Re: [PATCH v3 00/16] stop using variadic argument pack extension

On Wed, Feb 28, 2024 at 6:59 PM Tyler Retzlaff wrote: > > I find this new helper less tricky to use and easier to read than the > > RTE_FMT_* stuff that gets copy/pasted everywhere. > > The changes are quite mechanical, so even though we are past -rc1, +1 > > for me on the series. > > > > Can we f

RE: [PATCH 00/11] net/mlx5: flow insertion performance improvements

Hi Dariusz, > -Original Message- > From: Dariusz Sosnowski > Sent: Wednesday, February 28, 2024 7:01 PM > > Goal of this patchset is to improve the throughput of flow insertion > and deletion in mlx5 PMD when HW Steering flow engine is used. > > - Patch 1 - Use preallocated per-queue, p

RE: release candidate 24.03-rc1

> -Original Message- > From: Thomas Monjalon > Sent: Thursday, February 22, 2024 3:36 PM > To: annou...@dpdk.org > Subject: release candidate 24.03-rc1 > > A new DPDK release candidate is ready for testing: > https://git.dpdk.org/dpdk/tag/?id=v24.03-rc1 > > There are 521 new patche

Re: [PATCH] net/hns3: fix Rx packet truncation when KEEP CRC enabled

On 2/29/2024 3:58 AM, huangdengdui wrote: > > > On 2024/2/28 21:07, Ferruh Yigit wrote: >> On 2/28/2024 2:27 AM, huangdengdui wrote: >>> >>> >>> On 2024/2/27 0:43, Ferruh Yigit wrote: On 2/26/2024 3:16 AM, Jie Hai wrote: > On 2024/2/23 21:53, Ferruh Yigit wrote: >> On 2/20/2024 3:58

[PATCH v3 1/3] common/qat: isolate parser arguments configuration

This commit isolates qat device arguments from the common code. Now arguments are defined per service, and only appear in the application if the service is compiled-in. Signed-off-by: Arkadiusz Kusztal --- drivers/common/qat/qat_common.c | 11 +++ drivers/common/qat/qat_common.h | 3 +

[PATCH v3 2/3] common/qat: decouple pmds from the common code

Service specific functions were moved to services files. Weak symbols for device create/destroy were removed, named private devs were replaced by an opaque array. Signed-off-by: Arkadiusz Kusztal --- drivers/common/qat/qat_device.c | 112 ++-- drivers/common/q

[PATCH v3 3/3] common/qat: fix incorrectly placed legacy flag

This commit fixes a legacy flag, which was placed in a file that may not be included in a building process. Fixes: cffb726b7797 ("crypto/qat: enable insecure algorithms") Signed-off-by: Arkadiusz Kusztal --- drivers/common/qat/qat_device.c | 1 + drivers/crypto/qat/qat_sym.c| 1 - 2 files c

[PATCH] net/mlx5: fix the counters map in bonding mode

In the HW-LAG mode, there is only one mlx5 IB device with 2 ETH interfaces. In theory, the settings on both ports should be the same. But in the real life, some inconsistent settings may be done by the user and the PMD is not aware of this. In the previous implementation, the xstats map was genera

RE: [PATCH v4 0/4] add QAT GEN LCE device

> -Original Message- > From: Power, Ciara > Sent: Tuesday, February 27, 2024 10:55 AM > To: Nayak, Nishikanta ; dev@dpdk.org > Cc: Ji, Kai ; Kusztal, ArkadiuszX > ; S Joshi, Rakesh > Subject: RE: [PATCH v4 0/4] add QAT GEN LCE device > > > > > -Original Message- > > From: Na

RE: [PATCH v2 0/4] add new QAT gen3 and gen5

Series-acked-by: Arkadiusz Kusztal arkadiuszx.kusz...@intel.com Series-acked-by: Kai Ji mailto:kai...@intel.com>> From: Power, Ciara mailto:ciara.po...@intel.com>> Sent: 23 February 2024 15:12 To: dev@dpdk.org

Re: [PATCH v4 1/7] ethdev: support report register names and filter

26/02/2024 04:07, Jie Hai: > This patch adds "filter" and "names" fields to "rte_dev_reg_info" > structure. Names of registers in data fields can be reported and > the registers can be filtered by their names. > > The new API rte_eth_dev_get_reg_info_ext() is added to support > reporting names and

[PATCH] net/mlx5: fix action template expansion: support indirect actions list

MLX5 PMD actions template compilation may implicitly add MODIFY_HEADER to actions list provided by application. MLX5 actions in a template list must be arranged according to the HW supported order. The PMD must place new MODIFY_HEADER in the correct location relative to existing actions. The patch

[PATCH v4] common/qat: add virtual qat device (vQAT)

This commit adds virtual QAT device to the Intel QuickAssist Technology PMD. There are three kinds of virtual QAT device defined which offer different QAT service to the customers: symmetric crypto, asymmetric crypto and compression. Signed-off-by: Arkadiusz Kusztal --- v2: - added symmetric cryp

RE: [PATCH v4] common/qat: add virtual qat device (vQAT)

> -Original Message- > From: Kusztal, ArkadiuszX > Sent: Thursday, February 29, 2024 10:22 AM > To: dev@dpdk.org > Cc: gak...@marvell.com; Power, Ciara ; Kusztal, > ArkadiuszX > Subject: [PATCH v4] common/qat: add virtual qat device (vQAT) > > This commit adds virtual QAT device to th

RE: [PATCH v4 0/5] NAT64 support in mlx5 PMD

Hi, > -Original Message- > From: Bing Zhao > Sent: Wednesday, February 28, 2024 5:09 PM > To: Ori Kam ; aman.deep.si...@intel.com; Dariusz > Sosnowski ; Slava Ovsiienko > ; Suanming Mou ; > Matan Azrad ; NBU-Contact-Thomas Monjalon > (EXTERNAL) ; ferruh.yi...@amd.com; > dev@dpdk.org; Rasl

[PATCH] net/mlx5: fix sync meter processing in HWS setup

Synchronous calls for meter ASO try to pull pending completions from CQ, submit WR and return to caller. That avoids delays between WR post and HW response. If the template API was activated, PMD will use control queue for sync operations. PMD has different formats for the `user_data` context in

[PATCH] net/mlx5: fix non-masked indirect list meter translation in flow rule

Template table reuses DR5 action handle for non-masked indirect actions. Flow rule must explicitly translate non-masked indirect action and update DR5 handle with the rule indirect object. Current PMD assumed DR5 handle of non-masked indirect action was always NULL before the action translation.

Re: [PATCH] net/ixgbevf: fix RSS init for x550 nics

On Tue, Feb 27, 2024 at 05:26:06PM +, Medvedkin, Vladimir wrote: > > On 15/02/2024 13:31, edwin.brosse...@6wind.com wrote: > > From: Edwin Brossette > > > > Different Intel nics with the igxbe pmd do not handle RSS in the same > > way when working with virtualization. While some nics like In

[PATCH v2 00/11] net/mlx5: flow insertion performance improvements

Goal of this patchset is to improve the throughput of flow insertion and deletion in mlx5 PMD when HW Steering flow engine is used. - Patch 1 - Use preallocated per-queue, per-actions template buffer for storing translated flow actions, instead of allocating and filling it on demand, on each f

[PATCH v2 01/11] net/mlx5: allocate local DR rule action buffers

Goal of this is to remove the unnecessary copying of precalculated mlx5dr_rule_action structures used to create HWS flow rules. Before this patch, during template table creation an array of these structures was calculated for each actions template used. Each of these structures contained either fu

[PATCH v2 02/11] net/mlx5/hws: add check for matcher rule update support

From: Erez Shitrit The user want to know before trying to update a rule if that matcher that keeps the original rule supports updating. Signed-off-by: Erez Shitrit Acked-by: Ori Kam --- drivers/net/mlx5/hws/mlx5dr.h | 8 drivers/net/mlx5/hws/mlx5dr_matcher.c | 12 +++

[PATCH v2 04/11] net/mlx5: skip the unneeded resource index allocation

From: Bing Zhao The resource index was introduced to decouple the flow rule and its resources used by hardware steering. This is needed only when a rule update is supported. In some cases, the update is not supported on a table(matcher). E.g.: * Table is resizable * FW gets involved * Root

[PATCH v2 03/11] net/mlx5/hws: add check if matcher contains complex rules

From: Erez Shitrit The function returns true if that matcher can contain complicated rule, which means rule that needs more than one writing to the HW in order to have it. Signed-off-by: Erez Shitrit Acked-by: Ori Kam --- drivers/net/mlx5/hws/mlx5dr.h | 8 drivers/net/mlx5/h

[PATCH v2 05/11] net/mlx5: remove action params from job

mlx5_hw_q_job struct held references to buffers which contained: - modify header commands array, - encap/decap data buffer, - IPv6 routing data buffer. These buffers were passed as parameters to HWS layer during rule creation. They were needed only during the call to HWS layer when flow operation

[PATCH v2 07/11] net/mlx5: remove updated flow from job

mlx5_hw_q_job struct held a reference to a temporary flow rule struct, used during flow rule update operation. It serves as a container for flow actions data calculated during actions construction. After flow rule update operation succeeds, data from temporary flow rule is copied over to original f

[PATCH v2 06/11] net/mlx5: remove flow pattern from job

mlx5_hw_q_job struct held a reference to temporary flow rule pattern and contained temporary REPRESENTED_PORT and TAG items structs. They are used whenever it is required to prepend a flow rule pattern, provided by the application with one of such items. If prepending is required, then flow rule pa

[PATCH v2 09/11] net/mlx5: move rarely used flow fields outside

Some of the flow fields are either not always required or are used very rarely, e.g.: - AGE action reference, - direct METER/METER_MARK action reference, - matcher selector for resizable tables. This patch moves these fields to rte_flow_hw_aux struct in order to reduce the overall size of the flo

[PATCH v2 08/11] net/mlx5: use flow as operation container

While processing async flow operations in mlx5 PMD, mlx5_hw_q_job struct is used to hold the following data related to the ongoing operation. - operation type, - user data, - flow reference. Job itself is then passed to mlx5dr layer as its "user data". Other types of data required during flow ope

[PATCH v2 10/11] net/mlx5: reuse flow fields

Each time a flow is allocated in mlx5 PMD the whole buffer, both rte_flow_hw and mlx5dr_rule parts, are zeroed. This introduces some wasted work because: - mlx5dr layer does not assume that mlx5dr_rule must be initialized, - flow action translation in mlx5 PMD does not need most of the fields of

[PATCH v2 11/11] net/mlx5: remove unneeded device status checking

From: Bing Zhao The flow rule can be inserted even before the device started. The only exception is for a queue or RSS action. For the other interfaces of template API, the start status is not checked. The checking would cause some cache miss or eviction since the flag locates on some other cach

Re: [PATCH v2] net/iavf: remove error logs for vlan offloading

On Tue, Feb 06, 2024 at 11:34:20AM +0100, David Marchand wrote: > This was reported by RH QE. > > When a vlan is enforced on a VF via an administrative configuration on > the PF side, the net/iavf driver logs two error messages. > Those error messages have no consequence on the rest of the port >

[PATCH 0/7] vhost: FD manager improvements

This series aims at improving the Vhost FD manager. First patch is a fix necessary to have VDUSE devices destroy to work. I expect it to be taken into v24.03 release. The rest of the series are various improvements to the FD manager that can wait v24.07 release. Maxime Coquelin (7): vhost: fix

[PATCH 1/7] vhost: fix VDUSE device destruction failure

VDUSE_DESTROY_DEVICE ioctl can fail because the device's chardev is not released despite close syscall having been called. It happens because the events handler thread is still polling the file descriptor. fdset_pipe_notify() is not enough because it does not ensure the notification has been handl

[PATCH 2/7] vhost: rename polling mutex

This trivial patch fixes a typo in fd's manager polling mutex name. Signed-off-by: Maxime Coquelin --- lib/vhost/fd_man.c | 8 lib/vhost/fd_man.h | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/lib/vhost/fd_man.c b/lib/vhost/fd_man.c index 42ce059039..5dde40e51a 1

[PATCH 4/7] vhost: hide synchronization within FD manager

This patch forces synchronization for all FDs additions or deletions in the FD set. With that, it is no more necessary for the user to know about the FD set pipe, so hide its initialization in the FD manager. Signed-off-by: Maxime Coquelin --- lib/vhost/fd_man.c | 174 +--

[PATCH 3/7] vhost: make use of FD manager init function

Instead of statically initialize the fdset, this patch converts VDUSE and Vhost-user to use fdset_init() function, which now also initialize the mutexes. This is preliminary rework to hide FDs manager pipe from its users. Signed-off-by: Maxime Coquelin --- lib/vhost/fd_man.c | 9 +++-- lib

[PATCH 5/7] vhost: improve fdset initialization

This patch heavily reworks fdset initialization: - fdsets are now dynamically allocated by the FD manager - the event dispatcher is now created by the FD manager - struct fdset is now opaque to VDUSE and Vhost Signed-off-by: Maxime Coquelin --- lib/vhost/fd_man.c | 177 +++-- lib

[PATCH 6/7] vhost: convert fdset sync to eventfd

This patch converts fdset's sync mechanism from pipe to eventfd, as we only use it to send notification events. Signed-off-by: Maxime Coquelin --- lib/vhost/fd_man.c | 65 +++--- 1 file changed, 26 insertions(+), 39 deletions(-) diff --git a/lib/vhost/fd_

[PATCH 7/7] vhost: improve FD manager logging

Convert the logging macro to pass the fdset name as argument. Signed-off-by: Maxime Coquelin --- lib/vhost/fd_man.c | 25 +++-- 1 file changed, 11 insertions(+), 14 deletions(-) diff --git a/lib/vhost/fd_man.c b/lib/vhost/fd_man.c index 6a5bd74656..18426095b4 100644 --- a/li

[PATCH v1] ethdev: add Linux ethtool link mode conversion

Speed capabilities of a NIC may be discovered through its Linux kernel driver. It is especially useful for bifurcated drivers, so they don't have to duplicate the same logic in the DPDK driver. Parsing ethtool speed capabilities is made easy thanks to the functions added in ethdev for internal usa

[v10 0/3] net/af_xdp: fix multi interface support for K8s

The original `use_cni` implementation was limited to supporting only a single netdev in a DPDK pod. This patchset aims to fix this limitation transparently to the end user. It will also enable compatibility with the latest AF_XDP Device Plugin. Signed-off-by: Maryam Tahhan --- v10: * Add UDS acr

[v10 1/3] docs: AF_XDP Device Plugin

Fixup the references to the AF_XDP Device Plugin in the documentation (was referred to as CNI previously) and document the single netdev limitation for deploying an AF_XDP based DPDK pod. Also renames af_xdp_cni.rst to af_xdp_dp.rst Fixes: 7fc6ae50369d ("net/af_xdp: support CNI Integration") Cc: s

[v10 3/3] net/af_xdp: support AF_XDP DP pinned maps

Enable the AF_XDP PMD to retrieve the xskmap from a pinned eBPF map. This map is expected to be pinned by an external entity like the AF_XDP Device Plugin. This enabled unprivileged pods to create and use AF_XDP sockets. Signed-off-by: Maryam Tahhan --- doc/guides/howto/af_xdp_dp.rst | 3

[v10 2/3] net/af_xdp: fix multi interface support for K8s

The original 'use_cni' implementation, was added to enable support for the AF_XDP PMD in a K8s env without any escalated privileges. However 'use_cni' used a hardcoded socket rather than a configurable one. If a DPDK pod is requesting multiple net devices and these devices are from different pools,

Re: [v10 2/3] net/af_xdp: fix multi interface support for K8s

On 29/02/2024 13:01, Maryam Tahhan wrote: The original 'use_cni' implementation, was added to enable support for the AF_XDP PMD in a K8s env without any escalated privileges. However 'use_cni' used a hardcoded socket rather than a configurable one. If a DPDK pod is requesting multiple net devices

RE: [EXT] Re: [PATCH v10 4/4] app/dma-perf: add SG copy support

Hi Fengcheng, > -Original Message- > From: fengchengwen > Sent: Wednesday, February 28, 2024 3:02 PM > To: Amit Prakash Shukla ; Cheng Jiang > > Cc: dev@dpdk.org; Jerin Jacob ; Anoob Joseph > ; Kevin Laatz ; Bruce > Richardson ; Pavan Nikhilesh Bhagavatula > ; Gowrishankar Muthukrishnan

[v11 0/3] net/af_xdp: fix multi interface support for K8s

The original `use_cni` implementation was limited to supporting only a single netdev in a DPDK pod. This patchset aims to fix this limitation transparently to the end user. It will also enable compatibility with the latest AF_XDP Device Plugin. Signed-off-by: Maryam Tahhan --- v11: * Fixed up ty

[v11 1/3] docs: AF_XDP Device Plugin

Fixup the references to the AF_XDP Device Plugin in the documentation (was referred to as CNI previously) and document the single netdev limitation for deploying an AF_XDP based DPDK pod. Also renames af_xdp_cni.rst to af_xdp_dp.rst Fixes: 7fc6ae50369d ("net/af_xdp: support CNI Integration") Cc: s

[v11 2/3] net/af_xdp: fix multi interface support for K8s

The original 'use_cni' implementation, was added to enable support for the AF_XDP PMD in a K8s env without any escalated privileges. However 'use_cni' used a hardcoded socket rather than a configurable one. If a DPDK pod is requesting multiple net devices and these devices are from different pools,

[v11 3/3] net/af_xdp: support AF_XDP DP pinned maps

Enable the AF_XDP PMD to retrieve the xskmap from a pinned eBPF map. This map is expected to be pinned by an external entity like the AF_XDP Device Plugin. This enabled unprivileged pods to create and use AF_XDP sockets. Signed-off-by: Maryam Tahhan --- doc/guides/howto/af_xdp_dp.rst | 3

[PATCH v2 0/3] net/virtio: support IOVA as PA mode for vDPA backend

This patch series makes Virtio-user works in IOVA as PA mode for vDPA backend. First patch fixes the issue when having buffer IOVA address in control queue descriptors. Second and third patches helps to share descriptor IOVA address, to the vhost backend. And also disables the use_va flag for VDPA

[PATCH v2 1/3] net/virtio_user: avoid cq descriptor buffer address accessing

This patch makes changes to avoid descriptor buffer address accessing while processing shadow control queue. So that Virtio-user can work with having IOVA as descriptor buffer address. Signed-off-by: Srujana Challa --- .../net/virtio/virtio_user/virtio_user_dev.c | 68 +-- 1 fil

[PATCH v2 2/3] net/virtio: store desc IOVA address in vring data structure

Stores desc IOVA in the queue's vring data structure, as preliminary work to provide a way for Virtio-user to share desc IOVA to the vhost backend. Signed-off-by: Srujana Challa --- drivers/net/virtio/virtio_ring.h | 12 drivers/net/virtio/virtqueue.c | 4 ++-- 2 files changed, 1

[PATCH v2 3/3] net/virtio_user: support sharing vq descriptor IOVA to the backend

Adds support to share descriptor IOVA to the vhost backend. This makes virtio_user driver works in IOVA as PA mode when use_va flag is disabled. This patch also disables use_va flag for VDPA backend. Signed-off-by: Srujana Challa --- .../net/virtio/virtio_user/virtio_user_dev.c | 26 +++

Re: [PATCH 1/7] vhost: fix VDUSE device destruction failure

Hey Maxime, On Thu, Feb 29, 2024 at 1:25 PM Maxime Coquelin wrote: > > VDUSE_DESTROY_DEVICE ioctl can fail because the device's > chardev is not released despite close syscall having been > called. It happens because the events handler thread is > still polling the file descriptor. > > fdset_pipe

RE: [PATCH] net/mlx5: fix counter cache starvation

Hi, > -Original Message- > From: Dariusz Sosnowski > Sent: Wednesday, February 28, 2024 9:06 PM > To: Slava Ovsiienko ; Ori Kam ; > Suanming Mou ; Matan Azrad > ; Jack Min > Cc: dev@dpdk.org; sta...@dpdk.org; Bing Zhao > Subject: [PATCH] net/mlx5: fix counter cache starvation > > mlx5

RE: [PATCH] net/mlx5: fix the counters map in bonding mode

Hi, > -Original Message- > From: Bing Zhao > Sent: Thursday, February 29, 2024 11:35 AM > To: Slava Ovsiienko ; dev@dpdk.org; Raslan > Darawsheh > Cc: Ori Kam ; Dariusz Sosnowski > ; Suanming Mou ; > Matan Azrad ; Xueming Li ; > sta...@dpdk.org > Subject: [PATCH] net/mlx5: fix the counte

RE: [PATCH] net/mlx5: fix action template expansion: support indirect actions list

Hi, > -Original Message- > From: Gregory Etelson > Sent: Thursday, February 29, 2024 12:20 PM > To: dev@dpdk.org > Cc: Gregory Etelson ; Maayan Kashani > ; Raslan Darawsheh ; > sta...@dpdk.org; Suanming Mou ; Dariusz > Sosnowski ; Slava Ovsiienko > ; Ori Kam ; Matan Azrad > > Subject: [P

Re: [PATCH 2/7] vhost: rename polling mutex

On Thu, Feb 29, 2024 at 1:25 PM Maxime Coquelin wrote: > > This trivial patch fixes a typo in fd's manager polling > mutex name. > > Signed-off-by: Maxime Coquelin Reviewed-by: David Marchand -- David Marchand

Re: [PATCH 3/7] vhost: make use of FD manager init function

On Thu, Feb 29, 2024 at 1:25 PM Maxime Coquelin wrote: > > Instead of statically initialize the fdset, this patch > converts VDUSE and Vhost-user to use fdset_init() function, > which now also initialize the mutexes. > > This is preliminary rework to hide FDs manager pipe from > its users. > > Sig

[v11 0/4] PCI Dev and SG copy support

Improve dma-perf application to support PCI dev and SG copy, along with additional supports as below: - validate copied memory - skip tests if not opted. v11: - Review suggestions. Gowrishankar Muthukrishnan (4): app/dma-perf: add skip support app/dma-perf: add PCI device support app/dma-

[v11 2/4] app/dma-perf: add PCI device support

Add support to test performance for "device to memory" and "memory to device" data transfer. Signed-off-by: Amit Prakash Shukla Acked-by: Anoob Joseph Acked-by: Chengwen Feng --- v11: - config file formatting. app/test-dma-perf/benchmark.c | 119 ++ app/test-d

[v11 1/4] app/dma-perf: add skip support

Add support to skip running a dma-perf test-case. Signed-off-by: Amit Prakash Shukla Acked-by: Anoob Joseph Acked-by: Chengwen Feng --- v11: - config file formatting app/test-dma-perf/config.ini | 2 ++ app/test-dma-perf/main.c | 48 ++-- app/test-dma-per

[v11 3/4] app/dma-perf: validate copied memory

Validate copied memory to ensure DMA copy did not fail. Signed-off-by: Gowrishankar Muthukrishnan Acked-by: Anoob Joseph Acked-by: Chengwen Feng --- app/test-dma-perf/benchmark.c | 23 ++- app/test-dma-perf/main.c | 16 +++- app/test-dma-perf/main.h |

[v11 4/4] app/dma-perf: add SG copy support

Add SG copy support. Signed-off-by: Gowrishankar Muthukrishnan Acked-by: Anoob Joseph Acked-by: Chengwen Feng --- v11: - using struct for SGE config. app/test-dma-perf/benchmark.c | 283 ++ app/test-dma-perf/config.ini | 25 ++- app/test-dma-perf/main.c

[PATCH v2] net/mlx5: add HWS support for matching ingress metadata

Add support for matching metadata in HWS ingress rules. It using REG_B matching which is supported for each device supports HWS. Signed-off-by: Michael Baum --- v2: - Rebase. - Fix compilation issue. - Update documentation. doc/guides/nics/mlx5.rst | 3 +- drivers/net/mlx5/hw

RE: [EXT] Re: [PATCH v2] app/dma-perf: support bi-directional transfer

Hi Chengwen, I liked your suggestion and tried making changes, but encountered parsing issue for CFG files with line greater than CFG_VALUE_LEN=256(current value set). There is a discussion on the similar lines in another patch set: https://patchwork.dpdk.org/project/dpdk/patch/20231206112952.1

RE: [PATCH v2] net/mlx5: add HWS support for matching ingress metadata

> -Original Message- > From: Michael Baum > Sent: Thursday, February 29, 2024 14:49 > To: dev@dpdk.org > Cc: Matan Azrad ; Dariusz Sosnowski > ; Raslan Darawsheh ; Slava > Ovsiienko ; Ori Kam ; > Suanming Mou > Subject: [PATCH v2] net/mlx5: add HWS support for matching ingress > metadata

Re: [PATCH] net/ixgbe: increase vf reset timeout

On Tue, Feb 27, 2024 at 12:35:28PM +, Medvedkin, Vladimir wrote: > On 30/01/2024 10:00, Kevin Traynor wrote: > > When vf issues a reset to pf there is a 50 msec wait plus an additional > > max of 1 msec for the pf to indicate the reset is complete before > > timeout. > > > > In some cases, it

[PATCH v3] app/dma-perf: support bi-directional transfer

Adds bi-directional DMA transfer support to test performance. One DMA device on one core will do mem2dev transfer and another DMA device on another core will do dev2mem transfer. Depends-on: series-31298 ("PCI Dev and SG copy support") Signed-off-by: Amit Prakash Shukla --- v3: - Rebased with de

Re: [PATCH v5] app/testpmd: support updating flow rule actions

On 2/29/2024 2:15 AM, Oleksandr Kolomeiets wrote: > "flow update" updates a flow rule specified by a rule ID with a > new action list by making a call to "rte_flow_actions_update()": > > flow update {port_id} {rule_id} > actions {action} [/ {action} [...]] / end [user_id] > > Creating

Re: [PATCH v6 20/23] mbuf: remove and stop using rte marker fields

David Marchand writes: > On Wed, Feb 28, 2024 at 3:04 PM Dodji Seketeli wrote: >> > Btw, I see no way to suppress this (except a global [suppress_type] >> > name = rte_mbuf)... >> >> Right. >> >> To avoid having subsequent changes to that type from being "overly" >> suppressed, maybe do somethin

Re: [PATCH] net/iavf: fix access to null value

On Fri, Feb 09, 2024 at 03:27:56PM +0100, Burakov, Anatoly wrote: > On 1/24/2024 3:05 AM, Mingjin Ye wrote: > > The "vsi" may be null, so it needs to be used after checking. > > > > Fixes: ab28aad9c24f ("net/iavf: fix Rx Tx burst in multi-process") Fixes: 9f6186cf0d80 ("net/iavf: fix Rx/Tx burst

[PATCH v2] ethdev: add Linux ethtool link mode conversion

Speed capabilities of a NIC may be discovered through its Linux kernel driver. It is especially useful for bifurcated drivers, so they don't have to duplicate the same logic in the DPDK driver. Parsing ethtool speed capabilities is made easy thanks to the functions added in ethdev for internal usa

RE: [EXT] [PATCH v6 4/4] test/cryptodev: add tests for GCM with AAD

> Adding one new unit test code for validating the features > added as part of GCM with 64 byte AAD. > The new test case adds one new test for GCM algo for both > encrypt and decrypt operations. > > Signed-off-by: Nishikant Nayak > Acked-by: Ciara Power > --- What is the need for this new test v

[PATCH v6 1/4] eal: add pointer compression functions

Add a new utility header for compressing pointers. The provided functions can store pointers in 32-bit offsets. The compression takes advantage of the fact that pointers are usually located in a limited memory region (like a mempool). We can compress them by converting them to offsets from a base

[PATCH v6 0/4] add pointer compression API

This patchset is proposing adding a new EAL header with utility functions that allow compression of arrays of pointers. When passing caches full of pointers between threads, memory containing the pointers is copied multiple times which is especially costly between cores. A compression method will

[PATCH v6 2/4] test: add pointer compress tests to ring perf test

Add a test that runs a zero copy burst enqueue and dequeue on a ring of raw pointers and compressed pointers at different burst sizes to showcase performance benefits of newly added pointer compression APIs. Refactored threading code to pass more parameters to threads to reuse existing code. Added

[PATCH v6 3/4] docs: add pointer compression to the EAL guide

Documentation added in the EAL guide for the new utility functions for pointer compression showing example code and potential usecases. Signed-off-by: Paul Szczepanek Reviewed-by: Honnappa Nagarahalli --- .../prog_guide/env_abstraction_layer.rst | 142 ++ 1 file changed, 14

[PATCH v6 4/4] test: add unit test for ptr compression

Test compresses and decompresses pointers with various combinations of memory regions and alignments and verifies the pointers are recovered correctly. Signed-off-by: Paul Szczepanek --- app/test/meson.build | 1 + app/test/test_eal_ptr_compress.c | 108

RE: [EXT] [PATCH v6 3/4] crypto/qat: update headers for GEN LCE support

> This patch handles the changes required for updating the common > header fields specific to GEN LCE, Also added/updated of the response > processing APIs based on GEN LCE requirement. > > Signed-off-by: Nishikant Nayak > Acked-by: Ciara Power > --- > v2: > - Renamed device from GEN 5 to GE

[PATCH 1/2] net/mlx5: remove code duplications

Remove code duplications in DV L3 items validation translation. Fixes: 3193c2494eea ("net/mlx5: fix L4 protocol validation") Cc: sta...@dpdk.org Signed-off-by: Gregory Etelson Acked-by: Ori Kam --- drivers/net/mlx5/mlx5_flow_dv.c | 151 +--- 1 file changed, 43 inse

[PATCH 2/2] net/mlx5: fix IP-in-IP tunnels recognition

The patch fixes IP-in-IP tunnel recognition for the following patterns / [ipv4|ipv6] proto is [ipv4|ipv6] / end / [ipv4|ipv6] / [ipv4|ipv6] / Fixes: 3d69434113d1 ("net/mlx5: add Direct Verbs validation function") Signed-off-by: Gregory Etelson Acked-by: Ori Kam --- drivers/net/mlx5/mlx5_flo

[PATCH 0/2] fix IP-in-IP tunnels recognition

fix IP-in-IP tunnels validation and recognition Gregory Etelson (2): net/mlx5: remove code duplications net/mlx5: fix IP-in-IP tunnels recognition drivers/net/mlx5/mlx5_flow_dv.c | 243 +++- 1 file changed, 117 insertions(+), 126 deletions(-) Acked-by: Ori Kam

Re: [PATCH 0/6] remove incorrect code for loading 16B descriptors

On Thu, Feb 22, 2024 at 03:57:09PM +0100, Burakov, Anatoly wrote: > On 1/23/2024 12:40 PM, Bruce Richardson wrote: > > Inside the AVX2 code paths, there was special case code for loading two > > 16-byte descriptors simultaneously, if that build-time feature was > > enabled. As well as not being ena

RE: [EXT] [PATCH v6 1/4] common/qat: add files specific to GEN LCE

> a/drivers/common/qat/qat_adf/adf_transport_access_macros_gen_lce.h > b/drivers/common/qat/qat_adf/adf_transport_access_macros_gen_lce.h > new file mode 100644 > index 00..c9df8f5dd2 > --- /dev/null > +++ b/drivers/common/qat/qat_adf/adf_transport_access_macros_gen_lce.h > @@ -0,0 +1,51 @@

Re: [EXT] [PATCH v4 01/12] eventdev: improve doxygen introduction text

On Mon, Feb 26, 2024 at 3:29 PM Bruce Richardson wrote: > > On Mon, Feb 26, 2024 at 04:51:25AM +, Pavan Nikhilesh Bhagavatula wrote: > > > Make some textual improvements to the introduction to eventdev and event > > > devices in the eventdev header file. This text appears in the doxygen > > >

RE: [EXT] [PATCH v6 1/4] common/qat: add files specific to GEN LCE

> Adding GEN5 files for handling GEN LCE specific operations. > These files are inherited from the existing files/APIs > which has some changes specific GEN5 requirements It is not a good practice to use "adding files specific to .." Instead please explain what operation/feature is added for new d

Re: [PATCH v4] dts: add Dockerfile

Tested-by: Nicholas Pratte On Thu, Feb 29, 2024 at 10:48 AM Patrick Robb wrote: > > > -- Forwarded message - > From: > Date: Tue, Jan 16, 2024 at 2:18 PM > Subject: [PATCH v4] dts: add Dockerfile > To: , , < > tho...@monjalon.net>, , , < > paul.szczepa...@arm.com>, > Cc: , Jer

RE: [PATCH v4] crypto/ipsec_mb: unified IPsec MB interface

Hi folks, The introduction of a more unified IPsec MB library for DPDK is causing the snow3g tests to fail on ARM. Artifact here: https://lab.dpdk.org/results/dashboard/patchsets/29315/ PMDs using the direct API (KASUMI, CHACHA, ZUC, SNOW3G) will use the job API, from the AESNI MB PMD code. We

RE: [EXT] [PATCH v6 1/4] common/qat: add files specific to GEN LCE

> -Original Message- > From: Akhil Goyal > Sent: Thursday, February 29, 2024 4:14 PM > To: Nayak, Nishikanta ; dev@dpdk.org > Cc: Power, Ciara ; Ji, Kai ; Kusztal, > ArkadiuszX ; S Joshi, Rakesh > ; Thomas Monjalon ; > Burakov, Anatoly > Subject: RE: [EXT] [PATCH v6 1/4] common/qat: ad

RE: [PATCH v4] crypto/ipsec_mb: unified IPsec MB interface

> Hi folks, > > The introduction of a more unified IPsec MB library for DPDK is causing the > snow3g tests to fail on ARM. Artifact here: > https://lab.dpdk.org/results/dashboard/patchsets/29315/ > PMDs using the direct API (KASUMI, CHACHA, ZUC, SNOW3G) will use the job API, > from the AESNI MB PM

Re: [PATCH v2 1/1] net/octeon_ep: use devarg to enable ISM accesses

On Mon, Feb 26, 2024 at 2:55 PM Vamsi Attunuru wrote: > > Adds a devarg option to enable/disable ISM memory accesses > for reading packet count details. This option is disabled > by default, as ISM memory accesses effect throughput of > bigger size packets. > > Signed-off-by: Vamsi Attunuru Upda

RE: [EXT] [PATCH v6 4/4] test/cryptodev: add tests for GCM with AAD

> -Original Message- > From: Akhil Goyal > Sent: Thursday, February 29, 2024 3:52 PM > To: Nayak, Nishikanta ; dev@dpdk.org > Cc: Power, Ciara ; Ji, Kai ; Kusztal, > ArkadiuszX ; S Joshi, Rakesh > ; Fan Zhang > Subject: RE: [EXT] [PATCH v6 4/4] test/cryptodev: add tests for GCM with AA

Re: [PATCH v2] ethdev: add Linux ethtool link mode conversion

On Thu, 29 Feb 2024 16:42:56 +0100 Thomas Monjalon wrote: > +/* Link modes sorted with index as defined in ethtool. > + * Values are speed in Mbps with LSB indicating duplex. > + * > + * The ethtool bits definition should not change as it is a kernel API. > + * Using raw numbers directly avoids c

[PATCH] doc: update link to DevX integration on Windows

The older link no longer works. Cc: sta...@dpdk.org Signed-off-by: Ali Alnubani --- doc/guides/platform/mlx5.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/guides/platform/mlx5.rst b/doc/guides/platform/mlx5.rst index a66cf778d1..e9a1f52aca 100644 --- a/doc/guides/p

Re: [PATCH v3 3/3] event/cnxk: support DMA event functions

On Mon, Feb 26, 2024 at 3:30 PM Amit Prakash Shukla wrote: > > Added support of dma driver callback assignment to eventdev > enqueue and dequeue. The change also defines dma adapter > capabilities function. > > Depends-on: series-30612 ("lib/dmadev: get DMA device using device ID") > > Signed-off-

RE: [PATCH v2] ethdev: add Linux ethtool link mode conversion

> From: Stephen Hemminger [mailto:step...@networkplumber.org] > Sent: Thursday, 29 February 2024 17.45 > > On Thu, 29 Feb 2024 16:42:56 +0100 > Thomas Monjalon wrote: > > > +/* Link modes sorted with index as defined in ethtool. > > + * Values are speed in Mbps with LSB indicating duplex. > > +

RE: [PATCH] test/crypto: add ext mbuf test for aes-gcm aead algo

> Subject: [PATCH] test/crypto: add ext mbuf test for aes-gcm aead algo > > Add external mbuf test for AES GCM aead algo. > > Signed-off-by: Aakash Sasidharan Applied to dpdk-next-crypto Thanks.

RE: [EXT] [PATCH 0/2] app/test-crypto-perf: fix multi-segment issue

> This commit fixes the bugs in multi-segment size calculation. > > Suanming Mou (2): > app/test-crypto-perf: fix copy segment size calculation > app/test-crypto-perf: fix dst_mbuf size calculation > > app/test-crypto-perf/cperf_test_common.c | 5 +++-- > 1 file changed, 3 insertions(+), 2 d

  1   2   3   >