Re: [dpdk-dev] [PATCH v5 0/1] Support meter PPS(packet per second) in MLX5 PMD

Hi, > -Original Message- > From: Li Zhang > Sent: Tuesday, April 27, 2021 1:42 PM > To: Ori Kam ; Slava Ovsiienko > ; Matan Azrad ; Shahaf > Shuler > Cc: dev@dpdk.org; NBU-Contact-Thomas Monjalon > ; Raslan Darawsheh ; Roni > Bar Yanai > Subject: [PATCH v5 0/1] Support meter PPS(packet

[dpdk-dev] [PATCH V4 1/7] app/testpmd: fix forward lcores number when DCB test

For the DCB forwarding test, each core is assigned to each traffic class. Number of forwarding cores for DCB test must be equal or less than number of total TC. Otherwise, the following problems may occur: 1/ Redundant polling threads will be created when forwarding cores number is greater than

[dpdk-dev] [PATCH V4 2/7] app/testpmd: fix DCB forwarding configuration

After DCB mode is configured, the operations of port stop and port start change the value of the global variable "dcb_test", As a result, the forwarding configuration from DCB to RSS mode, namely, “dcb_fwd_config_setup()” to "rss_fwd_config_setup()". Currently, the 'dcb_flag' field in struct 'rte_

[dpdk-dev] [PATCH V4 3/7] app/testpmd: fix a segment fault when DCB test

After DCB mode is configured, if we decrease the number of RX and TX queues, fwd_config_setup() will be called to setup the DCB forwarding configuration. And forwarding streams are updated based on new queue numbers in fwd_config_setup(), but the mapping between the TC and queues obtained by rte_et

[dpdk-dev] [PATCH V4 5/7] app/testpmd: move position of verifying DCB test

Currently, the check for doing DCB test is assigned to start_packet_forwarding(), which will be called when run "start" cmd. But fwd_config_setup() is used in many scenarios, such as, "port config all rxq". This patch moves the check from start_packet_forwarding() to fwd_config_setup(). Fixes: 77

[dpdk-dev] [PATCH V4 6/7] app/testpmd: add forwarding configuration when config DCB

This patch adds fwd_config_setup() at the end of cmd_config_dcb_parsed() to update "cur_fwd_config", so that the actual forwarding streams can be queried by the "show config fwd" cmd. Signed-off-by: Huisong Li Signed-off-by: Lijun Ou --- app/test-pmd/cmdline.c | 4 ++-- 1 file changed, 2 insert

[dpdk-dev] [PATCH V4 0/7] modifications about DCB forwarding configuration

This patchset modifies some codes about DCB forwarding configuration. It mainly includes: - fix forward lcores number when DCB test - remove dcb_test - fix a segment fault when dcb test - add check for support of reporting DCB info - move position of verifying DCB test - add forwarding configuratio

[dpdk-dev] [PATCH V4 4/7] app/testpmd: add check for support of reporting DCB info

Currently, '.get_dcb_info' must be supported for the port doing DCB test, or all information in 'rte_eth_dcb_info' are zero. It should be prevented when user run cmd "port config 0 dcb vt off 4 pfc off". This patch adds the check for support of reporting dcb info. Signed-off-by: Huisong Li Signe

[dpdk-dev] [PATCH V4 7/7] app/testpmd: remove redundant fwd streams initialization

The fwd_config_setup() is called after init_fwd_streams(). The fwd_config_setup() will reinitialize forwarding streams. This patch removes init_fwd_streams() from init_config(). Signed-off-by: Huisong Li Signed-off-by: Lijun Ou --- app/test-pmd/testpmd.c | 4 1 file changed, 4 deletions(-)

Re: [dpdk-dev] [PATCH] net/virtio: fix kernel set memory table for multi-queue devices

> -Original Message- > From: Thierry Herbelot > Sent: Friday, April 23, 2021 8:25 PM > To: dev@dpdk.org > Cc: Thierry Herbelot ; Thomas Monjalon > ; Maxime Coquelin ; Xia, > Chenbo ; sta...@dpdk.org > Subject: [PATCH] net/virtio: fix kernel set memory table for multi-queue > devices > > R

Re: [dpdk-dev] nvgre inner rss problem in mlx5

rdma-core version is: rdma-core-52mlnx1-1.52104.x86_64 发件人：Asaf Penso 发送日期：2021-04-28 12:22:32 收件人："dev@dpdk.org" ,wenxu 主题：Re: [dpdk-dev] nvgre inner rss problem in mlx5 Hello Wenxu, Thank you for reaching us. I would like to know a few more details before I can provide an assistance. C

Re: [dpdk-dev] nvgre inner rss problem in mlx5

Hello Wenxu, Thank you for reaching us. I would like to know a few more details before I can provide an assistance. Can you share the version numbers for: rdma-core OFED OS Regards, Asaf Penso From: dev on behalf of wenxu Sent: Wednesday, April 28, 2021 6:47

[dpdk-dev] nvgre inner rss problem in mlx5

Hi mlnx teams, I test the upstream dpdk with the nvgre inner rss action with dpdk-testpmd # ./dpdk-testpmd -c 0x1f -n 4 -m 4096 -w ":19:00.1" --huge-dir=/mnt/ovsdpdk -- --forward-mode=rxonly --rxq=4 --txq=4 --auto-start --nb-cores=4 # testpmd>> flow create 0 ingress pattern eth / ipv

Re: [dpdk-dev] [PATCH v4 3/3] net/i40e: enable PCI bus master after reset

> -Original Message- > From: Wang, Haiyue > Sent: Tuesday, April 27, 2021 9:39 PM > To: dev@dpdk.org > Cc: Zhang, Qi Z ; Wang, Liang-min > ; david.march...@redhat.com; Wang, Haiyue > ; Xing, Beilei ; Guo, Jia > > Subject: [PATCH v4 3/3] net/i40e: enable PCI bus master after reset > >

Re: [dpdk-dev] [PATCH v4 2/3] net/iavf: enable PCI bus master after reset

> -Original Message- > From: Wang, Haiyue > Sent: Tuesday, April 27, 2021 9:39 PM > To: dev@dpdk.org > Cc: Zhang, Qi Z ; Wang, Liang-min > ; david.march...@redhat.com; Wang, Haiyue > ; Wu, Jingjing ; Xing, Beilei > > Subject: [PATCH v4 2/3] net/iavf: enable PCI bus master after reset >

Re: [dpdk-dev] [PATCH] vdpa/mlx5: improve portability of thread naming

> -Original Message- > From: dev On Behalf Of Thomas Monjalon > Sent: Thursday, April 22, 2021 1:59 AM > To: dev@dpdk.org > Cc: Matan Azrad ; Viacheslav Ovsiienko > > Subject: [dpdk-dev] [PATCH] vdpa/mlx5: improve portability of thread naming > > The function pthread_setname_np is non-po

Re: [dpdk-dev] [PATCH] net/virtio: fix getline memory leakage

> -Original Message- > From: stable On Behalf Of Xia, Chenbo > Sent: Wednesday, April 21, 2021 10:10 AM > To: Min Hu (Connor) ; dev@dpdk.org > Cc: Yigit, Ferruh ; maxime.coque...@redhat.com; > sta...@dpdk.org > Subject: Re: [dpdk-stable] [PATCH] net/virtio: fix getline memory leakage > >

Re: [dpdk-dev] [PATCH v5] vhost: allocate and free packets in bulk in Tx packed

> -Original Message- > From: dev On Behalf Of Maxime Coquelin > Sent: Wednesday, April 21, 2021 3:49 PM > To: Balazs Nemeth ; dev@dpdk.org > Cc: david.march...@redhat.com > Subject: Re: [dpdk-dev] [PATCH v5] vhost: allocate and free packets in bulk in > Tx packed > > > > On 4/16/21 12:2

Re: [dpdk-dev] [PATCH v3] vhost: don't track remaining packets separately

> -Original Message- > From: dev On Behalf Of Balazs Nemeth > Sent: Tuesday, April 13, 2021 9:31 PM > To: bnem...@redhat.com; dev@dpdk.org; maxime.coque...@redhat.com > Subject: [dpdk-dev] [PATCH v3] vhost: don't track remaining packets separately > > The remained variable stores the same

Re: [dpdk-dev] [PATCH v3] vhost: read last used index once

> -Original Message- > From: dev On Behalf Of Balazs Nemeth > Sent: Wednesday, April 7, 2021 5:55 PM > To: bnem...@redhat.com; dev@dpdk.org > Subject: [dpdk-dev] [PATCH v3] vhost: read last used index once > > Instead of calculating the address of a packed descriptor based on the > vq->de

Re: [dpdk-dev] [PATCH V3 3/7] app/testpmd: fix a segment fault when DCB test

在 2021/4/28 10:02, Li, Xiaoyun 写道: -Original Message- From: Huisong Li Sent: Tuesday, April 27, 2021 22:11 To: Li, Xiaoyun ; dev@dpdk.org Cc: Yigit, Ferruh Subject: Re: [PATCH V3 3/7] app/testpmd: fix a segment fault when DCB test 在 2021/4/27 19:19, Li, Xiaoyun 写道: -Original

Re: [dpdk-dev] [PATCH V3 6/7] app/testpmd: add forwarding config in start port

在 2021/4/27 19:43, Li, Xiaoyun 写道: -Original Message- From: Huisong Li Sent: Tuesday, April 20, 2021 17:01 To: dev@dpdk.org Cc: Yigit, Ferruh ; Li, Xiaoyun ; linux...@openeuler.org; lihuis...@huawei.com Subject: [PATCH V3 6/7] app/testpmd: add forwarding config in start port Most oper

Re: [dpdk-dev] [PATCH V2 1/7] app/testpmd: fix forward lcores number when DCB test

> -Original Message- > From: Huisong Li > Sent: Tuesday, April 27, 2021 21:50 > To: Li, Xiaoyun ; dev@dpdk.org > Cc: Yigit, Ferruh > Subject: Re: [PATCH V2 1/7] app/testpmd: fix forward lcores number when DCB > test > > > 在 2021/4/27 17:20, Li, Xiaoyun 写道: > > > >> -Original Messa

Re: [dpdk-dev] [PATCH] examples/vhost: fix ioat ring space in callbacks

> -Original Message- > From: Jiang, Cheng1 > Sent: Wednesday, March 17, 2021 1:41 PM > To: maxime.coque...@redhat.com; Xia, Chenbo > Cc: dev@dpdk.org; Hu, Jiayu ; Yang, YvonneX > ; Wang, Yinan ; Jiang, Cheng1 > ; sta...@dpdk.org > Subject: [PATCH] examples/vhost: fix ioat ring space in ca

Re: [dpdk-dev] [PATCH v3 0/4] Refactor async vhost control path

> -Original Message- > From: Hu, Jiayu > Sent: Tuesday, April 20, 2021 4:58 PM > To: dev@dpdk.org > Cc: maxime.coque...@redhat.com; Xia, Chenbo ; Wang, > Yinan ; Pai G, Sunil ; Jiang, > Cheng1 ; Hu, Jiayu > Subject: [PATCH v3 0/4] Refactor async vhost control path > > This patch set refa

Re: [dpdk-dev] [PATCH V3 3/7] app/testpmd: fix a segment fault when DCB test

> -Original Message- > From: Huisong Li > Sent: Tuesday, April 27, 2021 22:11 > To: Li, Xiaoyun ; dev@dpdk.org > Cc: Yigit, Ferruh > Subject: Re: [PATCH V3 3/7] app/testpmd: fix a segment fault when DCB test > > > 在 2021/4/27 19:19, Li, Xiaoyun 写道: > > > >> -Original Message-

Re: [dpdk-dev] [PATCH] net/iavf: fix VLAN tag extraction handling in AVX512 path

> -Original Message- > From: Sun, QinX > Sent: Tuesday, April 27, 2021 3:11 PM > To: Rong, Leyi ; Zhang, Qi Z ; Lu, > Wenzhuo > Cc: dev@dpdk.org; Rong, Leyi > Subject: RE: [dpdk-dev] [PATCH] net/iavf: fix VLAN tag extraction handling in > AVX512 path > > > -Original Message-

Re: [dpdk-dev] [PATCH v1] net/ice/base: clean code of finding GTPU dummy pkt

> -Original Message- > From: Zhang, Yuying > Sent: Monday, April 26, 2021 3:18 PM > To: dev@dpdk.org; Zhang, Qi Z > Cc: Zhang, Yuying > Subject: [PATCH v1] net/ice/base: clean code of finding GTPU dummy pkt > > Four gtpu tunnel types are used twice to find gtpu dummy packets > (ipv4_

Re: [dpdk-dev] [PATCH] net/e1000: fix param misusage when set rte flow error

> -Original Message- > From: dev On Behalf Of Wang, Haiyue > Sent: Tuesday, April 27, 2021 9:23 PM > To: Min Hu (Connor) ; dev@dpdk.org > Cc: Yigit, Ferruh ; Guo, Jia > Subject: Re: [dpdk-dev] [PATCH] net/e1000: fix param misusage when set rte > flow error > > > -Original Message-

Re: [dpdk-dev] [PATCH v1 2/2] net/i40e: allow get_monitor_addr for VF driver

> -Original Message- > From: dev On Behalf Of David Hunt > Sent: Monday, April 26, 2021 10:23 PM > To: Burakov, Anatoly ; dev@dpdk.org; Xing, > Beilei ; Guo, Jia ; Ananyev, > Konstantin > Subject: Re: [dpdk-dev] [PATCH v1 2/2] net/i40e: allow get_monitor_addr for > VF driver > > Hi Ana

Re: [dpdk-dev] [PATCH v1 1/2] net/ixgbe: allow get_monitor_addr for VF driver

> -Original Message- > From: dev On Behalf Of Wang, Haiyue > Sent: Tuesday, April 27, 2021 9:43 AM > To: Hunt, David ; Burakov, Anatoly > ; dev@dpdk.org; Guo, Jia ; > Ananyev, Konstantin ; Liang Ma > > Subject: Re: [dpdk-dev] [PATCH v1 1/2] net/ixgbe: allow get_monitor_addr for > VF dri

Re: [dpdk-dev] [PATCH] net/e1000: fix param misusage when set rte flow error

> -Original Message- > From: Min Hu (Connor) > Sent: Wednesday, April 28, 2021 09:16 > To: Wang, Haiyue ; dev@dpdk.org > Cc: Yigit, Ferruh ; Guo, Jia > Subject: Re: [PATCH] net/e1000: fix param misusage when set rte flow error > > > > 在 2021/4/27 21:22, Wang, Haiyue 写道: > >> -Origi

Re: [dpdk-dev] [PATCH] net/e1000: fix param misusage when set rte flow error

在 2021/4/27 21:22, Wang, Haiyue 写道: -Original Message- From: Min Hu (Connor) Sent: Tuesday, April 27, 2021 16:51 To: dev@dpdk.org Cc: Yigit, Ferruh ; Guo, Jia ; Wang, Haiyue Subject: [PATCH] net/e1000: fix param misusage when set rte flow error This patch fixes parameter misusage w

Re: [dpdk-dev] L3fwd mode in testpmd

> > > > > This discussion has ended up as a stats discussion. But, we also need > to be able to change the configurable parameters easily. > > > > > If we implement the stats and ability to change the configurable > > > > > parameters, then it is essentially bringing in some of the > > > > > capa

Re: [dpdk-dev] L3fwd mode in testpmd

> > On Thu, Mar 11, 2021 at 12:01 AM Honnappa Nagarahalli > > wrote: > > > > > > Hello, > > > Performance of L3fwd example application is one > > > of the key > > benchmarks in DPDK. However, the appli

Re: [dpdk-dev] L3fwd mode in testpmd

> > > > > > > On Tue, 27 Apr 2021 10:50:20 +0100 > > > Ferruh Yigit wrote: > > > > > > > Agree that it may help testing to have l3fwd support on the testpmd. > > > > > > > > Two concerns, > > > > 1) Testpmd already too complex. > > > > 2) Code duplication. > > > > > > > > For 1), if the l3fwd c

Re: [dpdk-dev] L3fwd mode in testpmd

> > > > On Tue, 27 Apr 2021 10:50:20 +0100 > > Ferruh Yigit wrote: > > > > > Agree that it may help testing to have l3fwd support on the testpmd. > > > > > > Two concerns, > > > 1) Testpmd already too complex. > > > 2) Code duplication. > > > > > > For 1), if the l3fwd can be implemented in test

Re: [dpdk-dev] L3fwd mode in testpmd

> > On Tue, 27 Apr 2021 10:50:20 +0100 > Ferruh Yigit wrote: > > > Agree that it may help testing to have l3fwd support on the testpmd. > > > > Two concerns, > > 1) Testpmd already too complex. > > 2) Code duplication. > > > > For 1), if the l3fwd can be implemented in testpmd as new, independ

[dpdk-dev] [PATCH v8 10/10] app/testpmd: enable building testpmd on Windows

From: Jie Zhou - Disable unsupported Apps on Windows - Enable building of testpmd on Windows Signed-off-by: Jie Zhou Signed-off-by: Jie Zhou --- app/meson.build| 4 app/pdump/meson.build | 6 ++ app/proc-info/meson.build | 6 ++ app/test-

[dpdk-dev] [PATCH v8 09/10] app/testpmd: fix unused function warnings

Function print_fdir_mask and print_fdir_flex_payload is only called when either i40e or ixgbe presents. Add #if defined to remove "unused function" compilation warning. Signed-Off-By: Jie Zhou Signed-off-by: Jie Zhou --- app/test-pmd/config.c | 82 +-- 1

[dpdk-dev] [PATCH v8 05/10] app/testpmd: add clock_gettime on Windows

Add clock_gettime for testpmd on Windows in config.h Signed-off-by: Jie Zhou Signed-off-by: Jie Zhou --- app/test-pmd/config.c | 10 ++- app/test-pmd/config.h | 66 +++ 2 files changed, 68 insertions(+), 8 deletions(-) create mode 100644 app/test-pmd

[dpdk-dev] [PATCH v8 07/10] app/testpmd: replace POSIX specific code

- Make printf format OS independent - Replace htons with RTE_BE16 - Replace POSIX specific inet_aton with OS independent inet_pton - Replace sleep with rte_delay_us_sleep - Repalce random with rte_rand - #ifndef mman related code for now Signed-off-by: Jie Zhou Signed-off-by: Jie Zhou ---

[dpdk-dev] [PATCH v8 06/10] app/testpmd: fix parse_fec_mode return type

Fix parse_fec_mode to return fec_capa instead of mode Signed-off-by: Jie Zhou Signed-off-by: Jie Zhou --- app/test-pmd/cmdline.c | 6 +++--- app/test-pmd/config.c | 4 ++-- app/test-pmd/testpmd.h | 2 +- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/app/test-pmd/cmdline.c b/ap

[dpdk-dev] [PATCH v8 08/10] app/testpmd: fix headers inclusion

- Include rte_os_shim.h in testpmd.h - Remove redundant headers Signed-off-by: Jie Zhou Signed-off-by: Jie Zhou --- app/test-pmd/cmdline.c | 3 --- app/test-pmd/cmdline_flow.c | 2 -- app/test-pmd/parameters.c | 1 - app/test-pmd/testpmd.h | 1 + 4 files changed, 1 insertion(+), 6 d

[dpdk-dev] [PATCH v8 04/10] app/testpmd: resolve name collisions

Resolve name collisions with Windows types Signed-off-by: Jie Zhou Signed-off-by: Jie Zhou --- app/test-pmd/cmdline_flow.c | 512 ++-- 1 file changed, 256 insertions(+), 256 deletions(-) diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c inde

[dpdk-dev] [PATCH v8 03/10] eal/windows: add device event stubs

Add device event stubs in eal_dev.c for Windows Signed-off-by: Jie Zhou Signed-off-by: Jie Zhou --- lib/eal/windows/eal_dev.c | 33 + lib/eal/windows/meson.build | 1 + 2 files changed, 34 insertions(+) create mode 100644 lib/eal/windows/eal_dev.c diff --git

[dpdk-dev] [PATCH v8 02/10] eal/windows: add necessary macros

Add required macros by testpmd on Windows in rte_os_shim.h Signed-off-by: Jie Zhou Signed-off-by: Jie Zhou --- lib/eal/windows/include/rte_os_shim.h | 9 + 1 file changed, 9 insertions(+) diff --git a/lib/eal/windows/include/rte_os_shim.h b/lib/eal/windows/include/rte_os_shim.h index

[dpdk-dev] [PATCH v8 00/10] app/testpmd: enable testpmd on Windows

This patchset is to enable testpmd on windows. It mainly includes: - Enable building libraries on Windows that testpmd depends on - Add necessary macros required by testpmd on Windows in rte_os_shim.h - Add device event stubs for Windows - Resolve name collisions with Windows types - Add clock_gett

[dpdk-dev] [PATCH v8 01/10] lib: build libraries that testpmd depends on

Enable building libraries that testpmd depends on Signed-off-by: Jie Zhou Signed-off-by: Jie Zhou --- lib/meson.build | 7 +++ 1 file changed, 7 insertions(+) diff --git a/lib/meson.build b/lib/meson.build index c9a20f65b..2d499b238 100644 --- a/lib/meson.build +++ b/lib/meson.build @@ -76

Re: [dpdk-dev] [PATCH v2 07/16] build: change infrastructure file tabs to spaces

On 4/20/21 3:22 AM, Bruce Richardson wrote: Switch from using tabs to 4 spaces for meson.build indentation, for the basic infrastructure and tooling files, as well as doc and kernel directories. Signed-off-by: Bruce Richardson --- buildtools/chkincs/meson.build| 24 +- buildtools/mes

Re: [dpdk-dev] [PATCH] autotest: disable lcores_autotest on ppc

On 4/20/21 4:45 AM, luca.bocca...@gmail.com wrote: From: Luca Boccassi This test consistently times out on ppc64 builds. Disable it. Cc: sta...@dpdk.org Signed-off-by: Luca Boccassi Is there something new about how/when this test is run during continuous integration? The test time is

Re: [dpdk-dev] [PATCH] autotest: disable lcores_autotest on ppc

First things first, let's Cc the PPC maintainer: David Christensen https://bugs.dpdk.org/show_bug.cgi?id=684 Tried GCC 10 on RHEL 8.3 and running the lcores_autotest individually does not produce any errors. I can't see in the log file how the test is called when it generates an error. Ca

Re: [dpdk-dev] [PATCH 5/5] vhost: fix offload flags in Rx path

On Tue, Apr 27, 2021 at 7:09 PM David Marchand wrote: > Investigating the other side (GUEST_ECN + the virtio pmd) could be > worth later, as I think GSO+ECN packets are dropped in the current > code. Errr, but that would be a problem only for vhost-kernel -> virtio pmd. Not sure this is a usecase

Re: [dpdk-dev] [PATCH 5/5] vhost: fix offload flags in Rx path

On Tue, Apr 13, 2021 at 5:27 PM Maxime Coquelin wrote: > On 4/8/21 8:38 PM, Flavio Leitner wrote: > > On Thu, Apr 01, 2021 at 11:52:43AM +0200, David Marchand wrote: > >> The vhost library current configures Tx offloading (PKT_TX_*) on any > >> packet received from a guest virtio device which asks

Re: [dpdk-dev] [PATCH 4/4] net/txgbe: add copyright owner

On 4/25/2021 9:03 AM, Jiawen Wu wrote: > All rights reserved by Beijing Wangxun Technology Co., Ltd. > Part of the code references Intel. > > Signed-off-by: Jiawen Wu Looks good to me, thanks for the update.

Re: [dpdk-dev] [PATCH 3/4] net/txgbe: remove port representor

On 4/25/2021 9:03 AM, Jiawen Wu wrote: > Remove port representor in device probe process. > It can be good to highlight that is because port representor is not supported by the driver. > Signed-off-by: Jiawen Wu > --- > drivers/net/txgbe/txgbe_ethdev.c | 21 ++--- > 1 file chan

Re: [dpdk-dev] [dpdk-stable] [PATCH 2/4] net/txgbe: fix VF MTU limit setting

On 4/25/2021 9:03 AM, Jiawen Wu wrote: > Fix to set MTU bigger than mbuf size when device is stopped, scattered > rx may be enabled in next starting, exempt from setting in EAL parameters. > The main change is 'dev_started' check, right? Can you please reword above paragraph to clarify it? > Fix

Re: [dpdk-dev] [dpdk-stable] [PATCH 1/4] net/txgbe: fix RSS in double VALN

On 4/25/2021 9:03 AM, Jiawen Wu wrote: > Regard QINQ as one case of extend VLAN, and always enable QINQ mode when > double VLAN is set, to avoid RSS does not work for QINQ packets. > It is hard to understand above paragraph, can you please reword it? Also in the patch title s/VALN/VLAN VLAN_EXT

Re: [dpdk-dev] L3fwd mode in testpmd

On Tue, 27 Apr 2021 10:50:20 +0100 Ferruh Yigit wrote: > Agree that it may help testing to have l3fwd support on the testpmd. > > Two concerns, > 1) Testpmd already too complex. > 2) Code duplication. > > For 1), if the l3fwd can be implemented in testpmd as new, independent > forwarding mode,

[dpdk-dev] [PATCH v5 2/2] mempool: distinguish debug counters from cache and pool

From: Joyce Kong If cache is enabled, objects will be retrieved/put from/to cache, subsequently from/to the common pool. Now the debug stats calculate the objects retrieved/put from/to cache and pool together, it is better to distinguish them. Signed-off-by: Joyce Kong Signed-off-by: Dharmik Th

[dpdk-dev] [PATCH v5 1/2] mempool: make stats macro generic

Make __MEMPOOL_STAT_ADD macro more generic and delete __MEMPOOL_CONTIG_BLOCKS_STAT_ADD macro Suggested-by: Olivier Matz Signed-off-by: Dharmik Thakkar Reviewed-by: Ruifeng Wang Reviewed-by: Honnappa Nagarahalli Acked-by: Olivier Matz --- lib/mempool/rte_mempool.h | 34 ---

[dpdk-dev] [PATCH v5 0/2] mempool: add debug stats

- Add debug counters for objects put/get to/from the common pool. - Make __MEMPOOL_STAT_ADD() more generic --- v5: - Fix compilation issue - Move 'increment get_success_bulk/objs' inside __mempool_generic_get v4: - Remove cache stats v3: - Add a patch to make stat add macro generic - Remove

[dpdk-dev] [PATCH 17/17] net/mlx5: add support of CT between two ports

After creating a connection tracking context, it could be used between two ports. For each port, the flow for one direction traffic will be created. The context can only be shared between the owner port and the peer port that specified when creating. Only the owner port could update the context or

[dpdk-dev] [PATCH 16/17] net/mlx5: reduce the reference count of CT

In order to release the DR actions and reuse the context of a CT, the reference count should be handled correctly, especially in the flow destroying. The CT index will be recorded in the rte_flow by reusing the age index. The action context type should be saved for CT also. When destroying a flow,

[dpdk-dev] [PATCH 15/17] net/mlx5: validation of CT item

The item of ASO connection tracking will be translated into the register value when matching. The validation of this item has no dependency on other layers, since the flow including this item should be jumped from another group. All the layers checking was already done in the previous groups. Only

[dpdk-dev] [PATCH 14/17] net/mlx5: validation of CT action

The validation of a CT action contains two parts. The first is the CT action configurations parameter. When creating a CT action context, some members need to be verified. The second is that when creating a flow, the DR action of CT should be validated with other actions and items as well. Current

[dpdk-dev] [PATCH 12/17] net/mlx5: add translation of CT item

The return register of the DR action will be used for matching. After the ASO CT checking of a TCP packet, the syndrome is filled in the register. Only the 8 LSB should be used. A converting from RTE_FLOW_CONNTRACK_FLAG* to the syndrome should be done after checing the spec and mask fields. Signed

[dpdk-dev] [PATCH 13/17] net/mlx5: add CT context update

When updating a connection tracking context, two separate parts could be updated. First, the direction. This will only update the traffic direction recorded in the software for flow creation. Second, the TCP parameters. The hardware context will be updated via the WQE. This update will be blocked u

[dpdk-dev] [PATCH 11/17] net/mlx5: add translation for CT action

When creating a flow with this action context for CT, it needs to be translated in 2 levels. First, retrieve from action context to RTE_FLOW action. Second, translate it to the correct DR action with traffic direction. Before using the DR action in a flow, the CT context should be available to use

[dpdk-dev] [PATCH 10/17] net/mlx5: add ASO CT destroy handling

When trying to destroy an ASO connection tracking context, the DR action created on this context should also be destroyed. Before inserting the related software object into the management free list, the reference count should be checked. Right now, the context object will not be freed to the syste

[dpdk-dev] [PATCH 09/17] net/mlx5: add ASO CT query implementation

After the connection tracking context is created and being used by the flows, the context will be updated by the HW automatically after a packet passed the CT validation. E.g., the ACK, SEQ, window and state of CT can be updated with both direction traffic. In order to query the updated contents o

[dpdk-dev] [PATCH 08/17] net/mlx5: close CT management structure

When freeing the IB shared context during stopping a device, the ASO connection tracking management structure should also be cleaned up. All the DR actions created should be destroyed. The structures need to be freed and ASO CT QP should be released. In the meanwhile, the allocated and registered

[dpdk-dev] [PATCH 07/17] net/mlx5: add actions creating for CT

Allocating a CT from the management pools and creating the DR actions for both directions by default. If there is no available connection tracking action, a new pool will be created with a fixed size bulk allocation. Right now, all the resources are controlled by the linked list. The ASO connecti

[dpdk-dev] [PATCH 06/17] net/mlx5: add modify support for CT

After the connection tracking object bulk is allocated, all the objects' contents are filled with zero by default. One object must be modified via WQE operation before using it. In order to reduce the latency for the flow creation, an asynchronous way is used instead of busy waiting for the CQE to

[dpdk-dev] [PATCH 05/17] common/mlx5: add Dexv CT objects creation

Adding support for connection tracking ASO creation via Devx command. Right now only bulk creation is supported. By default, the objects with zero contents will be created. Before using a single object, the modification via posting a WQE to the ASO CT SQ is needed. Signed-off-by: Bing Zhao ---

[dpdk-dev] [PATCH 04/17] net/mlx5: initialization of CT management

The definitions of ASO connection tracking objects management structures are added. Considering performance, the bulk allocation of ASO CT objects should be used. The maximal value per bulk and the granularity could be fetched from HCA capabilities 2. Right now, a fixed number of 64 is used for ea

[dpdk-dev] [PATCH 03/17] net/mlx5: use meter color reg for CT

Based on the capacity, 3 registers could be used. Due to the register allocation, only the one REG_C_3 for meter color could be reused right now. Then in the same flow, no more than one ASO action can be supported. Signed-off-by: Bing Zhao --- drivers/net/mlx5/mlx5_flow.c | 4 +++- drivers/net/

[dpdk-dev] [PATCH 02/17] common/mlx5: add CT offload capability checking

During startup, the ASO connection tracking offload capability could be queried via HCA_CAP_QUERY command. If the HW doesn't support ASO CT, the value would be 0 by default. The following initialization should be skipped and the creation of the CT object should return a failure directly. The follo

[dpdk-dev] [PATCH 01/17] common/mlx5: add connection tracking object definition

The structures of ASO connection tracking offload object are added based on the definitions in the PRM. One CT object context will be loaded into the cache completely in a reversed order of dwords. The valid bit should be the MSB of the last dword. This is used for the conntrack context creation an

[dpdk-dev] [PATCH 00/17] conntrack support in mlx5 PMD

This patch set adds the connection tracking offload support in the mlx5 driver. Bing Zhao (17): common/mlx5: add connection tracking object definition common/mlx5: add CT offload capability checking net/mlx5: use meter color reg for CT net/mlx5: initialization of CT management common/mlx

Re: [dpdk-dev] [PATCH] net/txgbe: fix null pointer check problem

On 4/25/2021 1:54 PM, Min Hu (Connor) wrote: > From: HongBo Zheng > > In function cons_parse_ntuple_filter, item->spec and item->mask > should be confirmed not null before use memcmp on it, current > judgement (item->spec || item->mask) just can confirm item->spec > or item->mask is not null, and

Re: [dpdk-dev] [PATCH] common/sfc_efx/base: fix the problem of using null pointer

On 4/25/2021 1:58 PM, Min Hu (Connor) wrote: > From: HongBo Zheng > > In function efx_pci_xilinx_cap_tbl_find, pointer entry_offsetp is used > before null pointer check, which may cause access to null pointer. > > This patch fix this problem. > > Fixes: ba9568b8b4b7 ("common/sfc_efx/base: add X

Re: [dpdk-dev] [PATCH v3] net/hns3: fix parse link fails code fail

On 4/27/2021 2:43 PM, Min Hu (Connor) wrote: > > > 在 2021/4/27 21:19, Ferruh Yigit 写道: >> On 4/27/2021 2:03 PM, Min Hu (Connor) wrote: >>> >>> >>> 在 2021/4/27 20:45, Ferruh Yigit 写道: On 4/27/2021 1:17 PM, Min Hu (Connor) wrote: > From: Chengwen Feng > > The link fails code shoul

[dpdk-dev] [dpdk-techboard] Minutes of Technical Board Meeting, 2021-04-21

Minutes of Technical Board Meeting, 2021-04-21 Members Attending --- -Aaron -Bruce -Ferruh -Hemant -Honnappa -Jerin -Kevin -Konstantin (chair) -Maxime -Olivier -Stephen -Thomas NOTE: The technical board meetings every second Wednesday at https://meet.jit.si/DPDK at 3 pm U

Re: [dpdk-dev] [PATCH v4 1/3] bus/pci: set PCI master in command register

On 27/04/2021 14:39, Haiyue Wang wrote: > Add the API to set 'Bus Master Enable' bit to be enabled or disabled in > the PCI command register. > > Signed-off-by: Haiyue Wang > --- > drivers/bus/pci/pci_common.c | 28 > drivers/bus/pci/rte_bus_pci.h | 14 ++

Re: [dpdk-dev] [PATCH 3/4] net: introduce functions to verify L4 checksums

> From: Olivier Matz [mailto:olivier.m...@6wind.com] > Sent: Tuesday, April 27, 2021 3:58 PM > > Since commit d5df2ae0428a ("net: fix unneeded replacement of TCP > checksum 0"), the functions rte_ipv4_udptcp_cksum() and > rte_ipv6_udptcp_cksum() can return either 0x or 0x when used to > ve

[dpdk-dev] Recall: [PATCH 3/4] net: introduce functions to verify L4 checksums

Morten Brørup would like to recall the message, "[PATCH 3/4] net: introduce functions to verify L4 checksums".

Re: [dpdk-dev] [PATCH 3/4] net: introduce functions to verify L4 checksums

> From: Olivier Matz [mailto:olivier.m...@6wind.com] > Sent: Tuesday, April 27, 2021 3:58 PM > > Since commit d5df2ae0428a ("net: fix unneeded replacement of TCP > checksum 0"), the functions rte_ipv4_udptcp_cksum() and > rte_ipv6_udptcp_cksum() can return either 0x or 0x when used to > ve

Re: [dpdk-dev] [PATCH 6/6] net/hns3: fix vector Rx burst default value

From:Ferruh Yigit To:humin (Q) ;dev Date:2021-04-27 21:47:08 Subject:Re: [dpdk-dev] [PATCH 6/6] net/hns3: fix vector Rx burst default value On 4/26/2021 4:34 AM, Min Hu (Connor) wrote: > From: Chengwen Feng < fengcheng...@huawei.com> > > Currently, driver uses t

Re: [dpdk-dev] [PATCH 1/6] net/hns3: delete some unused capabilities

On 4/27/2021 3:26 PM, Fengchengwen wrote: > > > *From:*Ferruh Yigit > *To:*humin (Q) ;dev > *Date:*2021-04-27 21:37:59 > *Subject:*Re: [dpdk-dev] [PATCH 1/6] net/hns3: delete some unused capabilities > > On 4/26/2021 4:34 AM, Min Hu (Connor) wrote: >> From: Chengwen Feng < fengcheng...@huawei.

Re: [dpdk-dev] [PATCH 5/6] net/hns3: log fdir configuration

From:Ferruh Yigit To:Fengchengwen ;humin (Q) ;dev Date:2021-04-27 22:26:09 Subject:Re: [dpdk-dev] [PATCH 5/6] net/hns3: log fdir configuration On 4/27/2021 3:15 PM, Fengchengwen wrote: > > > > > > *From:*Ferruh Yigit < ferruh.yi...@intel.com> > *To:*humin (Q) <

Re: [dpdk-dev] [PATCH 1/6] net/hns3: delete some unused capabilities

From:Ferruh Yigit To:humin (Q) ;dev Date:2021-04-27 21:37:59 Subject:Re: [dpdk-dev] [PATCH 1/6] net/hns3: delete some unused capabilities On 4/26/2021 4:34 AM, Min Hu (Connor) wrote: > From: Chengwen Feng < fengcheng...@huawei.com> > > This patch deletes some un

Re: [dpdk-dev] [PATCH 5/6] net/hns3: log fdir configuration

On 4/27/2021 3:15 PM, Fengchengwen wrote: > > > > > > *From:*Ferruh Yigit > *To:*humin (Q) ;dev > *Date:*2021-04-27 21:39:45 > *Subject:*Re: [dpdk-dev] [PATCH 5/6] net/hns3: log fdir configuration > > On 4/26/2021 4:34 AM, Min Hu (Connor) wrote: >> From: Chengwen Feng < fengcheng...@huawei.

Re: [dpdk-dev] [PATCH v3 2/2] mbuf: better document usage of packet pool initializers

Olivier Matz writes: > Clarify that the mempool private initializer and object initializer used > for packet pools require that the mempool private size is large enough. > > Also add an assert (only enabled when -DRTE_ENABLE_ASSERT is passed) to > check this constraint. > > Signed-off-by: Olivier

[dpdk-dev] [Bug 689] RTE bus list not populated causing EAL: failed to parse device "XX:XX.X" on CentOS 7 with static linkage

https://bugs.dpdk.org/show_bug.cgi?id=689 Thomas Monjalon (tho...@monjalon.net) changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolutio

Re: [dpdk-dev] [PATCH 5/6] net/hns3: log fdir configuration

From:Ferruh Yigit To:humin (Q) ;dev Date:2021-04-27 21:39:45 Subject:Re: [dpdk-dev] [PATCH 5/6] net/hns3: log fdir configuration On 4/26/2021 4:34 AM, Min Hu (Connor) wrote: > From: Chengwen Feng < fengcheng...@huawei.com> > > The rte flow interface does not

Re: [dpdk-dev] [PATCH V3 2/7] app/testpmd: fix DCB forwarding configuration

在 2021/4/27 18:59, Li, Xiaoyun 写道: -Original Message- From: Huisong Li Sent: Tuesday, April 20, 2021 17:01 To: dev@dpdk.org Cc: Yigit, Ferruh ; Li, Xiaoyun ; linux...@openeuler.org; lihuis...@huawei.com Subject: [PATCH V3 2/7] app/testpmd: fix DCB forwarding configuration After DCB

Re: [dpdk-dev] [PATCH V3 3/7] app/testpmd: fix a segment fault when DCB test

在 2021/4/27 19:19, Li, Xiaoyun 写道: -Original Message- From: Huisong Li Sent: Tuesday, April 20, 2021 17:01 To: dev@dpdk.org Cc: Yigit, Ferruh ; Li, Xiaoyun ; linux...@openeuler.org; lihuis...@huawei.com Subject: [PATCH V3 3/7] app/testpmd: fix a segment fault when DCB test After DCB

[dpdk-dev] [PATCH v4 3/3] net/i40e: enable PCI bus master after reset

The VF reset can be triggerred by the PF reset event, in this case, the PCI bus master will be cleared, then the VF is not allowed to issue any Memory or I/O Requests. So after the reset event is detected, always enable the PCI bus master. And align the VF reset event handling in device close mod

[dpdk-dev] [PATCH v4 2/3] net/iavf: enable PCI bus master after reset

The VF reset can be triggerred by the PF reset event, in this case, the PCI bus master will be cleared, then the VF is not allowed to issue any Memory or I/O Requests. So after the reset event is detected, always enable the PCI bus master. Signed-off-by: Haiyue Wang --- drivers/net/iavf/iavf_et

  1   2   >