efx/base/efx_mac.c | 2 +- drivers/common/sfc_efx/base/medford4_impl.h | 9 ++ drivers/common/sfc_efx/base/

[PATCH v2 32/45] common/sfc_efx/base: fill in software LUT for MAC statistics

To support MAC statistics, it is required to fill in a table which will be used by EFX to get DMA field IDs (offsets into the DMA buffer where MC FW uploads the data) by software IDs. Fill in the lookup table by parsing HW statistic descriptors. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreto

[PATCH v2 30/45] common/sfc_efx/base: introduce Medford4 stub for MAC methods

Provide only a small subset of methods for now. Next patches will augment the file with more, based on newer netport MCDI. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/sfc_efx/base/efx_mac.c | 32 +++- drivers/common

[PATCH v2 29/45] common/sfc_efx/base: implement PHY link control for Medford4

Use new MCDI to select loopback, speed, flow control and FEC. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/sfc_efx/base/ef10_impl.h | 6 + drivers/common/sfc_efx/base/ef10_phy.c | 2 +- drivers/common/sfc_efx/base/efx

[PATCH v2 25/45] common/sfc_efx/base: fill in loopback modes on netport probe

Indicate support for some of the loopback modes known to EFX. Such have been technology-dependent, but, on netport capable NICs, they are backed by new generic modes, that are tied to the stages in the processing chain where loopback can happen. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton

[PATCH v2 27/45] common/sfc_efx/base: refactor EF10 link mode decoding helper

Future patches will use this change for netport MCDI support. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/sfc_efx/base/ef10_impl.h | 13 +++ drivers/common/sfc_efx/base/ef10_phy.c | 31 ++--- 2 files

[PATCH v2 26/45] common/sfc_efx/base: introduce Medford4 stub for PHY methods

Provide only a couple of no-op methods for now. Next patches will augment the file with more, based on newer netport MCDI. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/sfc_efx/base/efx_impl.h | 4 +++ drivers/common/sfc_efx/

[PATCH v2 24/45] common/sfc_efx/base: decode netport link state on probe path

Retrieved properties form the advertised PHY capability mask. The new code also helps to check support for autonegotiation. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/sfc_efx/base/efx_impl.h | 12 + drivers/common/sfc_efx/ba

[PATCH v2 22/45] common/sfc_efx/base: allow for const in MCDI struct accessor

The existing code is fine, however, future patches will need to use this macro on buffers that are passed by callers with const qualifier. With 'cast-qual' enabled, this macro causes compile warnings in such cases. Rework it to allow for const. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton

[PATCH v2 20/45] common/sfc_efx/base: provide defaults on netport attach path

Fill in some port information, including legacy Siena fields. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/sfc_efx/base/efx_np.c | 44 1 file changed, 44 insertions(+) diff --git a/drivers/common/sfc_

[PATCH v2 21/45] common/sfc_efx/base: obtain assigned netport handle from NIC

Get an 'assigned' netport handle for the current MCDI entity. This handle will be used when sending other netport commands. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreton Reviewed-by: Pieter Jansen Van Vuuren --- drivers/common/sfc_efx/base/efx_impl.h | 4 +++ drivers/common/sfc_efx/base

[PATCH v2 19/45] common/sfc_efx/base: provide a stub for basic netport attach

New NICs (for instance, Medford4) offer new netport MCDI for managing physical ports, which will supersede the legacy one. Scope out the new interface initialisation on NIC probe path. That will be augmented with the actual code by later patches. Signed-off-by: Ivan Malov Reviewed-by: Andy Moreto

[PATCH v2 16/45] common/sfc_efx/base: add new X4 port mode

From: Denis Pryazhennikov Add handling for the port mode in the Medford4 driver logic, enabling support for two single network ports. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton Reviewed-by: Richard Houldsworth Reviewed-by: Ivan Malov --- .mailmap

[PATCH v2 12/45] common/sfc_efx/base: add Medford4 support to Tx module

From: Denis Pryazhennikov Define and use Medford4 specific method table. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton --- drivers/common/sfc_efx/base/efx_tx.c | 33 1 file changed, 33 insertions(+) diff --git a/drivers/common/sfc_efx/base/efx_tx.c

[PATCH v2 14/45] net/sfc: add Medford4 with only full feature datapath engine

From: Andy Moreton Add PCI IDs for Medford4 functions that only have the full feature datapath engine (no support for low latency datapath engine). Signed-off-by: Andy Moreton Reviewed-by: Ivan Malov --- drivers/net/sfc/sfc_ethdev.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/driver

[PATCH v2 09/45] common/sfc_efx/base: add Medford4 support to TUNNEL module

From: Denis Pryazhennikov Use common EF10 method table for that. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton --- drivers/common/sfc_efx/base/efx_tunnel.c | 18 -- 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/drivers/common/sfc_efx/base/efx_tu

[PATCH v2 06/45] common/sfc_efx/base: add Medford4 support to INTR module

From: Denis Pryazhennikov Use common EF10 method table for that. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton --- drivers/common/sfc_efx/base/efx_intr.c | 6 ++ 1 file changed, 6 insertions(+) diff --git a/drivers/common/sfc_efx/base/efx_intr.c b/drivers/common/sfc_efx/b

[PATCH v2 08/45] common/sfc_efx/base: add Medford4 support to PHY module

From: Denis Pryazhennikov Use common EF10 method table for that. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton --- drivers/common/sfc_efx/base/efx_phy.c | 6 ++ 1 file changed, 6 insertions(+) diff --git a/drivers/common/sfc_efx/base/efx_phy.c b/drivers/common/sfc_efx/bas

[PATCH v2 05/45] common/sfc_efx/base: add Medford4 support to FILTER module

From: Denis Pryazhennikov Use common EF10 method table for that. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton --- drivers/common/sfc_efx/base/efx_filter.c | 6 ++ 1 file changed, 6 insertions(+) diff --git a/drivers/common/sfc_efx/base/efx_filter.c b/drivers/common/sfc_e

[PATCH v2 04/45] common/sfc_efx/base: add Medford4 support to EV module

From: Denis Pryazhennikov Use common EF10 method table for that. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton --- drivers/common/sfc_efx/base/efx_ev.c | 6 ++ 1 file changed, 6 insertions(+) diff --git a/drivers/common/sfc_efx/base/efx_ev.c b/drivers/common/sfc_efx/base/

[PATCH v2 03/45] common/sfc_efx/base: add Medford4 support to NIC module

From: Denis Pryazhennikov Implement NIC family discovery and minimum probe support. Signed-off-by: Denis Pryazhennikov Signed-off-by: Andy Moreton --- drivers/common/sfc_efx/base/ef10_nic.c | 32 +++- drivers/common/sfc_efx/base/efx.h | 1 + drivers/common/sfc_efx/base/efx_i

[PATCH v2 00/45] Support AMD Solarflare X45xx adaptors

New X4522 (dual port SFP56) and X4542 (dual port QSFP56) adaptors are Medford4 (X4) chips that are based on EF10 architecture. An X4 NIC supports multiple network engine types. This series provides support only for the Medford2-alike, 'full-feature' (FF) network engine. This shall not be confused w

[PATCH v2 02/45] common/sfc_efx/base: add efsys option for Medford4

From: Denis Pryazhennikov Later patches will use this to implement support for Medford4. Signed-off-by: Denis Pryazhennikov Reviewed-by: Andy Moreton --- drivers/common/sfc_efx/base/efx_check.h | 25 ++--- drivers/common/sfc_efx/efsys.h | 2 ++ 2 files changed, 1

[PATCH v2 01/45] common/sfc_efx/base: add Medford4 PCI IDs to common code

From: Denis Pryazhennikov Later patches will use this to implement support for Medford4. Signed-off-by: Denis Pryazhennikov Signed-off-by: Andy Moreton --- drivers/common/sfc_efx/base/efx.h | 11 +++ 1 file changed, 11 insertions(+) diff --git a/drivers/common/sfc_efx/base/efx.h b/d

Re: [PATCH 2/5] net/qede: fix bad sanity check on Rx queue release

On Tue, 22 Apr 2025 17:51:40 +0200 edwin.brosse...@6wind.com wrote: > From: Edwin Brossette > > As per the rte_mbuf API: the driver is responsible of initializing all > the required fields. This is not done at qede alloc, meaning there can > be garbage data in mbufs memory, although this garbage

Re: [PATCH] net/mlx5: mitigate the Tx queue parameter adjustment

On Wed, 23 Apr 2025 15:28:07 +0300 Viacheslav Ovsiienko wrote: > + DRV_LOG(WARNING, > + "tso header inline requirements (%u) are not satisfied > (%u) on port %u," > + " the Tx queue capacity (%d) may not be guaranteed", > +

[PATCH] net/mlx5: mitigate the Tx queue parameter adjustment

he DPDK API rte_eth_tx_queue_setup() has a parameter nb_tx_desc specifying the desired queue capacity, measured in packets. The ConnectX NIC series has a hardware-imposed queue size limit of 32K WQEs (packet hardware descriptors). Typically, one packet requires one WQE to be sent. There is a spec

DPDK 24.11.2 released

Hi all, Here is a new stable release: https://fast.dpdk.org/rel/dpdk-24.11.2.tar.xz The git tree is at: https://dpdk.org/browse/dpdk-stable/?h=24.11 This LTS release contains ~200 fixes from main branch up to DPDK 25.03. Thanks to the authors who helped with backports and to the

[PATCH] crypto/qat: disable ZUC 256 on selected devices

Disable ZUC 256 cipher on selected devices. Signed-off-by: Radu Nicolau --- drivers/common/qat/qat_common.h | 2 ++ drivers/common/qat/qat_device.c | 9 + drivers/crypto/qat/dev/qat_crypto_pmd_gen3.c | 5 - drivers/crypto/qat/dev/qat_crypto_pmd_gen5.c | 9 ++

Re: 24.11.2 patches review and test

On 21/04/2025 07:18, Xu, HailinX wrote: >> -Original Message- >> From: Kevin Traynor >> Sent: Friday, March 28, 2025 6:15 PM >> To: sta...@dpdk.org >> Cc: dev@dpdk.org; Abhishek Marathe ; >> Ali Alnubani ; David Christensen >> ; Hemant Agrawal ; >> Stokes, Ian ; Jerin Jacob ; >> Mcnamara,

RE: [EXTERNAL] Re: [PATCH v1 04/12] node: add process callback for IP4 FIB

Hi Vladimir, >Hi Ankur, > >пт, 18 апр. 2025 г. в 15:45, Ankur Dwivedi >: > > > > Hi Vladimir, > >> diff --git a/lib/node/ip4_lookup_fib.c b/lib/node/ip4_lookup_fib.c > >> index e87864e672..c535b191f8 100644 > >> --- a/lib/node/ip4_lookup_fib.