Re: Difference between i40e and i40en?

Hello Édouard, On Tue, May 3, 2022 at 9:05 PM Edouard Gaulué wrote: > > I just received a brand new server with an Intel X710 T4L inside. I > hoped I could use it at 5GBaseT speed as claimed by Intel, but I couldn't. > > In fact, I run ESXi 6.7U3 on this new server and it looks it doesn't > suppo

Re: [PATCH] avoid AltiVec keyword vector

On Tue, May 03, 2022 at 05:44:26PM +0200, Thomas Monjalon wrote: > 03/05/2022 14:30, Bruce Richardson: > > On Tue, May 03, 2022 at 02:03:21PM +0200, Thomas Monjalon wrote: > > > The keyword "vector" may conflict easily. > > > As a rule, it is better to use the alternative keyword "__vector". > > I

[dpdk-dev][PATCH v2 3/3] net/cnxk: adding cnxk support to configure custom sa index

From: Kiran Kumar K Adding cnxk device driver support to configure custom sa index. Custom sa index can be configured as part of the session create as SPI, and later original SPI can be updated using session update. Signed-off-by: Kiran Kumar K --- doc/api/doxy-api-index.md | 3 +-

[dpdk-dev][PATCH v2 3/3] net/cnxk: adding cnxk support to configure custom sa index

From: Kiran Kumar K Adding cnxk device driver support to configure custom sa index. Custom sa index can be configured as part of the session create as SPI, and later original SPI can be updated using session update. Signed-off-by: Kiran Kumar K --- doc/api/doxy-api-index.md | 3 +-

[dpdk-dev][PATCH v2 2/3] net/cnxk: add devargs support to parse custom SA action

From: Kiran Kumar K Adding devargs support to parse custom sa action. Devargs can be specified in the following way. -a 0002:02:00.0,custom_sa_act=1 Signed-off-by: Kiran Kumar K --- doc/guides/nics/cnxk.rst | 20 drivers/net/cnxk/cnxk_ethdev_devargs.c | 10 ++

[dpdk-dev][PATCH v2 1/3] common/cnxk: add ROC support to parse cnxk custom sa action

From: Kiran Kumar K Adding ROC Flow changes to parse custom SA action for cnxk device. When custom sa action is enabled, VTAG actions are not allowed. And custom SA index will be calculated based on SA_HI and SA_LO values. This allows the potential for a MCAM entry to match many SAs, rather than

[dpdk-dev][PATCH v2 2/3] net/cnxk: add devargs support to parse custom SA action

From: Kiran Kumar K Adding devargs support to parse custom sa action. Devargs can be specified in the following way. -a 0002:02:00.0,custom_sa_act=1 Signed-off-by: Kiran Kumar K --- doc/guides/nics/cnxk.rst | 20 drivers/net/cnxk/cnxk_ethdev_devargs.c | 10 ++

[dpdk-dev][PATCH v2 1/3] common/cnxk: add ROC support to parse cnxk custom sa action

From: Kiran Kumar K Adding ROC Flow changes to parse custom SA action for cnxk device. When custom sa action is enabled, VTAG actions are not allowed. And custom SA index will be calculated based on SA_HI and SA_LO values. This allows the potential for a MCAM entry to match many SAs, rather than

RE: [PATCH] eal: allow worker lcore stacks to be allocated from hugepage memory

> -Original Message- > From: Don Wallwork > Sent: Wednesday, May 4, 2022 03:47 > To: Wang, Haiyue ; dev@dpdk.org > Cc: step...@networkplumber.org; m...@smartsharesystems.com; Burakov, Anatoly > ; > dmitry.kozl...@gmail.com; Richardson, Bruce ; > honnappa.nagaraha...@arm.com; > n...@arm.c

RE: [PATCH v8 2/2] hash: unify crc32 selection for x86 and Arm

> -Original Message- > From: Pavan Nikhilesh > Sent: Friday, April 29, 2022 9:17 AM > To: ruifeng.w...@arm.com; Wang, Yipeng1 ; > Gobriel, Sameh ; Richardson, Bruce > ; Medvedkin, Vladimir > > Cc: jer...@marvell.com; dev@dpdk.org; Pavan Nikhilesh > > Subject: [PATCH v8 2/2] hash: unif

RE: [PATCH v8 1/2] hash: split x86 and SW hash CRC intrinsics

> -Original Message- > From: Pavan Nikhilesh > Sent: Friday, April 29, 2022 9:17 AM > To: ruifeng.w...@arm.com; Wang, Yipeng1 ; > Gobriel, Sameh ; Richardson, Bruce > ; Medvedkin, Vladimir > > Cc: jer...@marvell.com; dev@dpdk.org; Pavan Nikhilesh > > Subject: [PATCH v8 1/2] hash: split x

Multiple worker core groups

Hello, i am trying to setup an application with multiple parallel cores each with its own distributor workers. I am seeing panics and trying to figure out if there is a fundamemtnal limitation with such an approach. My setup looks like ( simplified) Core 1 -> (distributor cores) 3,4,5 Core

Re: [PATCH v3 3/5] app/testpmd: fix port status of slave device

03/05/2022 11:02, Min Hu (Connor) пишет: From: Huisong Li Starting or stopping a bonded port also starts or stops all active slaves under the bonded port. If this port is a bonded device, we need to modify the port status of all slaves. Fixes: 0e545d3047fe ("app/testpmd: check stopping port is

Re: librte_bpf: roadmap or any specific plans for this library

AFAIK linux BPF is restricted to work with a single argument only. I don't want DPDK version to fork too far away from 'canonical' version. Though, as I said above, nothing prevents you to create a struct with several fields, and pass pointer to that struct to your BPF program. Would such app

RE: [Patch v2] net/netvsc: report correct stats values

> Subject: Re: [Patch v2] net/netvsc: report correct stats values > > On 5/3/2022 8:14 PM, Long Li wrote: > >> Subject: Re: [Patch v2] net/netvsc: report correct stats values > >> > >> On 5/3/2022 7:18 PM, Long Li wrote: > Subject: Re: [Patch v2] net/netvsc: report correct stats values >

Re: [Patch v2] net/netvsc: report correct stats values

On 5/3/2022 8:14 PM, Long Li wrote: Subject: Re: [Patch v2] net/netvsc: report correct stats values On 5/3/2022 7:18 PM, Long Li wrote: Subject: Re: [Patch v2] net/netvsc: report correct stats values On Tue, 26 Apr 2022 22:56:14 +0100 Ferruh Yigit wrote: if (i < RTE_ETHDEV_Q

Re: [PATCH] eal: allow worker lcore stacks to be allocated from hugepage memory

On 5/3/2022 9:08 AM, Wang, Haiyue wrote: -Original Message- From: Don Wallwork Sent: Monday, May 2, 2022 22:11 To: dev@dpdk.org Cc: d...@xsightlabs.com; step...@networkplumber.org; m...@smartsharesystems.com; Burakov, Anatoly ; dmitry.kozl...@gmail.com; Richardson, Bruce ; honnappa.nag

RE: OVS DPDK DMA-Dev library/Design Discussion

> -Original Message- > From: Ilya Maximets > Sent: Thursday, April 28, 2022 2:00 PM > To: Richardson, Bruce > Cc: i.maxim...@ovn.org; Mcnamara, John ; Hu, Jiayu > ; Maxime Coquelin ; Van > Haaren, Harry ; Morten Brørup > ; Pai G, Sunil ; Stokes, > Ian > ; Ferriter, Cian ; ovs- > d...@ope

RE: [Patch v2] net/netvsc: report correct stats values

> Subject: Re: [Patch v2] net/netvsc: report correct stats values > > On 5/3/2022 7:18 PM, Long Li wrote: > >> Subject: Re: [Patch v2] net/netvsc: report correct stats values > >> > >> On Tue, 26 Apr 2022 22:56:14 +0100 > >> Ferruh Yigit wrote: > >> > if (i < RTE_ETHDEV_QUEU

Re: [PATCH 0/7] bugfix for hns3 PMD

On 4/6/2022 10:22 AM, Min Hu (Connor) wrote: This patch set contains 7 bugfix for hns3 PMD. Chengwen Feng (1): net/hns3: fix more mbufs are freed when Tx done cleanup Huisong Li (6): net/hns3: fix order of clearing imissed register in PF net/hns3: fix MAC and queues HW statistics rever

Re: [PATCH 7/7] net/hns3: remove redundant RSS tuple field

On 4/6/2022 10:22 AM, Min Hu (Connor) wrote: From: Huisong Li The 'rss_tuple_fields' in struct struct hns3_rss_conf::rss_tuple_sets is redundant. Because the enabled RSS tuple in PMD is already managed by the 'types' in struct hns3_rss_conf::conf. This patch removes this rebundant typo, redun

Re: [PATCH 5/7] net/hns3: fix RSS disable

On 4/6/2022 10:22 AM, Min Hu (Connor) wrote: From: Huisong Li Currently, hns3 PMD disable RSS by resetting redirection table when user set rss_hf to 0 so as to all packets go to queue 0. The implementation may cause following problems: 1) the same type packet may go to different queue on the ca

Re: [PATCH 3/7] net/hns3: fix pseudo-sharing between threads

On 4/6/2022 10:22 AM, Min Hu (Connor) wrote: From: Huisong Li Some fields in the end of 'struct hns3_rx_queue' and 'struct hns3_tx_queue' are not accessed in the I/O path. But these fields may be accessed in other threads, which may lead to the problem of pseudo-sharing of IO threads. ... pro

Re: [PATCH 2/7] net/hns3: fix MAC and queues HW statistics reversion

On 4/6/2022 10:22 AM, Min Hu (Connor) wrote: From: Huisong Li The MAC and queues statistics aren't 64-bit registers in hardware. If hardware statistics are not obtained for a long time, these statistics will be reversed. I think using 'overflow' instead of 'reversed' is more clear. As far a

Re: [PATCH v2] net/nfp: update how MAX MTU is read

On 4/20/2022 2:46 PM, Walter Heymans wrote: The 'max_rx_pktlen' value was previously read from hardware, which was set by the running firmware. This caused confusion due to different meanings of 'MAX_MTU'. This patch updates the 'max_rx_pktlen' to the maximum value that the NFP NIC can support. T

Re: [PATCH V2 1/4] net/bonding: fix non-active slaves aren't stopped

On 5/3/2022 7:54 AM, Min Hu (Connor) wrote: Hi, Ferruh, 在 2022/4/29 21:31, Ferruh Yigit 写道: On 4/29/2022 7:45 AM, Min Hu (Connor) wrote: Hi, Ferruh, 在 2022/4/27 2:19, Ferruh Yigit 写道: On 3/24/2022 3:00 AM, Min Hu (Connor) wrote: From: Huisong Li When stopping a bonded port, all slaves sho

Re: [Patch v2] net/netvsc: report correct stats values

On 5/3/2022 7:18 PM, Long Li wrote: Subject: Re: [Patch v2] net/netvsc: report correct stats values On Tue, 26 Apr 2022 22:56:14 +0100 Ferruh Yigit wrote: if (i < RTE_ETHDEV_QUEUE_STAT_CNTRS) { - stats->q_opackets[i] = txq->stats.packets; -

Re: [PATCH 0/8] vmxnet3 version V5 and V6

Reviewed-by: Jochen Behrens jbehr...@vmware.com From: Pankaj Gupta Date: Monday, May 2, 2022 at 9:23 PM To: Jochen Behrens , Yong Wang Cc: dev@dpdk.org , Pankaj Gupta Subject: [PATCH 0/8] vmxnet3 version V5 and V6 Pankaj Gupta (8): vmxnet3: Added V5 support vmx

RE: [Patch v2] net/netvsc: report correct stats values

> Subject: Re: [Patch v2] net/netvsc: report correct stats values > > On Tue, 26 Apr 2022 22:56:14 +0100 > Ferruh Yigit wrote: > > > > if (i < RTE_ETHDEV_QUEUE_STAT_CNTRS) { > > > - stats->q_opackets[i] = txq->stats.packets; > > > - stats->q_obyt

Re: [PATCH v2 28/28] net/cnxk: fix multi-seg extraction in vwqe path

On Fri, Apr 22, 2022 at 4:20 PM Nithin Dabilpuram wrote: > > Fix multi-seg extraction in vwqe path to avoid updating mbuf[] > array until it is used via cq0 path. > > Fixes: 7fbbc981d54f ("event/cnxk: support vectorized Rx event fast path") > Cc: pbhagavat...@marvell.com > Cc: sta...@dpdk.org > >

Re: [PATCH] avoid AltiVec keyword vector

03/05/2022 14:30, Bruce Richardson: > On Tue, May 03, 2022 at 02:03:21PM +0200, Thomas Monjalon wrote: > > The keyword "vector" may conflict easily. > > As a rule, it is better to use the alternative keyword "__vector". I forgot to say that we should #undef vector. > hi Thomas, > > could you per

[PATCH v2] tap: fix write-after-free and double free of intr_handle

rte_pmd_tun/tap_probe() allocates pmd->intr_handle in eth_dev_tap_create() and it should not be freed until rte_pmd_tap_remove() is called. Inspection of tap_rx_intr_vec_set() shows that the call to tap_tx_intr_vec_uninstall() was calling rte_intr_instance_free() but tap_tx_intr_vec_install() can

Re: [PATCH 1/1] tap: fix write-after-free and double free of intr_handle

On Mon, May 2, 2022 at 9:23 PM Quentin Armitage wrote: > > rte_pmd_tun/tap_probe() allocates pmd->intr_handle in > eth_dev_tap_create() and it should not be freed until > rte_pmd_tap_remove() is called. > > Inspection of tap_rx_intr_vec_set() shows that the call to > tap_tx_intr_vec_uninstall() wa

Re: [PATCH v8 2/2] hash: unify crc32 selection for x86 and Arm

On Fri, Apr 29, 2022 at 6:17 PM Pavan Nikhilesh wrote: > > Merge crc32 hash calculation public API implementation for x86 and Arm. > Select the best available CRC32 algorithm when unsupported algorithm > on a given CPU architecture is requested by an application. > > Previously, if an application

Difference between i40e and i40en?

Hi community, I just received a brand new server with an Intel X710 T4L inside. I hoped I could use it at 5GBaseT speed as claimed by Intel, but I couldn't. In fact, I run ESXi 6.7U3 on this new server and it looks it doesn't support this speed (2.5G neither). I downloaded Intel driver source

RE: [PATCH] eal: allow worker lcore stacks to be allocated from hugepage memory

> -Original Message- > From: Don Wallwork > Sent: Monday, May 2, 2022 22:11 > To: dev@dpdk.org > Cc: d...@xsightlabs.com; step...@networkplumber.org; > m...@smartsharesystems.com; Burakov, Anatoly > ; dmitry.kozl...@gmail.com; Richardson, Bruce > ; > honnappa.nagaraha...@arm.com; n...@ar

Re: [PATCH] avoid AltiVec keyword vector

On Tue, May 03, 2022 at 02:03:21PM +0200, Thomas Monjalon wrote: > The keyword "vector" may conflict easily. > As a rule, it is better to use the alternative keyword "__vector". > hi Thomas, could you perhaps clarify a bit more in the log, I'm not aware of it being a keyword generally. What woul

[PATCH] avoid AltiVec keyword vector

The keyword "vector" may conflict easily. As a rule, it is better to use the alternative keyword "__vector". Signed-off-by: Thomas Monjalon --- drivers/net/i40e/i40e_rxtx_vec_altivec.c | 214 ++-- drivers/net/mlx5/mlx5_rxtx_vec_altivec.h | 1078 - .../net/virtio/virtio

NVIDIA roadmap for 22.07

Please find below NVIDIA roadmap for 22.07 release: rte_flow new APIs: === [1] Add port flags for pre-configuration flow hints. [RFC,1/2] ethdev: port flags for pre-configuration flow hints. Introduce a flag to specify if the operations on each flow will be executed on the same queue.

Re: rte_bus_probe() does not exit on error

03/05/2022 11:52, Tyler Retzlaff: > On Mon, May 02, 2022 at 11:54:29PM +0200, Thomas Monjalon wrote: > > Hello, > > > > 02/05/2022 23:20, McDaniel, Timothy: > > > Hello DPDK community, > > > > > > I am following up on a question/comment that I submitted on April 18, for > > > which > > > I have

Re: [PATCH] net/netvsc: fix issues with hot adding multiple VF PCI devices

On 3/24/2022 5:46 PM, lon...@linuxonhyperv.com wrote: From: Long Li This patch fixes two issues with hot removing/adding a VF PCI device: 1. The original device argument is lost when it's hot added 2. If there are multiple VFs hot adding at the same time, some of the VFs may not get added succe

Re: rte_bus_probe() does not exit on error

On Mon, May 02, 2022 at 11:54:29PM +0200, Thomas Monjalon wrote: > Hello, > > 02/05/2022 23:20, McDaniel, Timothy: > > Hello DPDK community, > > > > I am following up on a question/comment that I submitted on April 18, for > > which > > I have not received any responses. See the original comment

Re: [PATCH v4 3/3] test/threads: add unit test for thread API

On Sun, May 01, 2022 at 12:18:36PM +0300, Dmitry Kozlyuk wrote: > 2022-04-26 00:50 (UTC-0700), Tyler Retzlaff: > > Establish unit test for testing thread api. Initial unit tests > > for rte_thread_{get,set}_affinity_by_id(). > > > > Signed-off-by: Narcisa Vasile > > Signed-off-by: Tyler Retzlaff

Softnic test failed + questions

Hi all, I'm a beginner with dpdk. With my team, we developed a dpdk app with the following pipeline: NIC RX -> RX Thread -> Worker Thread -> TX Thread -> NIC TX. Within the RX Thread, we parse some headers. Within the worker thread, we're using the hierarchical scheduler. To sum up, we want to r

Re: [dpdk-dev] [PATCH v3 6/7] app/proc-info: provide way to request info on owned ports

24/04/2022 07:34, Subendu Santra: > Hi Stephen, > > We were going through the patch set: > https://inbox.dpdk.org/dev/20200715212228.28010-7-step...@networkplumber.org/ > and hoping to get clarification on the behaviour if post mask is not > specified in the input to `dpdk-proc-info` tool. > >

Re: [dpdk-dev] [PATCH v3 6/7] app/proc-info: provide way to request info on owned ports

Hi Stephen, Could you please help us understand the rationale behind showing just the last non-owned port in case the port mask was not specified? I really appreciate your help in this regard. Regards, Subendu. On Sun, Apr 24, 2022 at 11:04 AM Subendu Santra wrote: > Hi Stephen, > > We were

Re: librte_bpf: roadmap or any specific plans for this library

Thanks for the feedback. >>Example of the new API: >> struct rte_bpf * >> rte_bpf_elf_image_load(const struct rte_bpf_prm *prm, char *image, >> size_t size, const char *sname); >> > Did you look at rte_bpf_load()? > Basically it works with already pre-load

Re: [dpdk-dev] [PATCH] common/cnxk: fix unaligned access to device memory

On Wed, Apr 20, 2022 at 12:08 PM wrote: > > From: Satheesh Paul > > Use mbox_memcpy() instead of memcpy() to prevent unaligned access. > > Fixes: 1f66919817e ("common/cnxk: improve MCAM entries management") > Cc: sta...@dpdk.org > > Signed-off-by: Satheesh Paul Acked-by: Jerin Jacob Applied to