Re: [dpdk-dev] [PATCH v6 1/6] drivers: add generic API to find PCI extended cap

On Mon, Sep 28, 2020 at 4:03 PM Ferruh Yigit wrote: > > On 9/28/2020 9:59 AM, David Marchand wrote: > > On Sun, Sep 27, 2020 at 2:21 PM Jerin Jacob wrote: > >> > >> On Fri, Sep 25, 2020 at 5:26 PM Manish Chopra wrote: > >>> > >>> By adding generic API, this patch removes individual > >>> functio

[dpdk-dev] [RFC 0/5] decouple dependency from superseded structures

As annouced in the deprecation.rst, the legacy filter API and "rte_eth_conf.fdir_conf" field will be removed in DPDK 20.11. This patch set removes these superseded data structures from PMD and replaces by redefined data structures in PMD side. "rte_eth_conf.fdir_conf" field will be removed but IXGB

Re: [dpdk-dev] [PATCH] net/octeontx2: set max vtag insertion size

On Fri, Aug 21, 2020 at 12:29 PM wrote: > > From: Kiran Kumar K > > When TX side VTAG insertion is enabled, As we are not setting the max > vtag insertion size an interrupt has been received. This patch will fix > the issue by configuring the max vtag insertion size to 8B. > > Signed-off-by: Kira

[dpdk-dev] [RFC 4/5] net/ixgbe: decouple dependency from superseded structures

The legacy filter API will be removed, the associated rte_eth_ctrl.h will also be removed. This patch replaces these superseded structures by the PMD internal structures. The macros RTE_ETH_FILTER_GENERIC and RTE_ETH_FILTER_GET are not replaced, they are needed to follow librte to change. The "rte_

[dpdk-dev] [RFC 3/5] net/e1000: decouple dependency from superseded structures

The legacy filter API will be removed, the associated rte_eth_ctrl.h will also be removed. This patch replaces these superseded structures by the PMD internal structures. The macros RTE_ETH_FILTER_GENERIC and RTE_ETH_FILTER_GET are not replaced, they are needed to follow librte to change. Signed-o

[dpdk-dev] [RFC 2/5] net/igc: decouple dependency from superseded structures

From: Junyu Jiang The legacy filter API will be removed, the associated rte_eth_ctrl.h will also be removed. This patch replaces these superseded structures by the PMD internal structures. Signed-off-by: Junyu Jiang --- drivers/net/igc/igc_filter.c | 2 +- drivers/net/igc/igc_filter.h | 2 ++

[dpdk-dev] [RFC 1/5] net/i40e: decouple dependency from superseded structures

From: Junyu Jiang The legacy filter API will be removed, the associated rte_eth_ctrl.h will also be removed. This patch replaces these superseded structures by the PMD internal structures. The macros RTE_ETH_FILTER_GENERIC and RTE_ETH_FILTER_GET are not replaced, they are needed to follow librte

[dpdk-dev] [RFC 5/5] app/testpmd: add support for ixgbe FDIR configuration

As annouced in deprecation.rst, the "rte_eth_conf.fdir_conf" field will be removed, but IXGBE generic flow still needs to depend on these configurations. This patch adds setting FDIR configuration for IXGBE PMD by calling private API. Signed-off-by: Chenxu Di --- app/test-pmd/parameters.c | 63 +

Re: [dpdk-dev] [PATCH V15 1/3] ethdev: introduce FEC API

On Mon, Sep 28, 2020 at 6:06 PM Min Hu (Connor) wrote: > This patch adds Forward error correction(FEC) support for ethdev. > Introduce APIs which support query and config FEC information in > hardware. > > Signed-off-by: Min Hu (Connor) > Reviewed-by: Wei Hu (Xavier) > Reviewed-by: Chengwen Fen

Re: [dpdk-dev] [PATCH v5] bus/pci: netuio interface for windows

> Subject: [PATCH v5] bus/pci: netuio interface for windows > > This patch adds implementations to probe PCI devices bound to netuio with > the help of "netuio" class device changes. > Now Windows will support both "netuio" and "net" device class and can set > kernel driver type based on the devic

Re: [dpdk-dev] [PATCH v2 8/8] net/virtio: introduce Vhost-vDPA backend

Hi Maxime, I see there's a compile issue in patchwork. In vhost_vdpa_dma_unmap, len is size_t. And you use '0ULL -1' outside. In i686, ULL is 64 bit but size_t is 32 bit. Although -1 is all F and the code logic is fine to convert 64 '1's to 32 '1's, the compiler will treat it as an error? Thank

Re: [dpdk-dev] [PATCH v2] net/ring: advertise multi segment TX and scatter RX.

On Mon, Sep 28, 2020 at 08:47:29PM +0200, Dumitru Ceara wrote: > Even though ring interfaces don't support any other TX/RX offloads they > do support sending multi segment packets and this should be advertised > in order to not break applications that use ring interfaces. > > Also advertise scatte

Re: [dpdk-dev] [PATCH V15 1/3] ethdev: introduce FEC API

Thank Ajit. Hi, all, Are there any suggestions for this set of patches? I wish this patch can be merged into 20.11,and this is urget. Please gives reply to me, thanks. 在 2020/9/29 12:18, Ajit Khaparde 写道: On Mon, Sep 28, 2020 at 6:06 PM Min Hu (Connor)

Re: [dpdk-dev] [PATCH v1] event/sw: performance improvements

On 9/28/2020 5:02 PM, Honnappa Nagarahalli wrote: Add minimum burst throughout the scheduler pipeline and a flush counter. Replace ring API calls with local single threaded implementation where possible. Signed-off-by: Radu Nicolau mailto:radu.nico...@intel.com Thanks for the patch, a few c

Re: [dpdk-dev] [v2 1/2] cryptodev: support enqueue callback functions

> > > > > > > > > > > > > > > > > > > > > > > > > > +#ifdef RTE_CRYPTODEV_CALLBACKS int > > > > > > > > > > > +rte_cryptodev_rcu_qsbr_add(uint8_t dev_id, struct > > > > > > > > > > > +rte_rcu_qsbr > > > > > > > > > > > +*qsbr) { > > > > > > > > > > > + > > > > > > > > > > > + struct rte_cryptode

Re: [dpdk-dev] [dpdk-techboard] [PATCH V5 1/2] dpdk: resolve compiling errors for per-queue stats

29/09/2020 06:49, Min Hu (Connor): > > 在 2020/9/28 21:53, Ferruh Yigit 写道: > > On 9/28/2020 10:16 AM, Thomas Monjalon wrote: > >> 28/09/2020 10:59, Ferruh Yigit: > >>> On 9/27/2020 4:16 AM, Min Hu (Connor) wrote: > From: Huisong Li > > Currently, only statistics of rx/tx queues wit

[dpdk-dev] [PATCH v3 0/4] optimize async data path

This series applies optimization and fixes to the vhost async data path. v3: - fix a typo in vhost error log (checkpatch warning) - fix travis-robot ci build warning on aarch64 v2: - minor rewordings on commit message - minor fix on poll_enenque_completion to correct a packet number calcul

[dpdk-dev] [PATCH v3 1/4] vhost: simplify async copy completion

Current async ops allows check_completed_copies() callback to return arbitrary number of async iov segments finished from backend async devices. This design creates complexity for vhost to handle breaking transfer of a single packet (i.e. transfer completes in the middle of a async descriptor) and

[dpdk-dev] [PATCH v3 2/4] vhost: dynamically allocate async memory

Allocate async internal memory buffer by rte_malloc(), replacing array declaration inside vq structure. Dynamic allocation can help to save memory footprint when async path is not registered. Signed-off-by: Patrick Fu --- lib/librte_vhost/vhost.c | 69 ++-- li

[dpdk-dev] [PATCH v3 3/4] vhost: fix async vector buffer overrun

Add check on the async vector buffer usage to prevent the buf overrun. If the unused vector buffer is not sufficient to prepare for next packet's iov creation, an async transfer will be triggered immediately to free the vector buffer. Fixes: 78639d54563a ("vhost: introduce async enqueue registrati

[dpdk-dev] [PATCH v3 4/4] vhost: fix async register/unregister deadlock

When async register/unregister function is invoked in certain vhost event callbacks (e.g. vring state change), deadlock may occur due to recursive spinlock acquire. This patch removes unnecessary spinlock from register API and use trylock() primitive in the unregister API to avoid deadlock. Fixes:

Re: [dpdk-dev] [PATCH] net/octeontx2: add support VLAN insert and strip action

On Mon, Aug 24, 2020 at 12:34 PM wrote: > > From: Kiran Kumar K > > Adding support for RTE Flow VLAN insert and strip actions for octeontx2 > device. > > Signed-off-by: Kiran Kumar K > --- > doc/guides/nics/octeontx2.rst | 68 +++-- > drivers/net/octeontx2/otx2_flow.c |

[dpdk-dev] [PATCH v9 00/11] adding support for python 3 only

This patch set converts all python scripts in the project to use python3 only and removes all deprecation notices associated with these changes. This is due to python2 being EOL in January 2020. --- v6: - rebased, removing conflict with make removal patchset. - added changes to buildtools/map_

[dpdk-dev] [PATCH v9 02/11] usertools/dpdk-devbind: support python3 only

From: Louise Kilheeney Changed script to explicitly use python3 only to avoid maintaining python 2 and removed deprecation notice. Signed-off-by: Louise Kilheeney Reviewed-by: Bruce Richardson Acked-by: Robin Jarry --- usertools/dpdk-devbind.py | 22 -- 1 file changed, 4

[dpdk-dev] [PATCH v9 03/11] usertools/dpdk-pmdinfo: support python3 only

From: Louise Kilheeney Changed script to explicitly use python3 only to avoid maintaining python 2 and removed deprecation notice. Cc: Neil Horman Signed-off-by: Louise Kilheeney Reviewed-by: Bruce Richardson Acked-by: Neil Horman Acked-by: Robin Jarry --- usertools/dpdk-pmdinfo.py | 7 +-

[dpdk-dev] [PATCH v9 01/11] usertools/dpdk-telemetry-client: support python3 only

From: Louise Kilheeney Changed script to explicitly use python3 only to avoid maintaining python 2 and removed deprecation notice. Cc: Kevin Laatz Signed-off-by: Louise Kilheeney Signed-off-by: Kevin Laatz Acked-by: Bruce Richardson Acked-by: Robin Jarry --- v8: removed cleanup not relate

[dpdk-dev] [PATCH v9 04/11] usertools/cpu_layout: support python3 only

From: Louise Kilheeney Changed script to explicitly use python3 only to avoid maintaining python 2 and removed deprecation notice. Signed-off-by: Louise Kilheeney Reviewed-by: Bruce Richardson Acked-by: Robin Jarry --- usertools/cpu_layout.py | 13 ++--- 1 file changed, 2 insertions(

[dpdk-dev] [PATCH v9 05/11] app/test-cmdline: support python3 only

Changed script to explicitly use python3 only to avoid maintaining python 2 and removed deprecation notice. Cc: Olivier Matz Signed-off-by: Louise Kilheeney Signed-off-by: Kevin Laatz Reviewed-by: Bruce Richardson Acked-by: Robin Jarry --- v5: - fixed python3 issue causing script to fail.

Re: [dpdk-dev] [PATCH v3 28/29] ethdev: reset all when releasing a port

On 9/29/20 1:14 AM, Thomas Monjalon wrote: > The function rte_eth_dev_release_port() was resetting partially > the struct rte_eth_dev. The drivers were completing it > with more pointers set to NULL in the close or remove operations. > > A full memset is done so most of those assignments become

[dpdk-dev] [PATCH v9 06/11] app/test: support python3 only

Changed script to explicitly use python3 only to avoid maintaining python 2 and removed deprecation notice. Signed-off-by: Louise Kilheeney Signed-off-by: Kevin Laatz Acked-by: Robin Jarry --- v6: - Removed changes to mk/rte.sdktest.mk since the file no longer exists --- app/test/autotest.p

[dpdk-dev] [PATCH v9 08/11] config/arm: support python3 only

From: Louise Kilheeney Changed script to explicitly use python3 only to avoid maintaining python 2. Cc: Thomas Monjalon Signed-off-by: Louise Kilheeney Acked-by: Robin Jarry --- config/arm/armv8_machine.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/config/arm/armv8_

[dpdk-dev] [PATCH v9 09/11] app/test-bbdev: support python3 only

From: Louise Kilheeney Changed script to explicitly use python3 only to avoid maintaining python 2 and removed deprecation notice. Cc: Nicolas Chautru Signed-off-by: Louise Kilheeney Acked-by: Robin Jarry --- app/test-bbdev/test-bbdev.py | 7 +-- 1 file changed, 1 insertion(+), 6 deleti

[dpdk-dev] [PATCH v9 07/11] devtools: support python3 only

From: Louise Kilheeney Changed script to explicitly use python3 only to avoid maintaining python 2 and removed deprecation notice. Cc: Neil Horman Cc: Ray Kinsella Signed-off-by: Louise Kilheeney Acked-by: Ray Kinsella Acked-by: Neil Horman Acked-by: Robin Jarry --- devtools/update_versi

[dpdk-dev] [PATCH v9 10/11] buildtools: support python3 only

Changed script to explicitly use python3 only to avoid maintaining python 2 and removed deprecation notice. Cc: Dmitry Kozlyuk Cc: Narcisa Ana Maria Vasile Cc: Dmitry Malloy Cc: Pallavi Kadam Signed-off-by: Kevin Laatz Acked-by: Robin Jarry --- v7: - Cc Maintainers --- buildtools/map_to

[dpdk-dev] [PATCH v9 11/11] doc: support python3 only

Remove references to python 2 from the documentation. The deprecation notice for python 2 support has been removed. A section was added to the rel_notes for this patchset. Cc: Ray Kinsella Cc: Neil Horman Cc: John McNamara Cc: Marko Kovacevic Signed-off-by: Kevin Laatz --- doc/guides/contr

Re: [dpdk-dev] [PATCH v3 28/29] ethdev: reset all when releasing a port

29/09/2020 12:26, Maxime Coquelin: > On 9/29/20 1:14 AM, Thomas Monjalon wrote: > > The function rte_eth_dev_release_port() was resetting partially > > the struct rte_eth_dev. The drivers were completing it > > with more pointers set to NULL in the close or remove operations. > > > > A full memset

Re: [dpdk-dev] [PATCH v3 21/29] ethdev: remove old close behaviour

On 9/29/20 2:14 AM, Thomas Monjalon wrote: > The temporary flag RTE_ETH_DEV_CLOSE_REMOVE is removed. > It was introduced in DPDK 18.11 in order to give time for PMDs to migrate. > > The old behaviour was to free only queues when closing a port. > The new behaviour is calling rte_eth_dev_release_po

Re: [dpdk-dev] [PATCH v2] kni: fix build with Linux 5.9

> Starting from Linux 5.9 'get_user_pages_remote()' API doesn't get > 'struct task_struct' parameter: > commit 64019a2e467a ("mm/gup: remove task_struct pointer for all gup code") > > The change reflected to the KNI with version check. > > Cc: sta...@dpdk.org > > Signed-off-by: Ferruh Yigit Ap

Re: [dpdk-dev] [PATCH v3 23/29] drivers/net: check process type in close operation

On 9/29/20 2:14 AM, Thomas Monjalon wrote: > The secondary processes are not allowed to release shared resources. > Only process-private resources should be freed in a secondary process. > Most of the time, there is no process-private resource, > so the close operation is just forbidden in a second

Re: [dpdk-dev] [PATCH v3 27/29] ethdev: remove forcing stopped state upon close

On 9/29/20 2:14 AM, Thomas Monjalon wrote: > When closing a port, it is supposed to be already stopped, > and marked as such with "dev_started" state zeroed. > > Resetting "dev_started" before calling the driver close operation > was hiding the case of not properly stopped port being closed. > The

Re: [dpdk-dev] [PATCH v3 28/29] ethdev: reset all when releasing a port

On 9/29/20 2:14 AM, Thomas Monjalon wrote: > The function rte_eth_dev_release_port() was resetting partially > the struct rte_eth_dev. The drivers were completing it > with more pointers set to NULL in the close or remove operations. > > A full memset is done so most of those assignments become us

Re: [dpdk-dev] [PATCH v3 01/29] ethdev: reset device and interrupt pointers on release

On 9/29/20 2:14 AM, Thomas Monjalon wrote: > The pointers .device and .intr_handle were already reset by the helper > rte_eth_dev_pci_generic_remove(). > It is now made part of rte_eth_dev_release_port(). > > It makes rte_eth_dev_pci_release() meaningless, > so it is replaced with a call to rte_et

Re: [dpdk-dev] [PATCH v3 02/29] ethdev: allow drivers to return error on close

On 9/29/20 2:14 AM, Thomas Monjalon wrote: > The device operation .dev_close was returning void. > This driver interface is changed to return an int. > > Note that the API rte_eth_dev_close() is still returning void, > although a deprecation notice is pending to change it as well. > > Signed-off-

Re: [dpdk-dev] [PATCH v3 29/29] ethdev: allow close function to return an error

On 9/29/20 2:14 AM, Thomas Monjalon wrote: > The API function rte_eth_dev_close() was returning void. > The return type is changed to int for notifying of errors. > > If an error happens during a close operation, > the status of the port is undefined, > a maximum of resources having been freed. >

[dpdk-dev] [PATCH 1/9] net/hns3: expand the number of queues for one TC up to 512

From: Huisong Li The maximum number of queues for one TC hns3 PF PMD driver supported is 64 based on hns3 network engine with revison_id equals 0x21, while it is expanded up to 512 on hns3 network engine with revision_id equals 0x30. So the following points need to be modified to maintain better

[dpdk-dev] [PATCH 2/9] net/hns3: maximize the queue number

From: "Wei Hu (Xavier)" The maximum number of queues for hns3 PF and VF driver is 64 based on hns3 network engine with revision_id equals 0x21. Based on hns3 network engine with revision_id equals 0x30, the hns3 PF PMD driver can support up to 1280 queues, and hns3 VF PMD driver can support up to

[dpdk-dev] [PATCH 4/9] net/hns3: set suitable type when initial flow error struct

From: "Wei Hu (Xavier)" The API of rte_flow_error_set is used to pass detail error information to caller, this patch sets suitable type when calling rte_flow_error_set API. Fixes: fcba820d9b9e ("net/hns3: support flow director") Fixes: c37ca66f2b27 ("net/hns3: support RSS") Cc: sta...@dpdk.org

[dpdk-dev] [PATCH 0/9] updates and fixes for hns3 PMD driver

This series are updates and fixes for hns3 PMD driver. Chengchang Tang (1): net/hns3: support start and stop Tx or Rx queue Hongbo Zheng (2): net/hns3: check return value when reading PCI config space net/hns3: remove redundant return value assignments Huisong Li (3): net/hns3: expand th

[dpdk-dev] [PATCH 3/9] net/hns3: fix error type when validating RSS flow action

From: "Wei Hu (Xavier)" Because the macro named RTE_FLOW_ERROR_TYPE_ACTION_CONF indicates a action configuration and the macro named RTE_FLOW_ERROR_TYPE_ACTION indicates a specific action, the driver needs to return RTE_FLOW_ERROR_ACTION_CONF type and notify the user when a RSS configuration is i

[dpdk-dev] [PATCH 6/9] net/hns3: set max scheduling rate based on actual board

From: Huisong Li Currently, max scheduling rates configuration of pg, pri and port are set to 10Mbps, which is the maximum bandwidth of hns3 network engine with revision_id equals 0x21. However, max scheduling rate configuration should be set to hardware based on the actual hardware board env

[dpdk-dev] [PATCH 5/9] net/hns3: offload calculating the shapping para to firmware

From: Huisong Li In order to have more flexible selection of shapping algorithm based on different versions of hns3 network engine, moves the algorithm of calculating shapping parameter to firmware to execute. If bit HNS3_TM_RATE_VLD_B of flag field of struct named hns3_pri_shapping_cmd, hns3_pg_

[dpdk-dev] [PATCH 9/9] net/hns3: remove redundant return value assignments

From: Hongbo Zheng When an error occurs in the reset process, -EIO is returned. The assignment of ret here is redundant, so deleted it. Signed-off-by: Hongbo Zheng Signed-off-by: Wei Hu (Xavier) --- drivers/net/hns3/hns3_intr.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/hn

[dpdk-dev] [PATCH 7/9] net/hns3: support start and stop Tx or Rx queue

From: Chengchang Tang The new generation hns3 network engine supports independent enabling and disabling of a single Tx/Rx queue. So, it can support the queue start and stop feature. In addition, when different numbers of Tx and Rx queues need to be enabled in some applications, hns3 pmd does not

[dpdk-dev] [PATCH 8/9] net/hns3: check return value when reading PCI config space

From: Hongbo Zheng This patch add return value check when calling rte_pci_read_config function. Fixes: cea37e513329 ("net/hns3: fix FLR reset") Cc: sta...@dpdk.org Signed-off-by: Hongbo Zheng Signed-off-by: Wei Hu (Xavier) --- drivers/net/hns3/hns3_ethdev_vf.c | 62 ++

Re: [dpdk-dev] [PATCH v3 00/60] common/sfc_efx: support Riverhead NIC family

On 9/24/2020 1:11 PM, Andrew Rybchenko wrote: Riverhead is a new NIC family, the first one with EF100 architecture. Control interface is an MCDI similar to EF10. Datapath interface is completely new. Hi Andrew, Can you please update documentation and release notes for new Riverhead devices, i

Re: [dpdk-dev] [PATCH v3 00/60] common/sfc_efx: support Riverhead NIC family

On 9/29/20 2:32 PM, Ferruh Yigit wrote: > On 9/24/2020 1:11 PM, Andrew Rybchenko wrote: >> Riverhead is a new NIC family, the first one with EF100 architecture. >> Control interface is an MCDI similar to EF10. Datapath interface is >> completely new. > > Hi Andrew, > > Can you please update docum

Re: [dpdk-dev] [PATCH v3 29/29] ethdev: allow close function to return an error

29/09/2020 13:05, Andrew Rybchenko: > On 9/29/20 2:14 AM, Thomas Monjalon wrote: > > The API function rte_eth_dev_close() was returning void. > > The return type is changed to int for notifying of errors. > > > > If an error happens during a close operation, > > the status of the port is undefined

Re: [dpdk-dev] [PATCH v3 00/60] common/sfc_efx: support Riverhead NIC family

On 9/29/2020 12:41 PM, Andrew Rybchenko wrote: On 9/29/20 2:32 PM, Ferruh Yigit wrote: On 9/24/2020 1:11 PM, Andrew Rybchenko wrote: Riverhead is a new NIC family, the first one with EF100 architecture. Control interface is an MCDI similar to EF10. Datapath interface is completely new. Hi And

Re: [dpdk-dev] [PATCH v3 00/60] common/sfc_efx: support Riverhead NIC family

On 9/24/2020 1:11 PM, Andrew Rybchenko wrote: Riverhead is a new NIC family, the first one with EF100 architecture. Control interface is an MCDI similar to EF10. Datapath interface is completely new. The patch series is build tested with and without EFSYS_OPT_RIVERHEAD enabled. EF10-based NICs

Re: [dpdk-dev] [PATCH v3 29/29] ethdev: allow close function to return an error

On 9/29/20 2:47 PM, Thomas Monjalon wrote: > 29/09/2020 13:05, Andrew Rybchenko: >> On 9/29/20 2:14 AM, Thomas Monjalon wrote: >>> The API function rte_eth_dev_close() was returning void. >>> The return type is changed to int for notifying of errors. >>> >>> If an error happens during a close opera

Re: [dpdk-dev] [PATCH v3 28/29] ethdev: reset all when releasing a port

> -Original Message- > From: Thomas Monjalon > Sent: Tuesday, September 29, 2020 18:36 > To: Maxime Coquelin > Cc: dev@dpdk.org; Yigit, Ferruh ; > arybche...@solarflare.com; Shepard Siegel > ; Ed Czeck ; John > Miller > ; Igor Russkikh ; > Pavel Belous > ; Somalapuram Amaranath ; Ajit

Re: [dpdk-dev] [PATCH v4 3/5] net/ice: fix max mtu size packets with vlan tag cannot be received by default

> -Original Message- > From: Yang, SteveX > Sent: Monday, September 28, 2020 2:56 PM > To: dev@dpdk.org > Cc: Zhao1, Wei ; Guo, Jia ; Yang, > Qiming ; Zhang, Qi Z ; Wu, > Jingjing ; Xing, Beilei ; > Ananyev, Konstantin ; Yang, SteveX > > Subject: [PATCH v4 3/5] net/ice: fix max mtu siz

[dpdk-dev] [PATCH v2 4/9] net/hns3: set suitable type when initial flow error struct

From: "Wei Hu (Xavier)" The API of rte_flow_error_set is used to pass detail error information to caller, this patch sets suitable type when calling rte_flow_error_set API. Fixes: fcba820d9b9e ("net/hns3: support flow director") Fixes: c37ca66f2b27 ("net/hns3: support RSS") Cc: sta...@dpdk.org

[dpdk-dev] [PATCH v2 0/9] updates and fixes for hns3 PMD driver

This series are updates and fixes for hns3 PMD driver. Chengchang Tang (1): net/hns3: support start and stop Tx or Rx queue Hongbo Zheng (2): net/hns3: check return value when reading PCI config space net/hns3: remove redundant return value assignments Huisong Li (3): net/hns3: expand th

[dpdk-dev] [PATCH v2 1/9] net/hns3: expand the number of queues for one TC up to 512

From: Huisong Li The maximum number of queues for one TC hns3 PF PMD driver supported is 64 based on hns3 network engine with revision_id equals 0x21, while it is expanded up to 512 on hns3 network engine with revision_id equals 0x30. So the following points need to be modified to maintain bette

[dpdk-dev] [PATCH v2 6/9] net/hns3: set max scheduling rate based on actual board

From: Huisong Li Currently, max scheduling rates configuration of pg, pri and port are set to 10Mbps, which is the maximum bandwidth of hns3 network engine with revision_id equals 0x21. However, max scheduling rate configuration should be set to hardware based on the actual hardware board env

[dpdk-dev] [PATCH v2 3/9] net/hns3: fix error type when validating RSS flow action

From: "Wei Hu (Xavier)" Because the macro named RTE_FLOW_ERROR_TYPE_ACTION_CONF indicates a action configuration and the macro named RTE_FLOW_ERROR_TYPE_ACTION indicates a specific action, the driver needs to return RTE_FLOW_ERROR_ACTION_CONF type and notify the user when a RSS configuration is i

[dpdk-dev] [PATCH v2 9/9] net/hns3: remove redundant return value assignments

From: Hongbo Zheng When an error occurs in the reset process, -EIO is returned. The assignment of ret here is redundant, so deleted it. Signed-off-by: Hongbo Zheng Signed-off-by: Wei Hu (Xavier) --- drivers/net/hns3/hns3_intr.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/hn

[dpdk-dev] [PATCH v2 8/9] net/hns3: check return value when reading PCI config space

From: Hongbo Zheng This patch add return value check when calling rte_pci_read_config function. Fixes: cea37e513329 ("net/hns3: fix FLR reset") Cc: sta...@dpdk.org Signed-off-by: Hongbo Zheng Signed-off-by: Wei Hu (Xavier) --- drivers/net/hns3/hns3_ethdev_vf.c | 62 ++

[dpdk-dev] [PATCH v2 2/9] net/hns3: maximize the queue number

From: "Wei Hu (Xavier)" The maximum number of queues for hns3 PF and VF driver is 64 based on hns3 network engine with revision_id equals 0x21. Based on hns3 network engine with revision_id equals 0x30, the hns3 PF PMD driver can support up to 1280 queues, and hns3 VF PMD driver can support up to

[dpdk-dev] [PATCH v2 5/9] net/hns3: offload calculating the shapping para to firmware

From: Huisong Li In order to have more flexible selection of shapping algorithm based on different versions of hns3 network engine, moves the algorithm of calculating shapping parameter to firmware to execute. If bit HNS3_TM_RATE_VLD_B of flag field of struct named hns3_pri_shapping_cmd, hns3_pg_

[dpdk-dev] [PATCH v2 7/9] net/hns3: support start and stop Tx or Rx queue

From: Chengchang Tang The new generation hns3 network engine supports independent enabling and disabling of a single Tx/Rx queue. So, it can support the queue start and stop feature. In addition, when different numbers of Tx and Rx queues need to be enabled in some applications, hns3 pmd does not

[dpdk-dev] [PATCH] ethdev: fix link speed helper documentation

When generating the documentation, a new warning can be seen: .../dpdk/lib/librte_ethdev/rte_ethdev.h:2441: warning: argument 'link_speed' of command @param is not found in the argument list of rte_eth_link_speed_to_str(uint32_t speed_link) .../dpdk/lib/librte_ethdev/rte_ethdev.h:2455: warning

Re: [dpdk-dev] [PATCH] ethdev: fix link speed helper documentation

On 9/29/20 3:12 PM, David Marchand wrote: > When generating the documentation, a new warning can be seen: > > .../dpdk/lib/librte_ethdev/rte_ethdev.h:2441: > warning: argument 'link_speed' of command @param is not found in the > argument list of rte_eth_link_speed_to_str(uint32_t speed_link) >

Re: [dpdk-dev] [PATCH v9 00/11] adding support for python 3 only

Hello Kevin, On Tue, Sep 29, 2020 at 12:27 PM Kevin Laatz wrote: > > This patch set converts all python scripts in the project to use > python3 only and removes all deprecation notices associated with these > changes. This is due to python2 being EOL in January 2020. > > --- > v6: > - rebased,

Re: [dpdk-dev] [PATCH v2 2/8] crypto/bcmfs: add vfio support

Hi Vikas, > > Hi Akhil, > > > > + > > > +#ifdef VFIO_PRESENT > > > > I cannot see VFIO_PRESENT flag defined in this patch. > > Hence the below code is a dead code and the patch > > Title is not justified as it says adding support for VFIO. > I believe VFIO_PRESENT flag is dependent on the platfor

Re: [dpdk-dev] [PATCH] ethdev: fix link speed helper documentation

29/09/2020 14:17, Andrew Rybchenko: > On 9/29/20 3:12 PM, David Marchand wrote: > > When generating the documentation, a new warning can be seen: > > > > .../dpdk/lib/librte_ethdev/rte_ethdev.h:2441: > > warning: argument 'link_speed' of command @param is not found in the > > argument list of

Re: [dpdk-dev] [PATCH v1] config: remap flags used for Arm platforms

Hi Jerin, Do you have any comments on this change? Thanks. /Ruifeng > -Original Message- > From: Ruifeng Wang > Sent: Friday, September 11, 2020 11:40 PM > To: Bruce Richardson ; Vladimir Medvedkin > ; tho...@monjalon.net; > hemant.agra...@nxp.com; sachin.sax...@nxp.com; Wei Zhao > ; Je

[dpdk-dev] [PATCH] net/netvsc: fix txq leak in error path

From: Yunjian Wang In hn_dev_tx_queue_setup() allocated memory for txq, we should free it when error happens, otherwise it will lead to memory leak. Fixes: cc0251813277 ("net/netvsc: split send buffers from Tx descriptors") Cc: sta...@dpdk.org Signed-off-by: Yunjian Wang --- drivers/net/netvs

Re: [dpdk-dev] [dpdk-techboard] [PATCH V5 1/2] dpdk: resolve compiling errors for per-queue stats

Hi, Thomas, I think what you suggest is resonable and helpful. But I should have time to fix it about this patch. So, this patch will be sent to community in future. By the way, the other patch is acked by you. Acked-by: Thomas Monjalon

[dpdk-dev] [PATCH] eventdev: fix adapter leak in error path

From: Yunjian Wang In rte_event_crypto_adapter_create_ext() allocated memory for adapter, we should free it when error happens, otherwise it will lead to memory leak. Fixes: 7901eac3409a ("eventdev: add crypto adapter implementation") Cc: sta...@dpdk.org Signed-off-by: Yunjian Wang --- lib/li

Re: [dpdk-dev] [PATCH v5 36/41] examples/pipeline: add new example application

On Wed, Sep 23, 2020 at 8:07 PM Cristian Dumitrescu wrote: > > Add new example application to showcase the API of the newly > introduced SWX pipeline type. > > Signed-off-by: Cristian Dumitrescu > --- > MAINTAINERS | 1 + > examples/meson.build | 1 + > examples/pi

Re: [dpdk-dev] [PATCH v5 38/41] examples/pipeline: add configuration commands

Caught while diffing with ip_pipeline code. On Wed, Sep 23, 2020 at 8:07 PM Cristian Dumitrescu wrote: [snip] > + snprintf(out, out_size, > + "\n" > + "%s: flags=<%s> mtu %u\n" > + "\tether %02X:%02X:%02X:%02X:%02X:%02X rxqueues %u txqueues > %u\n"

Re: [dpdk-dev] [PATCH] vhost: return ready when at least 1 vring is configured

Hi Maxime, Vhost-crypto has exactly the same issue. Changpeng's patch fixed it. Could you give me a shout when your patch is out, so I can have a test? Regards, Fan > -Original Message- > From: dev On Behalf Of Maxime Coquelin > Sent: Wednesday, September 23, 2020 9:05 AM > To: Liu, Cha

Re: [dpdk-dev] [PATCH] vhost: return ready when at least 1 vring is configured

Hi Fan, The patch is already merged in main branch: commit 09424c3f74311555c33d3d4cdc2ca3654ce13b1c Author: Maxime Coquelin Date: Wed Sep 23 11:49:02 2020 +0200 vhost: fix external backends readiness Commit d0fcc38f5fa4 ("vhost: improve device readiness notifications") makes the

Re: [dpdk-dev] [PATCH v5 00/41] Pipeline alignment with the P4 language

On Wed, Sep 23, 2020 at 8:06 PM Cristian Dumitrescu wrote: > > This patch set introduces a new pipeline type that combines the DPDK > performance with the flexibility of the P4-16 language[1]. The new API > can be used either by itself to code a complete software switch (SWX) > or data plane app,

Re: [dpdk-dev] [PATCH v9 00/11] adding support for python 3 only

On Tue, 29 Sep 2020 11:22:13 +0100 Kevin Laatz wrote: > This patch set converts all python scripts in the project to use > python3 only and removes all deprecation notices associated with these > changes. This is due to python2 being EOL in January 2020. > > --- > v6: > - rebased, removing con

Re: [dpdk-dev] [PATCH v2 1/8] crypto/bcmfs: add BCMFS driver

Hi Akhil, On Tue, Sep 29, 2020 at 12:19 AM Akhil Goyal wrote: > > Hi Vikas, > > > +BCMFS crypto PMD depend upon the devices present in the path > > +/sys/bus/platform/devices/fs/ on the platform. > > +Each cryptodev PMD instance can be attached to the nodes present > > +in the mentioned path. > >

Re: [dpdk-dev] [PATCH v2 2/8] crypto/bcmfs: add vfio support

Hi Akhil, On Tue, Sep 29, 2020 at 12:30 AM Akhil Goyal wrote: > > Hi Vikas, > > > Subject: [PATCH v2 2/8] crypto/bcmfs: add vfio support > > > > Add vfio support for device. > > > > Signed-off-by: Vikas Gupta > > Signed-off-by: Raveendra Padasalagi > > Reviewed-by: Ajit Khaparde > > --- > > d

Re: [dpdk-dev] [PATCH v2 3/8] crypto/bcmfs: add apis for queue pair management

Hi Akhil, On Tue, Sep 29, 2020 at 12:59 AM Akhil Goyal wrote: > > > Subject: [PATCH v2 3/8] crypto/bcmfs: add apis for queue pair management > > > > diff --git a/drivers/crypto/bcmfs/bcmfs_hw_defs.h > > b/drivers/crypto/bcmfs/bcmfs_hw_defs.h > > new file mode 100644 > > index 0..ecb0c09ba

[dpdk-dev] [PATCH v2 0/2] net: add CRC run-time checks and AVX512/VPCLMULQDQ based CRC

This patchset makes two significant enhancements to the CRC modules of the rte_net library: 1) Adds run-time selection of the optimal architecture-specific CRC path. Previously the selection was solely made at compile-time, meaning it could only be built and run on the same generation of CPU

[dpdk-dev] [PATCH v2 2/2] net: add support for AVX512/VPCLMULQDQ based CRC

This patch enables the optimized calculation of CRC32-Ethernet and CRC16-CCITT using the AVX512 and VPCLMULQDQ instruction sets. This CRC implementation is built if the compiler supports the required instruction sets. It is selected at run-time if the host CPU, again, supports the required instruct

[dpdk-dev] [PATCH v2 1/2] net: add run-time architecture specific CRC selection

This patch adds support for run-time selection of the optimal architecture-specific CRC path, based on the supported instruction set(s) of the CPU. The compiler option checks have been moved from the C files to the meson script. The rte_cpu_get_flag_enabled function is called automatically by the

Re: [dpdk-dev] [PATCH v2 6/8] crypto/bcmfs: add session handling and capabilities

Hi Akhil, On Tue, Sep 29, 2020 at 1:16 AM Akhil Goyal wrote: > > Hi Vikas, > > > diff --git a/doc/guides/cryptodevs/features/bcmfs.ini > > b/doc/guides/cryptodevs/features/bcmfs.ini > > new file mode 100644 > > index 0..82d2c639d > > --- /dev/null > > +++ b/doc/guides/cryptodevs/features/

Re: [dpdk-dev] [PATCH v4] net/mlx5: relaxed ordering for multi-packet RQ buffer refcnt

Hi Raslan, It seems that there are no more comments for this patch. So shall we proceed further? Thanks, Phil Yang > -Original Message- > From: Alexander Kozyrev > Sent: Thursday, September 10, 2020 9:37 AM > To: Honnappa Nagarahalli ; Phil Yang > ; akozy...@mellanox.com; rasl...@mellan

Re: [dpdk-dev] [PATCH v9 00/11] adding support for python 3 only

On 29/09/2020 13:22, David Marchand wrote: Hello Kevin, On Tue, Sep 29, 2020 at 12:27 PM Kevin Laatz wrote: This patch set converts all python scripts in the project to use python3 only and removes all deprecation notices associated with these changes. This is due to python2 being EOL in Janua

[dpdk-dev] [PATCH 2/4] doc/api: align output folder with sphinx guides

The API docs were output to "/doc/api/api" folder, which was ugly-looking with the repeated "api", and inconsistent with the sphinx guides which were written to "/doc/guides/html". Changing the doxgen output folder to "html" fixes both these issues. Signed-off-by: Bruce Richardson --- doc/api/me

[dpdk-dev] [PATCH 0/4] improve output when building docs

When building the documentation, a lot of text is output, meaning that any warnings can be missed in all the text. Unfortunately, ninja merges both stderr and stdout of all tasks so one cannot just redirect stdout to a separate location as part of the build command to rectify that. Therefore, since

[dpdk-dev] [PATCH 1/4] doc/api: hide verbose doxygen standard output

The standard output of doxygen is very verbose, and since ninja mixes stdout and stderr together it makes it difficult to see any warnings from the doxygen run. Therefore, we can just log the standard output to file, and only output the stderr to make warnings clear. Suggested-by: Thomas Monjalon

[dpdk-dev] [PATCH 4/4] doc/guides: suppress printing out standard output

To see only errors and warnings from the doc builds, we can send the standard output text to a logfile and have only the stderr messages printed. This is similar to what is done for the API documentation. Signed-off-by: Bruce Richardson --- buildtools/call-sphinx-build.py | 6 -- 1 file chan

  1   2   3   >