Re: [PATCH v5 09/32] net/sssnic/base: add control queue

Hello Stephen, The length of response_data could up to 1920 bytes. Thanks. On 2023/9/18 10:36, Stephen Hemminger wrote: On Mon, 4 Sep 2023 12:56:35 +0800 wrote: rte_memcpy(cmd->response_data, rte_pktmbuf_mtod(mbuf, void *), + cmd->response_len); For sm

Minutes of Technical Board Meeting, 2023-Jun-28

Minutes of Technical Board Meeting, 2023-Jun-28 Members Attending - -Aaron -Hemant (Chair) -Jerin -Keven -Konstantin -Maxime -Stephen -Nathan -Ben NOTE: The technical board meetings every second Wednesday at https://meet.jit.si/DPDK at 3 pm UTC. Meetings are public, and DPDK comm

Re: [PATCH v5 08/32] net/sssnic/base: add work queue

Hello Stephen,  I'll fix it next version. On 2023/9/18 10:33, Stephen Hemminger wrote: On Mon, 4 Sep 2023 12:56:34 +0800 wrote: +void * +sssnic_workq_consume(struct sssnic_workq *workq, uint16_t num_entries, + uint16_t *ci) +{ + void *e; + uint16_t current_ci; + + if

Re: [PATCH v5 07/32] net/sssnic/base: add mailbox support

Hello Stephen, I'll fix it next version. On 2023/9/18 10:32, Stephen Hemminger wrote: On Mon, 4 Sep 2023 12:56:33 +0800 wrote: +struct sssnic_mbox_send_ctrl1_reg { + union { + uint32_t u32; + struct { + uint32_t resvd0 : 10; +

Re: [PATCH v5 06/32] net/sssnic/base: add message definition and utility

Hello, Stephen, data_buf is just used by control plane process, and it will not be used very frequtely, it has a little impact on IO performance. I'm not going to fix it. Thanks. On 2023/9/18 10:31, Stephen Hemminger wrote: On Mon, 4 Sep 2023 12:56:32 +0800 wrote: +static int +sssnic_msg

Re: [PATCH v5 04/32] net/sssnic: initialize hardware base

Hello Stephen, hw struct is just used in control plane, I think it has little impact on performance, howerver I am going to take your advice and fix it in next version of patches. Thanks. On 2023/9/18 10:28, Stephen Hemminger wrote: On Mon, 4 Sep 2023 12:56:30 +0800 wrote: static int

Re: [PATCH v5 00/32] Introduce sssnic PMD for 3SNIC's 9x0 serials Ethernet adapters

Hello Stephen, SSSNIC PMD does not support primary/secondary process right now, it can only run in primary process. We are going to support multiprocessnext release. Thanks. On 2023/9/18 10:37, Stephen Hemminger wrote: On Mon, 4 Sep 2023 12:56:26 +0800 wrote: From: Renyong Wan The sssn

Re: confusion about dpdk asan

On Thu, 14 Sep 2023 12:42:20 +0200 David Marchand wrote: > On Thu, Sep 14, 2023 at 10:59 AM 杨佳昊 wrote: > > > > Thank you for your reply. > > Please, let's keep this discussion on the ml. > Anybody can help and this thread may help others having issues with ASan. > Let's try to understand if s

[PATCH v4 26/26] net/nfp: extend the usage of nfp BAR from 8 to 24

Sync the logic from kernel driver, adjust the definition of structure, and extend the usage of nfp BAR from 8 to 24. This will greatly enhance the scalability of nfp PMD. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfp_ethdev.c | 8 +- drivers/net/n

[PATCH v4 25/26] net/nfp: refact the cppcore and PCIe module

Sync the logic from kernel driver, use the new entry function from the PCIe module instead of the cppcore module. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfp_ethdev.c | 9 +- drivers/net/nfp/nfpcore/nfp6000_pcie.c | 197 ++---

[PATCH v4 24/26] net/nfp: refact the PCIe module

Sync the logic from kernel driver and remove the unneeded header file include statements. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp6000_pcie.c | 211 + drivers/net/nfp/nfpcore/nfp_cpp.h | 9 ++ 2 files changed, 150 in

[PATCH v4 23/26] net/nfp: refact the cppcore module

Modify the comment and standard the coding style. Move the definition of data structure into the implement file to make the API small. Also sync the logic from kernel driver and remove the unneeded header file include statements. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- driv

[PATCH v4 22/26] net/nfp: add header file for PCIe module

Add a new header file for the PCIe module. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/meson.build | 2 +- .../nfpcore/{nfp_cpp_pcie_ops.c => nfp6000_pcie.c} | 2 ++ drivers/net/nfp/nfpcore/nfp6000_pcie.h | 13

[PATCH v4 21/26] net/nfp: add the dev module

Add the nfp_dev module to simplify the logic. The data structure we import can gather and replace several macro which scatter everywhere. Which also make the logic of choose value for different devices unnecessary and so eliminate such logic. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderl

[PATCH v4 20/26] net/nfp: rename data field to sync with kernel driver

Rename the data fields to sync with the logic in kernel driver. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/flower/nfp_flower.c | 6 +++--- drivers/net/nfp/nfp_common.h| 4 ++-- drivers/net/nfp/nfp_ethdev.c| 16 3 files changed

[PATCH v4 19/26] net/nfp: refact the mutex module

Add a header file to holds the API declarations of this module. Also sync the logic from kernel driver and remove the unneeded header file include statements. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_cpp.h | 16 -- drivers/net/nfp/nfpcore/nf

[PATCH v4 18/26] net/nfp: refact the nsp module

Move the definition of data structure into the implement file. Also sync the logic from kernel driver and remove the unneeded header file include statements. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfp_ethdev.c | 2 +- drivers/net/nfp/nfpcore/nfp

[PATCH v4 17/26] net/nfp: add a new header file

Split out the macro which are not easily find a DPDK substitute into a new header file. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_nsp.c | 1 + drivers/net/nfp/nfpcore/nfp_nsp.h | 23 -- drivers/net/nfp/nfpcore/nfp_nsp_eth.c

[PATCH v4 16/26] net/nfp: refact the target module

Move the function declaration into the header file, and sync some logic from kernel driver. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp6000/nfp6000.h | 20 drivers/net/nfp/nfpcore/nfp_cpp_pcie_ops.c | 1 + drivers/net/nfp/nfpco

[PATCH v4 15/26] net/nfp: refact the resource module

Sync the logic from kernel driver and remove the unneeded header file include statements. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_resource.c | 21 - 1 file changed, 8 insertions(+), 13 deletions(-) diff --git a/drivers/net/nf

[PATCH v4 14/26] net/nfp: refact the rtsym module

Add several read/write APIs and remove the unneeded header file include statements. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_rtsym.c | 305 +--- drivers/net/nfp/nfpcore/nfp_rtsym.h | 53 ++--- 2 files changed, 299 insertio

[PATCH v4 13/26] net/nfp: refact the mip module

Remove the unneeded header file include statements. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_mip.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/net/nfp/nfpcore/nfp_mip.c b/drivers/net/nfp/nfpcore/nfp_mip.c inde

[PATCH v4 12/26] net/nfp: refact the nffw module

Move the definition of data structure and macro into the implement file. Also sync the logic from kernel driver and remove the unneeded header file include statements. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_mip.h | 4 +- drivers/net/nfp/nfpco

[PATCH v4 11/26] net/nfp: refact the hwinfo module

Move the definition of data structure and macro into the implement file. Also remove the unneeded header file include statements. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_hwinfo.c | 84 +--- drivers/net/nfp/nfpcore/nfp_hwin

[PATCH v4 10/26] net/nfp: rename some parameter and variable

Rename some parameter and variable to make the logic easier to understand. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_cpp.h | 10 ++--- drivers/net/nfp/nfpcore/nfp_cpp_pcie_ops.c | 42 - drivers/net/nfp/nfpcore/nfp_cppcore.c

[PATCH v4 09/26] net/nfp: unify the guide line of header file

Unify the guide line of header file, we choose '__FOO_BAR_H__' style. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_cpp.h | 2 +- drivers/net/nfp/nfpcore/nfp_hwinfo.h | 2 +- drivers/net/nfp/nfpcore/nfp_mip.h | 2 +- drivers/net/nfp/nfpcore

[PATCH v4 07/26] net/nfp: standard the comment style

Follow the DPDK coding style, use the kdoc comment style. Also move the comment of the functions to the implement file and add some comment to help understand logic. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_cpp.h | 504 ---

[PATCH v4 08/26] net/nfp: standard the blank character

Use space character to align instead of TAB character. There should one blank line to split the block of logic, no more no less. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp6000/nfp6000.h | 4 +- drivers/net/nfp/nfpcore/nfp_cpp.h | 18 +--

[PATCH v4 06/26] net/nfp: adjust the log statement

Add log statement to the important control logic, which means more strict check of the return value of function call. Also remove some verbose info log statement. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_cpp_pcie_ops.c | 37 ++

[PATCH v4 05/26] net/nfp: standard the local variable coding style

There should only declare one local variable in each line, and the local variable should be arranged from short to long in the function. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_cpp_pcie_ops.c | 58 -- drivers/net/nfp/nfpcore/n

[PATCH v4 04/26] net/nfp: remove the unneeded logic

Remove the unneeded logic. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_cpp.h | 107 drivers/net/nfp/nfpcore/nfp_cpp_pcie_ops.c | 12 - drivers/net/nfp/nfpcore/nfp_cppcore.c | 162 -- drivers/net/nfp/nfpcore/nfp_mutex.c

[PATCH v4 03/26] net/nfp: unify the type of integer variable

Unify the type of integer variable to the DPDK prefer style. Also change the return type of 'nfp_eth_speed2rate()' to the corresponding enum, which make it more readable. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_cpp.h | 53 ++-

[PATCH v4 02/26] net/nfp: unify the indent coding style

Each parameter of function should occupy one line, and indent two TAB character. All the statement which span multi line should indent two TAB character. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_cpp.h | 80 +- drivers/net/nfp/nfp

[PATCH v4 00/26] refact the nfpcore module

This patch series aims to: - Make the coding style satisfy with DPDK. - Sync the logic with kernel driver. - Make the sub-module more modular. - Extend the nfp configure BAR from 8 to 24. --- v4: * Drop the commit 'using the DPDK memory management API'. * Modify the commit message of 'standard the

[PATCH v4 01/26] net/nfp: explicitly compare to null and 0

To compliance with the coding standard, make the pointer variable explicitly comparing to 'NULL' and the integer variable explicitly comparing to '0'. Signed-off-by: Chaoyong He Reviewed-by: Niklas Söderlund --- drivers/net/nfp/nfpcore/nfp_cpp_pcie_ops.c | 12 ++-- drivers/net/nfp/nfpco

Re: [PATCH v5 09/32] net/sssnic/base: add control queue

On Mon, 4 Sep 2023 12:56:35 +0800 wrote: > rte_memcpy(cmd->response_data, rte_pktmbuf_mtod(mbuf, void *), > + cmd->response_len); For small sizes, regular memcpy is going to be the same or faster than rte_memcpy

Re: [PATCH v5 08/32] net/sssnic/base: add work queue

On Mon, 4 Sep 2023 12:56:34 +0800 wrote: > +void * > +sssnic_workq_consume(struct sssnic_workq *workq, uint16_t num_entries, > + uint16_t *ci) > +{ > + void *e; > + uint16_t current_ci; > + > + if (workq->idle_entries + num_entries > workq->num_entries) > + return NULL

Re: [PATCH v5 07/32] net/sssnic/base: add mailbox support

On Mon, 4 Sep 2023 12:56:33 +0800 wrote: > +struct sssnic_mbox_send_ctrl1_reg { > + union { > + uint32_t u32; > + struct { > + uint32_t resvd0 : 10; > + /* Destination eventq in the mgmt cpu */ > + uint32_t ds

Re: [PATCH v5 06/32] net/sssnic/base: add message definition and utility

On Mon, 4 Sep 2023 12:56:32 +0800 wrote: > +static int > +sssnic_msg_buf_alloc(struct sssnic_msg *msg, size_t size) > +{ > + msg->data_buf = rte_zmalloc("sssnic_msg_data", size, 1); Putting message buffer on same NUMA node as device might help performance on NUMA architecture.

Re: [PATCH v5 04/32] net/sssnic: initialize hardware base

On Mon, 4 Sep 2023 12:56:30 +0800 wrote: > static int > sssnic_ethdev_init(struct rte_eth_dev *ethdev) > { > - RTE_SET_USED(ethdev); > + int ret; > + struct sssnic_hw *hw; > + struct sssnic_netdev *netdev; > + struct rte_pci_device *pci_dev; > + > PMD_INIT_FUNC_TRACE(

RE: [PATCH] raw/ntb: add support for 5th and 6th Gen Intel Xeon

> -Original Message- > From: Wu, Jingjing > Sent: Friday, September 15, 2023 17:14 > To: Guo, Junfeng > Cc: dev@dpdk.org > Subject: RE: [PATCH] raw/ntb: add support for 5th and 6th Gen Intel Xeon > > > > > -Original Message- > > From: Guo, Junfeng > > Sent: Friday, September

Re: [PATCH v3 02/27] net/nfp: unify the indent coding style

On Mon, 18 Sep 2023 01:25:28 + Chaoyong He wrote: > > From: Ferruh Yigit > > Sent: Friday, September 15, 2023 9:41 PM > > To: Chaoyong He ; dev@dpdk.org > > Cc: oss-drivers ; Niklas Soderlund > > > > Subject: Re: [PATCH v3 02/27] net/nfp: unify the indent coding style > > > > On 9/15/2023

[PATCH v4 18/18] common/idpf/base: update version

Update README Signed-off-by: Simei Su Acked-by: Beilei Xing --- drivers/common/idpf/base/README | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/common/idpf/base/README b/drivers/common/idpf/base/README index 693049c057..ff26f736ec 100644 --- a/drivers/common/idpf/bas

[PATCH v4 17/18] common/idpf/base: remove unused Tx descriptor types

Remove the unused TX descriptor types and mark them as reserved. Signed-off-by: Pavan Kumar Linga Signed-off-by: Simei Su Acked-by: Beilei Xing --- drivers/common/idpf/base/idpf_lan_txrx.h | 132 ++- 1 file changed, 10 insertions(+), 122 deletions(-) diff --git a/drivers/c

[PATCH v4 15/18] common/idpf/base: use 'type functionname(args)' style

Instead of splitting the function name and function type into multiple lines, use then in a single line. Signed-off-by: Pavan Kumar Linga Signed-off-by: Simei Su Acked-by: Beilei Xing --- drivers/common/idpf/base/idpf_controlq.c | 5 ++--- drivers/common/idpf/base/idpf_controlq_setup.c |

[PATCH v4 16/18] common/idpf/base: don't declare union with 'flex'

In idpf_flex_tx_desc structure, instead of naming the union with 'flex', use no name union as the union name is not really necessary there. This reduces the level of indirection in the hotpath. Signed-off-by: Pavan Kumar Linga Signed-off-by: Simei Su Acked-by: Beilei Xing --- drivers/common/id

[PATCH v4 14/18] common/idpf/base: use GENMASK macro

Instead of using a custom defined macro for generating a mask, use the standard GENMASK macro. Signed-off-by: Pavan Kumar Linga Signed-off-by: Simei Su Acked-by: Beilei Xing --- drivers/common/idpf/base/idpf_lan_pf_regs.h | 26 ++--- drivers/common/idpf/base/idpf_lan_txrx.h| 116 +

[PATCH v4 12/18] common/idpf/base: refactor descriptor 'ret val' stripping

Conditional check is not necessary to strip and get status bits from the descriptor. Signed-off-by: Pavan Kumar Linga Signed-off-by: Simei Su Acked-by: Beilei Xing --- drivers/common/idpf/base/idpf_controlq.c | 7 ++- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/drivers/co

[PATCH v4 13/18] common/idpf/base: refine comments and alignment

Refine the macros and definitions by using 'tab' spaces and new lines wherever necessary. Also refine the comment in 'idpf_ctlq_setup_regs' and remove the TODO comment in idpf_rss_hash enum as it doesn't make any sense. Signed-off-by: Pavan Kumar Linga Signed-off-by: Simei Su Acked-by: Beilei Xi

[PATCH v4 11/18] common/idpf/base: use 'void' return type

As idpf_ctlq_deinit always returns success, make it 'void' instead of returning only success. This also changes the return type for idpf_deinit_hw as 'void'. Based on the upstream comments, explicit __le16 typecasting is not necessary as CPU_TO_LE16 is already being used. Signed-off-by: Pavan Kum

[PATCH v4 10/18] common/idpf/base: use local pointer before updating 'CQ out'

Instead of updating directly to 'cq_out' double pointer, use a local pointer and update only when we return success. Signed-off-by: Pavan Kumar Linga Signed-off-by: Simei Su Acked-by: Beilei Xing --- drivers/common/idpf/base/idpf_controlq.c | 43 +--- 1 file changed, 23 ins

[PATCH v4 09/18] common/idpf/base: define non-flexible size structure for ADI

Customer has a requirement to use the legacy fixed size, single chunk structure for ADI creation - one chunk for queue and one chunk for vector. This is described in detail in customer case https://issuetracker.google.com/issues/270157802. On the other hand, upstream code review patch has been pos

[PATCH v4 08/18] common/idpf/base: add union for SW cookie fields in ctlq msg

Instead of using something like a byte offset, we can add a union to the struct here to enable direct addressing. Signed-off-by: Alan Brady Signed-off-by: Simei Su Acked-by: Beilei Xing --- .mailmap | 1 + drivers/common/idpf/base/idpf_controlq_api.h | 5 +++

[PATCH v4 07/18] common/idpf/base: add necessary check

Add necessary check for payload and message buffer. Signed-off-by: Julianx Grajkowski Signed-off-by: Simei Su Acked-by: Beilei Xing --- .mailmap | 1 + drivers/common/idpf/base/idpf_common.c | 6 -- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git

[PATCH v4 06/18] common/idpf/base: add some adi specific fields

a) Add maximum ADI count in capabilities message b) Add PF side ADI index to create_adi message c) Define another constant to indicate 'Function active' state of ADI Signed-off-by: Shailendra Bhatnagar Signed-off-by: Simei Su Acked-by: Beilei Xing --- .mailmap | 1 +

[PATCH v4 05/18] common/idpf/base: remove mailbox registers

Removing mailbox register offsets as the mapping to device register offsets are different between CVL and MEV (they are swapped out) individual drivers will define the offsets based on how registers are hardware addressed. However the it will begin with VDEV_MBX_START offset. Signed-off-by: Madhu

[PATCH v4 04/18] common/idpf/base: initial PTP support

Adding a few PTP capabilities to determine which PTP features are enabled - legacy cross time, ptm, device clock control, PTP Tx timestamp with direct registers access, PTP Tx timestamp using virtchnl messages. Creating structures and opcodes to support feautres introduced by capabilities. Signed

[PATCH v4 03/18] common/idpf/base: add miss completion capabilities

Add miss completion tag to other capabilities list, to indicate support for detecting a miss completion based on the upper bit of the completion tag. Signed-off-by: Josh Hay Signed-off-by: Simei Su Acked-by: Beilei Xing --- .mailmap | 1 + drivers/common/idpf/base/v

[PATCH v4 02/18] common/idpf/base: enable support for physical port stats

Add support to indicate physical port representor and query its statistics. Signed-off-by: Zhenning Xiao Signed-off-by: Jayaprakash Shanmugam Signed-off-by: Simei Su Acked-by: Beilei Xing --- .mailmap | 2 + drivers/common/idpf/base/virtchnl2.h | 80 ++

[PATCH v4 01/18] common/idpf: refactor single queue Tx data path

Currently, single queue Tx data path uses flex Tx data descriptor(DTYPE3) which is removed in the latest idpf spec. This patch replaces flex Tx data descriptor with base Tx data descriptor for single queue Tx data path and refines Tx single queue setup to align with Tx data path. Signed-off-by: Si

[PATCH v4 00/18] update idpf base code

Currently, single queue Tx data path uses flex Tx data descriptor(DTYPE3) which is removed in the latest idpf spec. [PATCH v4 01/18] replaces flex Tx data descriptor with base Tx data descriptor for single queue Tx data path and refines Tx single queue setup to align with Tx data path. [PATCH v4 0

RE: [PATCH v3 07/27] net/nfp: standard the comment style

> On 9/15/2023 10:15 AM, Chaoyong He wrote: > > Follow the DPDK coding style, use the kdoc comment style. > > Also add some comment to help understand logic. > > > > Signed-off-by: Chaoyong He > > Reviewed-by: Niklas Söderlund > > <...> > > > > > -/* > > - * Set the model id > > - * > > - * @pa

[RFC v1 2/2] vhost: add reconnection support to VDUSE (WIP)

From: Maxime Coquelin this patch is changed from https://gitlab.com/mcoquelin/dpdk-next-virtio/-/commit/a89dc311f2d03e99b8180f377b4a60a0e94 the biggest change is moving the mmaping-related process to the Previous patch Signed-off-by: Cindy Lu --- lib/vhost/vduse.c | 43 ++

[RFC v1 1/2] vduse: add mapping process in vduse create and destroy

the change in the creation process is add 1> check if we need to do reconnect 2> Use ioctl get the reconnect info (size and max page number) from kernel 3> mapping 1 page for reconnect  status + vq_numbers pages for every vqs, The change in the destroy process is the add the related unmap process

[RFC v1 0/2] Vduse: Add support for reconncect

Hi All this patch is add support for reconnection This patch changes based on Maxime's patch https://gitlab.com/mcoquelin/dpdk-next-virtio/-/commit/a89dc311f2d03e99b8180f377b4a60a0e94 The biggest change is sync the information with kernel by mmap the kernel is based on https://gitlab.com/lulu6/vho

RE: [PATCH v3 12/27] net/nfp: refact the hwinfo module

> -Original Message- > From: Ferruh Yigit > Sent: Friday, September 15, 2023 9:46 PM > To: Chaoyong He ; dev@dpdk.org > Cc: oss-drivers ; Niklas Soderlund > > Subject: Re: [PATCH v3 12/27] net/nfp: refact the hwinfo module > > On 9/15/2023 10:15 AM, Chaoyong He wrote: > > Move the defi

Please help to review sssnic PMD patches

Hi, I sent sssnic PMD patches 2 weeks ago, the CI checks are all successful,  but so far no body reviewed them. I really hope sssnic PMD will be merged into release 23.11. What should I do next? The following is URL of sssnic PMD patches: DPDK - Patchwork

RE: [PATCH v3 08/27] net/nfp: using the DPDK memory management API

> On 9/15/2023 10:15 AM, Chaoyong He wrote: > > Revise the logic, using the DPDK memory management API to replace the > > malloc()/free(). > > > > DPDK memory management APIs are required for datapath, where it matters > from which socket it is allocated, it is accessed a lot (datapath), or it is

RE: [PATCH v3 07/27] net/nfp: standard the comment style

> On 9/15/2023 10:15 AM, Chaoyong He wrote: > > Follow the DPDK coding style, use the kdoc comment style. > > Also add some comment to help understand logic. > > > > Signed-off-by: Chaoyong He > > Reviewed-by: Niklas Söderlund > > <...> > > > > > -/* > > - * Set the model id > > - * > > - * @p

RE: [PATCH v3 03/27] net/nfp: unify the type of integer variable

> On 9/15/2023 10:15 AM, Chaoyong He wrote: > > Unify the type of integer variable to the DPDK prefer style. > > Also change the return type of 'nfp_eth_speed2rate()' to the > > corresponding enum, which make it more readable. > > > > Signed-off-by: Chaoyong He > > Reviewed-by: Niklas Söderlund >

RE: [PATCH v3 02/27] net/nfp: unify the indent coding style

> From: Ferruh Yigit > Sent: Friday, September 15, 2023 9:41 PM > To: Chaoyong He ; dev@dpdk.org > Cc: oss-drivers ; Niklas Soderlund > > Subject: Re: [PATCH v3 02/27] net/nfp: unify the indent coding style > > On 9/15/2023 10:15 AM, Chaoyong He wrote: > > Each parameter of function should occup

Re: [PATCH v1] examples/l3fwd: relax the RSS/Offload requirement

03/09/2023 05:01, Trevor Tao пишет: Now the port Rx mq_mode had been set to RTE_ETH_MQ_RX_RSS, and offload mode set to RTE_ETH_RX_OFFLOAD_CHECKSUM by default, but some hardware and/or virtual interface does not support the RSS and offload mode presupposed, e.g., some virtio interfaces in the clou

RE: [PATCH v3 1/3] lib: introduce dispatcher library

> -Original Message- > From: Mattias Rönnblom > Sent: Monday, September 4, 2023 6:33 PM > To: dev@dpdk.org > Cc: Jerin Jacob ; techbo...@dpdk.org; Van Haaren, > Harry ; hof...@lysator.liu.se; Nilsson, Peter > ; Heng Wang ; > Naga Harish K, S V ; Pavan Nikhilesh > ; Gujjar, Abhinandan S >

RE: [EXT] [RFC] cryptodev: refactor sm2, add plain message flag

> -Original Message- > From: Gowrishankar Muthukrishnan > Sent: Monday, August 14, 2023 9:49 AM > To: Kusztal, ArkadiuszX ; dev@dpdk.org > Cc: Akhil Goyal ; Ji, Kai ; Power, Ciara > > Subject: RE: [EXT] [RFC] cryptodev: refactor sm2, add plain message flag > > Hi, > Please find my com

[PATCH] crypto/qat: add sm2 ecdsa

Added SM2 ECDSA feature to the Intel QuickAssist Technology symmetric crypto PMD. Signed-off-by: Arkadiusz Kusztal --- doc/guides/cryptodevs/features/qat.ini| 1 + doc/guides/cryptodevs/qat.rst | 1 + doc/guides/rel_notes/release_23_11.rst| 3 + .../common/qat/

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

Service specific functions were moved to services files. Signed-off-by: Arkadiusz Kusztal --- drivers/common/qat/dev/qat_dev_gen1.c| 1 + drivers/common/qat/dev/qat_dev_gen2.c| 1 + drivers/common/qat/dev/qat_dev_gen3.c| 2 + drivers/common/qat/dev/qat_dev_gen4.c

[PATCH 2/3] common/qat: move command line params to the device struct

Command line parameters was moved to the device struct. Signed-off-by: Arkadiusz Kusztal --- drivers/common/qat/qat_device.c | 54 ++--- drivers/common/qat/qat_device.h | 35 drivers/compress/qat/qat_comp_pmd.c | 9 +++ driver

[PATCH 1/3] common/qat: limit configuration to the primary process

This change prevents certain configuration functions from being called by the secondary process. Signed-off-by: Arkadiusz Kusztal --- drivers/common/qat/qat_device.c | 113 +++- drivers/common/qat/qat_device.h | 2 + 2 files changed, 66 insertions(+), 49 del

Re: [PATCH v2 18/29] ip_frag: mark a couple of functions stable

There were two functions added in 22.11 which were marked as experimental. Remove the experimental tag. Signed-off-by: Stephen Hemminger --- lib/ip_frag/rte_ip_frag.h | 2 -- lib/ip_frag/version.map | 9 ++--- 2 files changed, 2 insertions(+), 9 deletions(-) diff --git a/lib/ip_frag/rte_

Re: [PATCH v2 23/29] ipsec: remove experimental from SA API

These API's were added in 21.11, remove experimental flag. Signed-off-by: Stephen Hemminger --- lib/ipsec/rte_ipsec.h | 2 -- lib/ipsec/version.map | 9 + 2 files changed, 1 insertion(+), 10 deletions(-) diff --git a/lib/ipsec/rte_ipsec.h b/lib/ipsec/rte_ipsec.h index 04129926b69f..816

Re: [PATCH v2 01/29] bpf: make rte_bpf_dump and rte_bpf_convert stable API's

These two API's were introduced in 23.11 and can now be made not experimental. Signed-off-by: Stephen Hemminger --- lib/bpf/rte_bpf.h | 2 -- lib/bpf/version.map | 9 ++--- 2 files changed, 2 insertions(+), 9 deletions(-) diff --git a/lib/bpf/rte_bpf.h b/lib/bpf/rte_bpf.h index 4d71120db

Minutes of Technical Board Meeting, 2023-Sep-06

Minutes of Technical Board Meeting, 2023-Spetember-06 Members Attending - -Aaron -Bruce -Hemant -Jerin -Kevin -Konstantin (Chair) -Maxime -Stephen -Thomas NOTE: The technical board meetings every second Wednesday at https://meet.jit.si/DPDK at 3 pm UTC. Meetings are public,

Re: [PATCH v3 00/11] rework thread management

13/09/2023 12:28, Thomas Monjalon пишет: The main effect of this patch series is to remove calls to pthread functions except for pthread_cancel and locks. The function rte_thread_create_control() does not take thread attributes settings anymore as it looks a useless complication of the API. Then