[PATCH 7/7] net/mlx5: fix the error in VLAN actions creation

When a failure occurs during the VLAN actions creating, the value of "rte_errno" is already set by the mlx5dr_action_create*. The value can be returned directly to reflect the actual reason. In the meanwhile, the "rte_flow_error" structure should also be set with explict message. Fixes: 773ca0e91

[PATCH 6/7] net/mlx5: fix the error set for age pool initialization

The rte_flow_error needs to be set when the age pool initialization has a failure. Or else the application will crash due to the access of the invalid "message" field. Fixes: 04a4de756e14 ("net/mlx5: support flow age action with HWS") Cc: michae...@nvidia.com Signed-off-by: Bing Zhao Acked-by: O

[PATCH 5/7] net/mlx5: fix the error set in control tables create

When some failure occurrs in the flow_hw_create_ctrl_tables(), the "rte_flow_error" should be set properly with all needed information. Then the rte_errno and the "message" will reflect the actual failure. This will also solve the crash when trying to access the "message". Fixes: 1939eb6f660c ("n

[PATCH 4/7] net/mlx5: fix the return value of vport action

The "rte_flow_error" should be set in case a invalid pointer access to the message member. Fixes: 1939eb6f660c ("net/mlx5: support flow port action with HWS") Cc: dsosnow...@nvidia.com Signed-off-by: Bing Zhao Acked-by: Ori Kam --- drivers/net/mlx5/mlx5_flow_hw.c | 3 ++- 1 file changed, 2 ins

[PATCH 3/7] net/mlx5: fix the error set in Tx representor tagging

In the previous implementation, the error information was not set when there was a failure during the initialization. The pointer from the user should be passed to the called functions to be set properly before returning. Fixes: 483181f7b6dd ("net/mlx5: support device control of representor match

Re: [PATCH 1/7] net/mlx5: fix the modify field check of tag

On Fri, 30 Jun 2023 08:43:03 +0300 Bing Zhao wrote: > @@ -1117,9 +1117,10 @@ flow_dv_fetch_field(const uint8_t *data, uint32_t size) > static inline bool > flow_modify_field_support_tag_array(enum rte_flow_field_id field) > { > - switch (field) { > + switch ((int)field) { > case

[PATCH v2] net/iavf: fix duplicate reset done check with large VF

When starting with large vf, need to reset VF to request queues, the reset process will execute VIRTCHNL commands to clean up resource. VF reset done check and reset watchdog read the same global register, resulting in the NIC not responding to the VIRTCHNL command. This patch turns off the watch

RE: [PATCH] vfio: do not coalesce DMA mappings

Hi Nipun, Replies are inline. > -Original Message- > From: Nipun Gupta > Sent: Friday, June 30, 2023 9:46 AM > To: Ding, Xuan ; dev@dpdk.org; > tho...@monjalon.net; Burakov, Anatoly ; > ferruh.yi...@amd.com > Cc: nikhil.agar...@amd.com; He, Xingguang ; Ling, > WeiX > Subject: Re: [PATCH

[PATCH 2/7] net/mlx5: fix the error set in quota init

When there was a failure in the quota initialization, only the return value was set. The "rte_errno" and the "rte_flow_error" were not set. The application may get a crash when trying to access the "message" field. Fixes: 15896eaf9d71 ("net/mlx5: support indirect quota create/query/modify") Cc: ge

[PATCH 1/7] net/mlx5: fix the modify field check of tag

The new member "tag_index" was added into the structure "rte_flow_action_modify_data". It is the index of the header inside encapsulation level. Both the "RTE_FLOW_FIELD_TAG" and "MLX5_RTE_FLOW_FIELD_META_REG" types will use this member instead of the "level", but the "level" will still be supporte

Re: [RFC] eventdev: remove single-event enqueue operation

On Fri, Jun 9, 2023 at 11:18 PM Mattias Rönnblom wrote: > > Eliminate non-burst enqueue operation from Eventdev. > > The effect of this change is to reduce Eventdev code complexity > somewhat and slightly improve performance. > > The single-event enqueue shortcut provided a very minor performance

Re: [dpdk-dev] [PATCH] eal/linux: add operation LOCK_NB to flock()

On Thu, 15 Apr 2021 15:24:01 +0100 "Burakov, Anatoly" wrote: > On 25-Mar-21 8:21 AM, xiangxia.m@gmail.com wrote: > > From: Tonghao Zhang > > > > The hugepage of different size, 2MB, 1GB may be mounted on > > the same directory (e.g /dev/hugepages). Then dpdk > > primary process will be bloc

Re: [dpdk-dev] [RFC 0/4] SocketPair Broker support for vhost and virtio-user.

On Wed, 17 Mar 2021 21:25:26 +0100 Ilya Maximets wrote: > TL;DR; > Managing socket files is too much fun. :) And here is how this > could be improved: > https://github.com/igsilya/one-socket > https://github.com/igsilya/one-socket/blob/main/doc/socketpair-broker.rst > In particular

[PATCH] vhost: add notify reply ops to avoid message deadlock

Since backend and frontend message are synchronous in the same thread, there will be a probability of message deadlock. Consider each driver to determine whether to wait for response. Signed-off-by: Rma Ma --- lib/vhost/vdpa_driver.h | 3 +++ lib/vhost/vhost_user.c | 23 ++-

[PATCH] doc: fix missing release note for Intel 4th Gen Xeon Processors NTB support

Add missing release note for NTB support on 4th Gen Intel® Xeon® Processors Platform. Fixes: 834d99f388ca ("raw/ntb: add PPD status check for Sapphire Rapids") Signed-off-by: Junfeng Guo --- doc/guides/rel_notes/release_23_07.rst | 4 1 file changed, 4 insertions(+) diff --git a/doc/guide

RE: [EXT] Re: [PATCH v3] net/bnx2x: offer maintainership for bnx2x

> -Original Message- > From: Thomas Monjalon > Sent: 28 June 2023 04:54 > To: Julien Aube > Cc: dev@dpdk.org; Alok Prasad ; Akhil Goyal > ; Jerin Jacob Kollanukkaran > > Subject: [EXT] Re: [PATCH v3] net/bnx2x: offer maintainership for bnx2x > > External Email > >

Re: [PATCH] vfio: do not coalesce DMA mappings

Hi Xuan, Thanks for pointing out the issue and figuring out the patch which introduced this. If you have answers to below queries, please let me know: Is there any other test cases which tests "--no-huge" which pass? Also, if we change the "-m" option to provide lower memory, does the test p

[PATCH] doc: fix missing release note for NTB Sapphire Rapids support

Add missing release note for NTB support on Sapphire Rapids Platform. Fixes: 834d99f388ca ("raw/ntb: add PPD status check for Sapphire Rapids") Signed-off-by: Junfeng Guo --- doc/guides/rel_notes/release_23_07.rst | 4 1 file changed, 4 insertions(+) diff --git a/doc/guides/rel_notes/rele

RE: [dpdk-dev] [PATCH 0/9] Introduce vfio-user library

Hi Stephen, > -Original Message- > From: Stephen Hemminger > Sent: Friday, June 30, 2023 12:10 AM > To: Thanos Makatos > Cc: David Marchand ; Xia, Chenbo > ; dev ; Thomas Monjalon > ; Liang, Cunming ; Lu, > Xiuchun ; Li, Miao ; Wu, Jingjing > ; john.g.john...@oracle.com; Stefan Hajnoczi

Re: [PATCH v3 2/2] net/hns3: add FDIR VLAN match mode runtime config

Hi Ferruh On 2023/6/29 23:52, Ferruh Yigit wrote: On 6/29/2023 2:21 PM, Dongdong Liu wrote: diff --git a/doc/guides/nics/hns3.rst b/doc/guides/nics/hns3.rst index 5373ec5a8f..001ccbdb62 100644 --- a/doc/guides/nics/hns3.rst +++ b/doc/guides/nics/hns3.rst @@ -140,6 +140,32 @@ Runtime Configuratio

Re: [PATCH v2] eal: add notes to SMP memory barrier APIs

On 2023-06-26 09:12, Ruifeng Wang wrote: The rte_smp_xx() APIs are deprecated. But it is not mentioned in the function header. Added notes in function header for clarification. Signed-off-by: Ruifeng Wang --- v2: Made the notes more specific. lib/eal/include/generic/rte_atomic.h | 19 +++

Re: [PATCH] eal: add notes to SMP memory barrier APIs

On 2023-06-25 10:17, Ruifeng Wang wrote: -Original Message- From: Mattias Rönnblom Sent: Friday, June 23, 2023 2:20 AM To: Ruifeng Wang ; tho...@monjalon.net; david.march...@redhat.com Cc: dev@dpdk.org; konstantin.v.anan...@yandex.ru; Honnappa Nagarahalli ; nd Subject: Re: [PATCH] eal:

Re: [PATCH] Updated to dpdk20.11

Hi David, David Young writes: > --- Thanks for the contribution! Very happy to have someone working on the docs. Thanks for getting a patch to the mailing list. Just a few nits to address before the patch can get merged. Customarily, a change to each area will describe that in the subject -

[PATCH v8] hash: add XOR32 hash function

An XOR32 hash is needed in the Software Switch (SWX) Pipeline for its use case in P4. We implement it in this patch so it could be easily registered in the pipeline later. Signed-off-by: Bili Dong --- v8: * Removed unnecessary helper function. (re Vladimir Medvedkin ) v7: * Simplified byte orde

Re: [dpdk-dev] [PATCH v3 0/2] Add example l3fwd-regex

On Sun, 15 Nov 2020 15:07:37 +0100 Thomas Monjalon wrote: > 18/10/2020 14:03, g...@marvell.com: > > From: Guy Kaneti > > > > The L3 Forwarding with Regex application is a simple example of > > packet processing using DPDK Regex framework. > > The application performs L3 LPM based forwarding whi

Re: [dpdk-dev] [PATCH v6 1/3] kni: rework rte_kni_update_link using ioctl

On Mon, 30 Aug 2021 17:27:29 +0300 Igor Ryzhov wrote: > Current implementation doesn't allow us to update KNI carrier if the > interface is not yet UP in kernel. It means that we can't use it in the > same thread which is processing rte_kni_ops.config_network_if, which is > very convenient, becau

Re: [dpdk-dev] [PATCH] ppc64le: fix build with Clang

On Thu, 14 Oct 2021 19:35:37 +0200 Piotr Kubaj wrote: > Couple of quite common issues: > 1. In rte_cycles.h, sys/platform/ppc.h and __ppc_get_timebase() are only for > glibc. > 2. In include/rte_atomic.h, the same issue already was encountered on arm64: > https://mails.dpdk.org/archives/dev/2019

[PATCH] net/mlx5: share QUOTA indirect action

Support shared indirect QUOTA action. flow configure 1 queues_number 4 queues_size 64 quotas_number 32 flow configure 0 queues_number 4 queues_size 64 quotas_number 0 \ host_port 1 flags 2 flow indirect_action 1 create ingress action_id 8 \ action quota_create limit 18000 mode l3 / end

Re: [dpdk-dev] [PATCH] net: announce changes in TCP header

On Fri, 6 Aug 2021 13:56:56 + Gregory Etelson wrote: > Hello, > > The TCP flags bitfield patch complements similar changes > proposed to IPv4 header in net: announce changes in IPv4 header - Patchwork > (dpdk.org)

Re: [dpdk-dev] [PATCH] lib: fix socket id type

On Tue, 27 Apr 2021 17:36:32 +0800 "Min Hu (Connor)" wrote: > From: Chengchang Tang > > The variable type of the current socket ID is uint8_t. When traversing all > nodes, the socket ID is compared with RTE_MAX_NUMA_NODES. Since > RTE_MAX_NUMA_NODES has not been verified, it may be larger than

Re: [dpdk-dev] [PATCH] test: make hugepage check more robust under Linux

On Fri, 19 Mar 2021 10:34:02 -0400 Aaron Conole wrote: > > The check is evaluated first when configuring as normal user and not > > reevaluated when running the tests as root in Travis/GHA, so the tests > > are started with no hugepage. > > > > Then, in Travis and GHA, we can see (sigbus o_O) cra

Re: [dpdk-dev] [PATCH v5 0/4] improve options help

On Mon, 5 Apr 2021 21:39:50 +0200 Thomas Monjalon wrote: > After v4, this series is split in several parts. > The remaining 4 patches of this series are low priority. > > Patches 1 and 3 are simple improvements. > > Patches 2 and 4 lead to a new formatting of the usage text. > It is a matter o

Re: [dpdk-dev] [PATCH 07/11] eal: add log level help

On Wed, 10 Mar 2021 00:31:11 +0100 Thomas Monjalon wrote: > The option --log-level was not completely described in the usage text, > and it was difficult to guess the names of the log types and levels. > > A new value "help" is accepted after --log-level to give more details > about the syntax a

Re: [PATCH] ethdev: fix Tx queue mask endianness

On 6/29/2023 4:42 PM, Thomas Monjalon wrote: > 29/06/2023 17:40, David Marchand: >> On Thu, Jun 29, 2023 at 5:31 PM Thomas Monjalon wrote: >>> 29/06/2023 15:58, David Marchand: - .tx_queue = RTE_BE16(0x), + .tx_queue = 0x, >>> >>> As I said in an earlier comment about the

Re: [dpdk-dev] [PATCH 0/9] Introduce vfio-user library

On Fri, 18 Dec 2020 14:07:20 + Thanos Makatos wrote: > > Hello, > > > > On Fri, Dec 18, 2020 at 8:54 AM Chenbo Xia wrote: > > > *librte_vfio_user* library is an implementation of VFIO-over-socket[1] > > (also > > > known as vfio-user) which is a protocol that allows a device to be >

[PATCH v6 6/6] eal: add option to put timestamp on console output

When debugging driver or startup issues, it is useful to have a timestamp on each message printed. The messages in syslog already have a timestamp, but often syslog is not available during testing. The timestamp format is chosen to look like the default Linux dmesg timestamp. The first few lines a

[PATCH v6 5/6] eal: allow user to set default log stream before init

It is useful for application to be able to set the default log stream before call rte_eal_init(). This makes all messages go to the new default. For example, to skip using syslog; just doing rte_openlog_stream(stderr); There is no reason for helper command line applications to clutter sys

[PATCH v6 4/6] eal: skip stdio on console logging

There is no need to use stdio when logging to console. Using the write system call directly avoids unnecessary copy to stdio output buffer. Signed-off-by: Stephen Hemminger --- lib/eal/unix/eal_log.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/lib/eal/unix/eal_log.c b

[PATCH v6 3/6] eal: fix help message for syslog option

The --syslog flag takes facility argument. Signed-off-by: Stephen Hemminger --- lib/eal/common/eal_common_options.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/eal/common/eal_common_options.c b/lib/eal/common/eal_common_options.c index 03059336987d..005da4d12001 1006

[PATCH v6 2/6] eal: turn off getopt_long error message during eal_log_level

If DPDK application is given a bogus option, the error message would get printed twice. Once during scan for log level and again during parsing of arguments. Example: # ./build/app/dpdk-testpmd --bogus ./build/app/dpdk-testpmd: unrecognized option '--bogus' EAL: Detected CP

[PATCH v6 1/6] eal: unify logging code

FreeBSD logging code was not using syslog and did not have the same options as Linux. Use a common set of functions for that. Pre-parsing for log level is common to all OS's. Now the pre-scan can return an error. A bad argument give to --log-level option was given the code would keep going. Sign

[PATCH v6 0/6] Logging related patches

Patchset that includes: - unified code for more of log argument handling - fix for duplicate option errors - timestamp option for logging v6 - fix build on freebsd and windows - add test for log timestamp Stephen Hemminger (6): eal: unify logging code eal: turn off getopt_long error

Re: [PATCH v3 0/2] net/hns3: add FDIR VLAN match mode runtime config

On 6/29/2023 2:21 PM, Dongdong Liu wrote: > This patchset contains two patches: > The first one is to delete duplicate macro definition for hns3. > The second one is to add FDIR VLAN match mode runtime config for hns3 > > v2->v3: > - Use double backquotes for values of the devargs and fix spellin

Re: [PATCH v3 2/2] net/hns3: add FDIR VLAN match mode runtime config

On 6/29/2023 2:21 PM, Dongdong Liu wrote: > diff --git a/doc/guides/nics/hns3.rst b/doc/guides/nics/hns3.rst > index 5373ec5a8f..001ccbdb62 100644 > --- a/doc/guides/nics/hns3.rst > +++ b/doc/guides/nics/hns3.rst > @@ -140,6 +140,32 @@ Runtime Configuration > For example:: > -a :7d:00.0

Re: [PATCH] net/netvsc: fix bogus sizeof calculation

On Wed, Jun 28, 2023 at 05:29:57PM -0700, lon...@linuxonhyperv.com wrote: > From: Stephen Hemminger > > Found by cppcheck. > > drivers/net/netvsc/hn_rndis.c:332:21: warning: Found calculation inside > sizeof(). [sizeofCalculation] if (len < sizeof(3 * sizeof(uint32_t))) { > > Fixes: 4e9c73e96

Re: [PATCH] ethdev: fix Tx queue mask endianness

29/06/2023 17:40, David Marchand: > On Thu, Jun 29, 2023 at 5:31 PM Thomas Monjalon wrote: > > 29/06/2023 15:58, David Marchand: > > > - .tx_queue = RTE_BE16(0x), > > > + .tx_queue = 0x, > > > > As I said in an earlier comment about the same issue, > > UINT16_MAX would be better. >

Re: [PATCH] ethdev: fix Tx queue mask endianness

On Thu, Jun 29, 2023 at 5:31 PM Thomas Monjalon wrote: > > 29/06/2023 15:58, David Marchand: > > Even if harmless, this endianness tag is incorrect as the tx_queue field > > is declared as a host integer. > > Additionally, this breaks OVS compilation with sparse. > > > > Fixes: 41f6bdc7615a ("ethd

Re: [PATCH v2] ethdev: add Tx queue flow matching item

08/06/2023 21:59, Thomas Monjalon: > 08/05/2023 15:49, kirankum...@marvell.com: > > From: Kiran Kumar K > > > > Adding support for Tx queue flow matching item. > > This item is valid only for egress rules. > > An example use case would be that application can > > set different vlan insert rules w

Re: [PATCH] ethdev: fix Tx queue mask endianness

29/06/2023 15:58, David Marchand: > Even if harmless, this endianness tag is incorrect as the tx_queue field > is declared as a host integer. > Additionally, this breaks OVS compilation with sparse. > > Fixes: 41f6bdc7615a ("ethdev: add Tx queue flow matching item") > > Signed-off-by: David March

Re: [PATCH] ethdev: fix Tx queue mask endianness

On 6/29/2023 2:58 PM, David Marchand wrote: > Even if harmless, this endianness tag is incorrect as the tx_queue field > is declared as a host integer. > Additionally, this breaks OVS compilation with sparse. > > Fixes: 41f6bdc7615a ("ethdev: add Tx queue flow matching item") > > Signed-off-by: D

Re: [PATCH] net/txgbe: fix blocking system events

On 6/29/2023 11:07 AM, Jiawen Wu wrote: > Refer to commit 819d0d1d57f1 ("net/ixgbe: fix blocking system events"). > Fix the same issue as ixgbe. > Can you please describe the problem in this commit log, it is OK to keep reference to the ixgbe too, but better to describe problem here. > Replace r

[PATCH v11] app/dma-perf: introduce dma-perf application

There are many high-performance DMA devices supported in DPDK now, and these DMA devices can also be integrated into other modules of DPDK as accelerators, such as Vhost. Before integrating DMA into applications, developers need to know the performance of these DMA devices in various scenarios and

RE: [PATCH v10] app/dma-perf: introduce dma-perf application

> -Original Message- > From: Thomas Monjalon > Sent: Thursday, June 29, 2023 9:19 PM > To: Hu, Jiayu ; Wang, YuanX ; > Jiang, Cheng1 > Cc: Richardson, Bruce ; > m...@smartsharesystems.com; Xia, Chenbo ; > amitpraka...@marvell.com; ano...@marvell.com; > huangdeng...@huawei.com; Laatz, K

[PATCH] ethdev: fix Tx queue mask endianness

Even if harmless, this endianness tag is incorrect as the tx_queue field is declared as a host integer. Additionally, this breaks OVS compilation with sparse. Fixes: 41f6bdc7615a ("ethdev: add Tx queue flow matching item") Signed-off-by: David Marchand --- lib/ethdev/rte_flow.h | 2 +- 1 file c

[PATCH] cryptodev: fix device socket ID type

The socket ID field for a cryptodev device data was unsigned int. Due to recent changes to the default device socket ID, this caused an issue when the socket ID was unknown and set to -1. The device socket ID wrapped around to 255, and caused errors when allocating memory. Changing this field type

Re: [PATCH v2 2/2] net/hns3: add FDIR VLAN match mode runtime config

On 2023/6/29 19:08, Ferruh Yigit wrote: On 6/28/2023 3:59 AM, Dongdong Liu wrote: Hi Thomas Many thanks for your review. On 2023/6/28 7:12, Thomas Monjalon wrote: 27/06/2023 09:11, Dongdong Liu: +- ``fdir_vlan_match_mode`` (default `strict`) + + Used to select VLAN match mode. This runti

Re: [PATCH] test: add additional tests for thash library

On Thu, Jun 29, 2023 at 3:44 PM Medvedkin, Vladimir wrote: > > Hi David, > > Yes, I sent them separately because the bugfix in mentioned patch should be > backported unlike adding a new test. Well, having the two patches in a single series won't prevent us from only backporting the fix, and the

RE: [PATCH] test: add additional tests for thash library

Hi David, Yes, I sent them separately because the bugfix in mentioned patch should be backported unlike adding a new test. -Original Message- From: David Marchand Sent: Thursday, June 29, 2023 2:00 PM To: Medvedkin, Vladimir Cc: dev@dpdk.org; konstantin.v.anan...@yandex.ru Subject: Re

[PATCH v3 2/2] net/hns3: add FDIR VLAN match mode runtime config

From: Huisong Li The VLAN number in FDIR meta data is used to enable that hardware bases on VLAN number to strictly match the input flow. And it is enabled by default. For the following two rules: rule0: pattern: eth type is 0x0806 actions: queue index 3 rule1: pattern: eth typ

[PATCH v3 0/2] net/hns3: add FDIR VLAN match mode runtime config

This patchset contains two patches: The first one is to delete duplicate macro definition for hns3. The second one is to add FDIR VLAN match mode runtime config for hns3 v2->v3: - Use double backquotes for values of the devargs and fix spelling errors. v1->v2: - Fix the hns3.rst doc build warini

[PATCH v3 1/2] net/hns3: delete duplicate macro definition

From: Huisong Li This patch delete some duplicate macro definitions. Fixes: a4c7152d0581 ("net/hns3: extract common code to its own file") Cc: sta...@dpdk.org Signed-off-by: Huisong Li Signed-off-by: Dongdong Liu --- drivers/net/hns3/hns3_ethdev.h | 7 --- 1 file changed, 7 deletions(-)

Re: [PATCH v10] app/dma-perf: introduce dma-perf application

29/06/2023 14:50, Jiang, Cheng1: > From: Thomas Monjalon > > 28/06/2023 03:20, Cheng Jiang: > > > There are many high-performance DMA devices supported in DPDK now, > > and > > > these DMA devices can also be integrated into other modules of DPDK as > > > accelerators, such as Vhost. Before integr

Re: [PATCH] doc: ensure sphinx output is reproducible

On Thu, Jun 29, 2023 at 2:58 PM wrote: > > From: Christian Ehrhardt > > By adding -j we build in parallel, to make building on multiprocessor > machines more effective. While that works it does also break > reproducible builds as the order of the sphinx generated searchindex.js > is depending on

Re: [PATCH] test: add additional tests for thash library

On Wed, Jun 28, 2023 at 9:12 PM Vladimir Medvedkin wrote: > > Adds tests comparing the results of applying the output > of rte_thash_get_complement() to the tuple with the result > of calling rte_thash_adjust_tuple(). > > Suggested-by: Konstantin Ananyev > Signed-off-by: Konstantin Ananyev > Sig

[PATCH] doc: ensure sphinx output is reproducible

From: Christian Ehrhardt By adding -j we build in parallel, to make building on multiprocessor machines more effective. While that works it does also break reproducible builds as the order of the sphinx generated searchindex.js is depending on execution speed of the individual processes. Signed-

Re: [PATCH v6 1/2] build: select deprecated libraries

On Thu, Jun 29, 2023 at 2:48 PM Aaron Conole wrote: > > David Marchand writes: > > > Rework deprecated libraries selection by introducing a new configuration > > option. > > > > This breaks existing configurations that were relying on disable_libs='' > > for enabling deprecated libraries. > > On

RE: [PATCH v10] app/dma-perf: introduce dma-perf application

Hi Thomas, > -Original Message- > From: Thomas Monjalon > Sent: Thursday, June 29, 2023 5:38 PM > To: Jiang, Cheng1 > Cc: Richardson, Bruce ; > m...@smartsharesystems.com; Xia, Chenbo ; > amitpraka...@marvell.com; ano...@marvell.com; > huangdeng...@huawei.com; Laatz, Kevin ; > fengcheng.

RE: [PATCH v10] app/dma-perf: introduce dma-perf application

Hi Thomas, > -Original Message- > From: Thomas Monjalon > Sent: Thursday, June 29, 2023 5:09 PM > To: Hu, Jiayu ; Wang, YuanX ; > Jiang, Cheng1 > Cc: Richardson, Bruce ; > m...@smartsharesystems.com; Xia, Chenbo ; > amitpraka...@marvell.com; ano...@marvell.com; > huangdeng...@huawei.com;

Re: [PATCH v6 1/2] build: select deprecated libraries

David Marchand writes: > Rework deprecated libraries selection by introducing a new configuration > option. > > This breaks existing configurations that were relying on disable_libs='' > for enabling deprecated libraries. > On the other hand, it will make enabling optional libraries more > straig

Re: [PATCH v5 1/2] build: select deprecated libraries

David Marchand writes: > Rework deprecated libraries selection by introducing a new configuration > option. > > This breaks existing configurations that were relying on disable_libs='' > for enabling deprecated libraries. > On the other hand, it will make enabling optional libraries more > straig

Re: [PATCH] app/testpmd: fix 32-bit system forwarding failure

On 6/29/2023 11:09 AM, Shiyang He wrote: > l2_len is not calculated correctly on 32-bit systems, resulting in > packet forwarding failure. > > This patch fixes this issue by changing the l2_len calculation. > > Fixes: dd827fa42a30 ("app/testpmd: fix GTP L2 length in checksum engine") > Cc: sta...

Re: [PATCH] hash: fix reading unaligned bits implementation

Fixes: 28ebff11c2dc ("hash: add predictable RSS") Cc: sta...@dpdk.org Signed-off-by: Vladimir Medvedkin --- lib/hash/rte_thash.c | 12 +++- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/lib/hash/rte_thash.c b/lib/hash/rte_thash.c index 0249883b8d..2228af576b 100644

Re: [PATCH] net/netvsc: fix bogus sizeof calculation

On 6/29/2023 1:29 AM, lon...@linuxonhyperv.com wrote: > From: Stephen Hemminger > > Found by cppcheck. > > drivers/net/netvsc/hn_rndis.c:332:21: warning: Found calculation inside > sizeof(). [sizeofCalculation] if (len < sizeof(3 * sizeof(uint32_t))) { > > Fixes: 4e9c73e96e83 ("net/netvsc: ad

Re: [PATCH v2 2/2] net/hns3: add FDIR VLAN match mode runtime config

On 6/28/2023 3:59 AM, Dongdong Liu wrote: > Hi Thomas > > Many thanks for your review. > > On 2023/6/28 7:12, Thomas Monjalon wrote: >> 27/06/2023 09:11, Dongdong Liu: >>> +- ``fdir_vlan_match_mode`` (default `strict`) >>> + >>> +  Used to select VLAN match mode. This runtime config can be `stric

DPDK Release Status Meeting 2023-06-29

Release status meeting minutes 2023-06-29 = Agenda: * Release Dates * Subtrees * Roadmaps * LTS * Defects * Opens Participants: * AMD * Intel * Nvidia * Red Hat Release Dates - The following are the proposed current dates for 23.07: * V1:

[PATCH] net/mlx5: fix flow dump for modify field

When using other process to dump flow information, the modify field information cannot be dumpped due to incorrect lcore value. Add loop to get modify field information from all lcore to fix the issue. Fixes: a7ac7fae4919 ("net/mlx5: enhance flow dump") Cc: sta...@dpdk.org Signed-off-by: Sean Zha

[PATCH] net/txgbe: fix blocking system events

Refer to commit 819d0d1d57f1 ("net/ixgbe: fix blocking system events"). Fix the same issue as ixgbe. Replace rte alarm thread with an independent thread. Fixes: 0c061eadec59 ("net/txgbe: add link status change") Cc: sta...@dpdk.org Signed-off-by: Jiawen Wu --- drivers/net/txgbe/txgbe_ethdev.c

Re: [PATCH v4 2/9] event/dlb2: skip configuration if no eventdev lib

On Wed, Jun 28, 2023 at 12:19 PM David Marchand wrote: > > On Fri, Jun 23, 2023 at 5:07 PM Bruce Richardson > wrote: > > > > While the build system will skip building most libs and drivers when a > > dependency is missing for a component, for DLB2 driver, the > > "static_rte_eventdev" object is r

Re: [PATCH v10] app/dma-perf: introduce dma-perf application

28/06/2023 03:20, Cheng Jiang: > --- a/doc/guides/tools/index.rst > +++ b/doc/guides/tools/index.rst > @@ -23,3 +23,4 @@ DPDK Tools User Guides > testregex > testmldev > dts > +dmaperf Would be better to have upper in the list (DTS should be last). I suggest before flow-perf as

Re: [PATCH v10] app/dma-perf: introduce dma-perf application

28/06/2023 03:20, Cheng Jiang: > There are many high-performance DMA devices supported in DPDK now, and > these DMA devices can also be integrated into other modules of DPDK as > accelerators, such as Vhost. Before integrating DMA into applications, > developers need to know the performance of thes

Re: [PATCH v2 0/5] net/mlx5: introduce Tx datapath tracing

Slava Ovsiienko, Jun 29, 2023 at 09:16: > Hi, Robin > > Thank you for your courtesy about script reviewing. > Please see an attachment - the raw data gathered as a result of tracing, and > brief description. Thanks for the details. I think that most of the contents of the included pdf file should

[PATCH v6 2/2] build: select optional libraries

There is currently no way to know which libraries are optional. Introduce a enable_libs option (close to what we have for drivers) so that packagers or projects consuming DPDK can more easily select the optional libraries that matter to them and disable other optional libraries. Signed-off-by: Dav

[PATCH v6 1/2] build: select deprecated libraries

Rework deprecated libraries selection by introducing a new configuration option. This breaks existing configurations that were relying on disable_libs='' for enabling deprecated libraries. On the other hand, it will make enabling optional libraries more straightforward by taking the deprecated lib

[PATCH v6 0/2] Select optional libraries

This series is one implementation to try and please users who want to select more easily which parts of DPDK are built. It introduces a change in behavior for enabling deprecated libraries: this series is aimed at the next release but sent early as a demo of what changes are required. A deprecati

RE: [PATCH] net/iavf: fix SCTP tunnel packet forwarding issue

> -Original Message- > From: Jiale, SongX > Sent: Wednesday, June 28, 2023 3:33 PM > To: Lu, Wenzhuo ; dev@dpdk.org > Cc: Lu, Wenzhuo > Subject: RE: [PATCH] net/iavf: fix SCTP tunnel packet forwarding issue > > > -Original Message- > > From: Wenzhuo Lu > > Sent: Wednesday, Ju

RE: [PATCH] vfio: do not coalesce DMA mappings

Hi Nipun, I'd like to appreciate your time reading this email. Our QA team found that since this commit "a399d7b5a994: do not coalesce DMA mappings" is introduced, the dpdk testpmd start with "--no-huge" parameters will failed, and shows "EAL: Cannot set up DMA remapping, error 28 (No space lef

RE: [PATCH v2] raw/ntb: add check for disabling interrupt in dev close ops

> -Original Message- > From: Junfeng Guo > Sent: Wednesday, June 28, 2023 5:12 PM > To: Wu, Jingjing > Cc: dev@dpdk.org; sta...@dpdk.org; He, Xingguang > ; Laatz, Kevin ; Guo, > Junfeng > Subject: [PATCH v2] raw/ntb: add check for disabling interrupt in dev close > ops > > During EAL cl

[PATCH 5/5] net/mlx5/hws: support default miss action on FDB

From: Alex Vesker Add the support for default miss on HWS FDB, this implementation was missing until now. Default miss can be used for more efficient miss flow instead of going to an empty matcher or some defecated empty table. Signed-off-by: Alex Vesker --- drivers/net/mlx5/hws/mlx5dr_table.c

[PATCH 4/5] net/mlx5/hws: renaming FT to TBL

An action naming change is made to describe the action better and avoid implicit meaning. Signed-off-by: Itamar Gozlan --- drivers/net/mlx5/hws/mlx5dr.h| 2 +- drivers/net/mlx5/hws/mlx5dr_action.c | 16 drivers/net/mlx5/hws/mlx5dr_debug.c | 2 +- drivers/net/mlx5/mlx5

[PATCH 3/5] net/mlx5/hws: remove duplicated reformat type

From: Shun Hao Currently there are two enum definitions for reformat type: mlx5dr_action_reformat_type and mlx5dr_action_type. They are actually the same. This patch remove the unnecessary mlx5dr_action_reformat_type so use only one definition for reformat type. Signed-off-by: Shun Hao --- dri

[PATCH 2/5] net/mlx5/hws: allow destroying rule resources on error

From: Alex Vesker In case the send engine is in error state API should provide a way to free rule resources and provide a fake completion. Otherwise there is no way to release these resources. Signed-off-by: Alex Vesker --- drivers/net/mlx5/hws/mlx5dr_rule.c | 10 +- 1 file changed, 5

[PATCH 1/5] net/mlx5/hws: remove uneeded new line for DR_LOG

From: Alex Vesker In some places an extra new line was added, remove to have clean prints. Signed-off-by: Alex Vesker --- drivers/net/mlx5/hws/mlx5dr_action.c | 30 +-- drivers/net/mlx5/hws/mlx5dr_cmd.c | 2 +- drivers/net/mlx5/hws/mlx5dr_definer.c | 4 ++-- driv