[PATCH v2 4/4] net/nfp: vector Rx function supports parsing ptype

From: Long Wu Vector AVX2 Rx function supports parsing packet type and set it to mbuf. Signed-off-by: Long Wu Reviewed-by: Chaoyong He --- drivers/net/nfp/nfp_net_common.c| 2 +- drivers/net/nfp/nfp_rxtx.c | 2 +- drivers/net/nfp/nfp_rxtx.h | 3 +++ drivers/net/nfp/nfp_r

[PATCH v2 3/4] net/nfp: support AVX2 Rx function

From: Long Wu Use AVX2 instructions to accelerate Rx performance. The acceleration only works on X86 machine. Signed-off-by: Peng Zhang Signed-off-by: Long Wu Reviewed-by: Chaoyong He --- drivers/net/nfp/nfp_ethdev.c| 2 +- drivers/net/nfp/nfp_ethdev_vf.c | 2 +- drivers/net/

[PATCH v2 2/4] net/nfp: support AVX2 Tx function

From: Long Wu Use AVX2 instructions to accelerate Tx performance. The acceleration only works on X86 machine. Signed-off-by: Long Wu Reviewed-by: Chaoyong He --- drivers/net/nfp/meson.build | 15 + drivers/net/nfp/nfdk/nfp_nfdk.h | 1 + drivers/net/nfp/nfdk/nfp_

[PATCH v2 1/4] net/nfp: export more interfaces of NFDk

From: Long Wu NFP will support NFDk vector Tx function, so move some functions to header file for use by vector Tx function. Signed-off-by: Long Wu Reviewed-by: Chaoyong He --- drivers/net/nfp/nfdk/nfp_nfdk.h| 45 ++ drivers/net/nfp/nfdk/nfp_nfdk_dp.c | 43 +---

[PATCH v2 0/4] support AVX2 instruction Rx/Tx function

This patch series add the support of Rx/Tx function using the AVX2 instruction. --- v2: rebase to the latest main branch. --- Long Wu (4): net/nfp: export more interfaces of NFDk net/nfp: support AVX2 Tx function net/nfp: support AVX2 Rx function net/nfp: vector Rx function supports parsi

RE: [PATCH 0/4] support AVX2 instruction Rx/Tx function

> On 6/19/2024 3:59 AM, Chaoyong He wrote: > > This patch series add the support of Rx/Tx function using the > > AVX2 instruction. > > > > Long Wu (4): > > net/nfp: export more interfaces of NFDk > > net/nfp: support AVX2 Tx function > > net/nfp: support AVX2 Rx function > > net/nfp: vector

答复: [PATCH v2] r8125: add r8125 ethernet poll mode driver

Dear Ferruh, I'm delighted to receive your email. I'd like to address some of the questions you raised. The drivers supported by the PMD uploaded by Realtek will include all in-market Realtek PCIe network card chips. Currently, we plan to first upload drivers supporting 2.5G and 5G network car

[PATCH] net/nfp: fix offload drop flow rule failed

From: Long Wu The issue is caused by an invalid exit logic in flow compile function. Fixes: 4839d4b4e180 ("net/nfp: forbid offload flow rules with empty action list") Cc: chaoyong...@corigine.com Cc: sta...@dpdk.org Signed-off-by: Long Wu Reviewed-by: Chaoyong He Reviewed-by: Peng Zhang ---

[PATCH v4] vhost: fix crash caused by accessing a freed vsocket

From: Gongming Chen When a vhost user message handling error in the event dispatch thread, vsocket reconn is added to the reconnection list of the reconnection thread. Since the reconnection, event dispatching and app configuration thread do not have common thread protection restrictions, the app

Re: [PATCH 07/15] net/ena/base: update copyrights comments

On 08-07-2024 00:27, Ferruh Yigit wrote: On 7/2/2024 3:46 PM, shaib...@amazon.com wrote: From: Shai Brandes copyright dates are not mandatory to be maintained, therefore the range of years was removed. In addition, the copyrights lines were separated into two comments. Signed-off-by: Shai B

Re: [PATCH v2] net/memif: fix buffer overflow in zero copy Rx

On 07/07/2024 21:46, Mihai Brodschi wrote: > > > On 07/07/2024 18:18, Mihai Brodschi wrote: >> >> >> On 07/07/2024 17:05, Ferruh Yigit wrote: >>> >>> My expectation is numbers should be like following: >>> >>> Initially: >>> size = 256 >>> head = 0 >>> tail = 0 >>> >>> In first refill: >>>

[PATCH v2] net/nfp: fix port action core dump

From: Long Wu The app will be destroyed if add a port flow action which its destination port id is invalid. Add a check to avoid the problem. Fixes: bf4416016c66 ("net/nfp: split out flow item check logic") Fixes: 18c0ba1d69a8 ("net/nfp: split out flow action check logic") Cc: chaoyong...@corig

[PATCH v2 7/7] net/nfp: add trace point about flow rule action

Add 13 trace points of control message related with flow rule action list. Signed-off-by: Chaoyong He Reviewed-by: Long Wu Reviewed-by: Peng Zhang --- drivers/net/nfp/flower/nfp_flower_cmsg.c | 78 ++ drivers/net/nfp/nfp_trace.c | 39 +++ drivers/net/nfp/nfp_trace.h

[PATCH v2 6/7] net/nfp: add trace point about flow rule pattern

Add 11 trace points of control message related with flow rule pattern. Signed-off-by: Chaoyong He Reviewed-by: Long Wu Reviewed-by: Peng Zhang --- drivers/net/nfp/flower/nfp_flower_cmsg.c | 152 + drivers/net/nfp/nfp_trace.c | 33 drivers/net/nfp/nfp_trace.h

[PATCH v2 5/7] net/nfp: add trace point about flow rule

Add the trace point of control message related with flow rule meta data. Signed-off-by: Chaoyong He Reviewed-by: Long Wu Reviewed-by: Peng Zhang --- drivers/net/nfp/flower/nfp_flower_cmsg.c | 12 drivers/net/nfp/nfp_trace.c | 6 ++ drivers/net/nfp/nfp_trace.h

[PATCH v2 4/7] net/nfp: refactor to prepare for add flow trace point

Refactor 'nfp_flower_flow' module, move some macro into the header file in order to prepare for flow trace points. Signed-off-by: Chaoyong He Reviewed-by: Long Wu Reviewed-by: Peng Zhang --- drivers/net/nfp/flower/nfp_flower_flow.c | 46 drivers/net/nfp/flower/nfp_flow

[PATCH v2 3/7] net/nfp: add trace point about Qos

Add 1 trace point of control message related with Qos. Signed-off-by: Chaoyong He Reviewed-by: Long Wu Reviewed-by: Peng Zhang --- drivers/net/nfp/flower/nfp_flower_cmsg.c | 4 drivers/net/nfp/nfp_trace.c | 3 +++ drivers/net/nfp/nfp_trace.h | 29 ++

[PATCH v2 2/7] net/nfp: add trace point about tunnel

Add 6 trace points of control message related with tunnel. Signed-off-by: Chaoyong He Reviewed-by: Long Wu Reviewed-by: Peng Zhang --- drivers/net/nfp/flower/nfp_flower_cmsg.c | 14 +++ drivers/net/nfp/nfp_trace.c | 18 +++ drivers/net/nfp/nfp_trace.h | 133

[PATCH v2 1/7] net/nfp: add trace points about port

Add 3 trace points of control message related with representor port, also modify the needed logic. Signed-off-by: Chaoyong He Reviewed-by: Long Wu Reviewed-by: Peng Zhang --- drivers/net/nfp/flower/nfp_flower_cmsg.c | 17 +-- drivers/net/nfp/flower/nfp_flower_cmsg.h | 14 +++--- drivers/ne

[PATCH v2 0/7] add trace support for control message

This patch series add trace support for control message send to flower firmware. --- v2: rebase to the latest main branch. --- Chaoyong He (7): net/nfp: add trace points about port net/nfp: add trace point about tunnel net/nfp: add trace point about Qos net/nfp: refactor to prepare for ad

[PATCH v3] vhost: fix crash caused by accessing a freed vsocket

From: Gongming Chen When a vhost user message handling error in the event dispatch thread, vsocket reconn is added to the reconnection list of the reconnection thread. Since the reconnection, event dispatching and app configuration thread do not have common thread protection restrictions, the app

[PATCH v2] vhost: fix crash caused by accessing a freed vsocket

From: Gongming Chen When a vhost user message handling error in the event dispatch thread, vsocket reconn is added to the reconnection list of the reconnection thread. Since the reconnection, event dispatching and app configuration thread do not have common thread protection restrictions, the app

RE: [PATCH 0/7] add trace support for control message

> On 6/19/2024 11:07 AM, Chaoyong He wrote: > > This patch series add trace support for control message send to flower > > firmware. > > > > Chaoyong He (7): > > net/nfp: add trace points about port > > net/nfp: add trace point about tunnel > > net/nfp: add trace point about Qos > > net/nfp

RE: [PATCH 01/23] net/nfp: fix dereference of null pointer

> > Fixes: 149850c5b8ec ("net/nfp: support xstats for flower firmware") > > Cc: chaoyong...@corigine.com > > Cc: sta...@dpdk.org > > > > Signed-off-by: Peng Zhang > > > > Hi Chaoyong, > > As maintainer of the driver, I expect your review on the patch, and since > whole > series has it but not t

Re: [PATCH 00/17] NFP bugfix

On 6/24/2024 2:57 AM, Chaoyong He wrote: > This patch series consist a branch of bugfix patches from NFP PMD, > in order to speed up the review process and avoid send them out one > by one. > > Chaoyong He (4): > net/nfp: fix coredump caused by firmware abnormal cleanup > net/nfp: forbid offlo

Re: [PATCH 17/17] net/nfp: fix port action core dump

On 6/24/2024 2:57 AM, Chaoyong He wrote: > From: Long Wu > > The app will be destroyed if add a port flow action which its > destination port id is invalid. > > Add a check to avoid the problem. > > Cc: chaoyong...@corigine.com > Cc: sta...@dpdk.org > > Signed-off-by: Long Wu > This patch in

Re: [PATCH] net/hns3: support more VLAN fields matching

On 7/4/2024 3:15 AM, Jie Hai wrote: > The commit 09315fc83861 ("ethdev: add VLAN attributes to ethernet > and VLAN items") introduces ``has_vlan`` and ``has_more_vlan`` > fields in items ETH and VLAN. This patch adds support for these > fields. The usage is documented in hns3.rst. > > Signed-off-b

Re: [PATCH 0/7] add trace support for control message

On 6/19/2024 11:07 AM, Chaoyong He wrote: > This patch series add trace support for control message send > to flower firmware. > > Chaoyong He (7): > net/nfp: add trace points about port > net/nfp: add trace point about tunnel > net/nfp: add trace point about Qos > net/nfp: refactor to pre

Re: [PATCH] net/gve: fix RSS hash endianness in DQO format

On 7/3/2024 2:31 PM, Shreesh Adiga wrote: > The hash field in struct gve_rx_compl_desc_dqo defined in > gve_desc_dqo.h has the type __le32. Therefore the hash must be > read in little endian rather than big endian. > > Bugzilla ID: 1441 > > Signed-off-by: Shreesh Adiga <16567adigashre...@gmail.co

Re: [PATCH] net/nfp: add partial flow control message

On 6/19/2024 11:03 AM, Chaoyong He wrote: > From: Long Wu > > With new firmware, NFP PMD will use partial flow control > message to notify firmware the configuration of mark&queue > action. Moreover, if partial flow control message contains > both queue&mark configuration, the whole flow control

Re: [PATCH 00/15] net/ena: driver release 2.10.0

On 7/2/2024 3:46 PM, shaib...@amazon.com wrote: > From: Shai Brandes > > Hi all, > This release contains: > 1. HAL upgrade to latest divided into separate patches. > 2. three bug fixes. > 3. restructuring of the Rx checksum code for readability > 4. restructuring of the device uninit flow which i

Re: [PATCH 12/15] net/ena: fix invalid return value check

On 7/2/2024 3:46 PM, shaib...@amazon.com wrote: > From: Shai Brandes > > Removed the sign inversion for when checking if > ena_com_set_host_attributes returns ENA_COM_UNSUPPORTED. > ENA_COM_UNSUPPORTED is defined as -EOPNOTSUPP, so the extra sign > inversion is wrong. > > Fixes: 3adcba9a8987 ("n

Re: [PATCH 07/15] net/ena/base: update copyrights comments

On 7/2/2024 3:46 PM, shaib...@amazon.com wrote: > From: Shai Brandes > > copyright dates are not mandatory to be maintained, > therefore the range of years was removed. > In addition, the copyrights lines were separated > into two comments. > > Signed-off-by: Shai Brandes > --- > drivers/net/e

Re: [PATCH 00/23] support flower firmware with multiple PF

On 6/19/2024 10:58 AM, Chaoyong He wrote: > This patch series refactor some data structure and logic in order > to support the flower firmware with multiple PF. > > Peng Zhang (23): > net/nfp: fix dereference of null pointer > net/nfp: disable ctrl VNIC queues > net/nfp: fix dereference of n

Re: [PATCH 01/23] net/nfp: fix dereference of null pointer

On 6/19/2024 10:58 AM, Chaoyong He wrote: > From: Peng Zhang > > The original logic can call the uninitialized parameter 'process_private' > to get 'struct nfp_net_hw_priv', so the null pointer will cause a segment > fault problem. > > This commit will add the "struct nfp_repr_init" as the param

Re: [PATCH] net/ionic: fix double-free of mbufs when emptying array

On 7/1/2024 4:19 PM, Andrew Boyer wrote: > The bulk-allocation array is used back to front, so we need to free > everything before the marker, not after it. Flip ionic_empty_array() > so that it frees from 0 to the provided index. Adjust the callers > as needed. > > Fixes: 218afd825bca ("net/ionic

Re: [PATCH] net/nfp: fix flow mask table entry problem

On 6/19/2024 10:53 AM, Chaoyong He wrote: > The mask table entry using reference count to manage the flow rules > which have the same mask, and the new mask table entry will indicated by > flag 'NFP_FL_META_FLAG_MANAGE_MASK'. > > The abuse of 'NFP_FL_META_FLAG_MANAGE_MASK' will cause the firmware

Re: [PATCH] net/nfp: fix the allocate of switch domain

On 6/19/2024 10:51 AM, Chaoyong He wrote: > The check of 'RTE_ETH_DEV_SWITCH_DOMAIN_ID_INVALID' cause the following > call of 'rte_eth_switch_domain_alloc()' never trigger. > > Fix this by just remove the check logic. > > Fixes: e1124c4f8a45 ("net/nfp: add flower representor framework") > Cc: sta

Re: [PATCH v2] net/memif: fix buffer overflow in zero copy Rx

On 07/07/2024 18:18, Mihai Brodschi wrote: > > > On 07/07/2024 17:05, Ferruh Yigit wrote: >> >> My expectation is numbers should be like following: >> >> Initially: >> size = 256 >> head = 0 >> tail = 0 >> >> In first refill: >> n_slots = 256 >> head = 256 >> tail = 0 >> >> Subsequent ru

Re: [EXTERNAL] [PATCH v2 2/2] eal: add Arm WFET in power management intrinsics

> On Jul 5, 2024, at 5:10 PM, Pavan Nikhilesh Bhagavatula > wrote: > >> 04/07/2024 16:55, Stephen Hemminger: >>> On Thu, 04 Jul 2024 16:14:42 +0200 >>> Thomas Monjalon wrote: >>> >> Let’s ask Pavan why this flag is used in cn10k driver. >> >> From our perspective, WFE is availab

Re: [PATCH v2] net/memif: fix buffer overflow in zero copy Rx

On 07/07/2024 17:05, Ferruh Yigit wrote: > On 7/7/2024 6:50 AM, Mihai Brodschi wrote: >> Hi Ferruh, >> >> On 07/07/2024 05:12, Ferruh Yigit wrote: >>> On 6/28/2024 10:01 PM, Mihai Brodschi wrote: rte_pktmbuf_alloc_bulk is called by the zero-copy receiver to allocate new mbufs to be prov

Re: [PATCH] memzone: remove unnecessary null checks

19/06/2024 02:15, Tyler Retzlaff: > On Tue, Jun 18, 2024 at 04:25:08PM -0700, Stephen Hemminger wrote: > > The function rte_memzone_free() already checks if the argument > > is NULL. Add it to the coccinelle fix script and correct > > all the existing code. > > > > Signed-off-by: Stephen Hemminger

Re: [PATCH] telemetry: fix potential connection issue

26/06/2024 04:13, Jiale, SongX: > From: Zhichao Zeng > > > > For calling memcpy when the source and destination addresses are the same, > > there is a small probability that there will be a copy error issue in some > > environments, resulting in a failure to connect. > > > > This patch uses memm

Re: [DPDK][PATCH v4] config/arm: add Ampere AmpereOneAC04 platform

01/07/2024 11:27, Ruifeng Wang: > > On 2024/6/28 11:22 AM, Yutang Jiang wrote: > > The AmpereOneAC04 is efficient Cloud Native CPU: > >Up to 192 Cores > >2MB Private L2 Cache per Core > >8 channel DDR5 > >128 lanes PCIe Gen5 > > > > Signed-off-by: Yutang Jiang > > Acked-by: Ruif

Re: [PATCH v2] net/memif: fix buffer overflow in zero copy Rx

On 7/7/2024 6:50 AM, Mihai Brodschi wrote: > Hi Ferruh, > > On 07/07/2024 05:12, Ferruh Yigit wrote: >> On 6/28/2024 10:01 PM, Mihai Brodschi wrote: >>> rte_pktmbuf_alloc_bulk is called by the zero-copy receiver to allocate >>> new mbufs to be provided to the sender. The allocated mbuf pointers >>

Re: [PATCH] net/gve: fix RSS hash endianness in DQO format

On 7/7/2024 3:53 AM, Shreesh Adiga wrote: > Thank you Joshua for your review and suggestions.  > Appreciate the help with dpdk process. > I've sent a new patch with "Fixes" tag to sta...@dpdk.org > .  > I'm not sure if same needs to be sent here to dev@dpdk.org >

Re: [PATCH v10 1/4] hash: pack the hitmask for hash in bulk lookup

05/07/2024 19:43, Yoan Picchi: > On 7/4/24 21:31, David Marchand wrote: > > On Wed, Jul 3, 2024 at 7:13 PM Yoan Picchi wrote: > >> --- /dev/null > >> +++ b/lib/hash/compare_signatures_arm_pvt.h > > > > I guess pvt stands for private. > > No need for such suffix, this header won't be exported in a

[PATCH 10/10] net/mlx5/hws: fix NA64 copy TOS field instead of TTL

From: Erez Shitrit We don't have enough registers to copy TTL and TOS, so we will set TTL to be the default value (64) and will copy TOS. Fixes: 06d969a8c5b8 ("net/mlx5/hws: support NAT64 flow action") Cc: ere...@nvidia.com Cc: sta...@dpdk.org Signed-off-by: Erez Shitrit Acked-by: Matan Azrad

[PATCH 09/10] net/mlx5/hws: fix NAT64 csum issue

From: Erez Shitrit Due to HW limitation we got two csum's that were not correct, udp and ip, both of them were not calculated correctly by the HW. By adding the next W/A we allow the HW to collect it well: Separate the protocol field and zero all the addresses before fixed the UDP csum. We saw th

[PATCH 08/10] net/mlx5/hws: take out not needed variable

From: Erez Shitrit Removing a redundant variable. Was there from day 1, not in use. Fixes: f8c8a6d8440d ("net/mlx5/hws: add action object") Cc: ere...@nvidia.com Cc: sta...@dpdk.org Signed-off-by: Erez Shitrit Acked-by: Matan Azrad --- drivers/net/mlx5/hws/mlx5dr_pat_arg.h | 1 - 1 file chan

[PATCH 07/10] net/mlx5/hws: fix incorrect port ID on root item convert

From: Alex Vesker When calling item convert function we need to pass the port_id in the attributes. This value should be passed not only for cases that match on PORT related items, to resolve we will always pass it. Fixes: 572fe9ef2f46 ("net/mlx5/hws: fix port ID for root table") Cc: ere...@nvid

[PATCH 06/10] net/mlx5/hws: fix deletion of action vport

From: Erez Shitrit No more ignoring this action while destroying it. Fixes: f8c8a6d8440d ("net/mlx5/hws: add action object") Cc: ere...@nvidia.com Cc: sta...@dpdk.org Signed-off-by: Erez Shitrit Acked-by: Matan Azrad --- drivers/net/mlx5/hws/mlx5dr_action.c | 4 drivers/net/mlx5/hws/mlx

[PATCH 05/10] net/mlx5/hws: strictly range templates check fix

Using range and non range templates is not allowed, and in HWS there is a check that enforce that limitation with constantly check that, in a loop, if the current template defined as range, the last one should also be defined as range. But, in the case where there are two templates in the following

[PATCH 04/10] net/mlx5/hws: fix memory leak in modify header free

From: Erez Shitrit When creating action from type MLX5DR_ACTION_TYP_REFORMAT_TNL_L3_TO_L2 we use modify-header object, we support few of that type at the same time over this action depends on the number of headers. Now when destroying the modify-header object we run over the number_of_patterns, t

[PATCH 03/10] net/mlx5/hws: set eswitch owner vhc ID valid accordingly

From: Erez Shitrit eswitch_owner_vhca_id_valid value should be set to 1 only on merged-eswitch device. Signed-off-by: Erez Shitrit Acked-by: Matan Azrad --- drivers/net/mlx5/hws/mlx5dr_action.c | 12 drivers/net/mlx5/hws/mlx5dr_cmd.c| 3 ++- drivers/net/mlx5/hws/mlx5dr_cmd.h

[PATCH 02/10] net/mlx5: fix matcher object memory leak

From: Mahmoud Maatuq This makes sure that the allocated matcher object is freed for all branches that return NULL. Coverity issue: 426424 Fixes: 27d171b88031 ("net/mlx5: abstract flow action and enable reconfigure") Signed-off-by: Mahmoud Maatuq Acked-by: Bing Zhao Acked-by: Dariusz Sosnowski

[PATCH 01/10] net/mlx5: add hairpin out of buffer counter

From: Shani Peretz Currently mlx5 PMD exposes rx_out_of_buffer counter that tracks packets dropped when Rx queue was full. To provide more granular statistics, this patch splits the `rx_out_of_buffer` counter into two separate counters: 1. hairpin_out_of_buffer - This counter specifically tracks

[PATCH] net/mlx5: fix IPv6-in-IPv6 tunnel recognition

The PMD did not recognize IPv6-in-IPv6 tunnel if IPv6 routing extension was placed between IPv6 outer and inner headers. The patch fixes IPv6-in-IPv6 tunnel recognition. Fixes: 80c676259a04 ("net/mlx5: validate HWS template items") Signed-off-by: Gregory Etelson Acked-by: Dariusz Sosnowski ---

[PATCH] net/mlx5: fix compilation warning in GCC-9.1

GCC has introduced a bugfix in 9.1 that changed GCC ABI in ARM setups: https://gcc.gnu.org/gcc-9/changes.html ``` On Arm targets (arm*-*-*), a bug in the implementation of the procedure call standard (AAPCS) in the GCC 6, 7 and 8 releases has been fixed: a structure containing a bit-field based on

[PATCH] net/mlx5: fix indexed pool resize

On success, indexed pool resize sets maximal pool entries number to the `num_entries` parameter value. The patch fixes maximal pool entries assignment. The patch also adds `error` parameter to log error types. Fixes: 89578504edd9 ("net/mlx5: add ipool resize function") Signed-off-by: Gregory Et