Re: [dpdk-dev] [PATCH v1] doc: policy on promotion of experimental APIs

On 01/07/2021 16:09, Tyler Retzlaff wrote: > On Thu, Jul 01, 2021 at 11:19:27AM +0100, Kinsella, Ray wrote: >> >> >> On 30/06/2021 20:56, Tyler Retzlaff wrote: >>> On Tue, Jun 29, 2021 at 07:38:05PM +0100, Kinsella, Ray wrote: >> +Promotion to stable >> +~~~ >>>

Re: [dpdk-dev] 19.11.9 patches review and test

On Thu, Jul 1, 2021 at 5:50 PM Ali Alnubani wrote: > > Hi, > > > -Original Message- > > From: Christian Ehrhardt > > Sent: Thursday, June 17, 2021 9:38 AM > > To: sta...@dpdk.org > > Cc: dev@dpdk.org; Abhishek Marathe ; > > Akhil Goyal ; Ali Alnubani ; > > benjamin.wal...@intel.com; David

[dpdk-dev] [PATCH v3 22/22] net/mlx5: optimize Rx queue match

As hrxq struct has the indirect table pointer, while matching the hrxq, better to use the hrxq indirect table instead of searching from the list. This commit optimizes the hrxq indirect table matching. Signed-off-by: Suanming Mou Acked-by: Matan Azrad --- drivers/net/mlx5/mlx5_rxq.c | 18 +

[dpdk-dev] [PATCH v3 21/22] net/mlx5: support list none local core operations

This commit supports the list none local core operations with an extra sub-list. Signed-off-by: Suanming Mou Acked-by: Matan Azrad --- drivers/common/mlx5/mlx5_common_utils.c | 92 + drivers/common/mlx5/mlx5_common_utils.h | 9 ++- 2 files changed, 71 insertions(+), 30

[dpdk-dev] [PATCH v3 20/22] net/mlx5: support index pool none local core operations

This commit supports the index pool none local core operations with an extra cache. Signed-off-by: Suanming Mou Acked-by: Matan Azrad --- drivers/net/mlx5/mlx5_utils.c | 75 +-- drivers/net/mlx5/mlx5_utils.h | 3 +- 2 files changed, 56 insertions(+), 22 deletion

[dpdk-dev] [PATCH v3 19/22] net/mlx5: change memory release configuration

This commit changes the index pool memory release configuration to 0 when memory reclaim mode is not required. Signed-off-by: Suanming Mou Acked-by: Matan Azrad --- drivers/net/mlx5/mlx5.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/mlx5/mlx5.c b/drivers/net/mlx5/mlx5.c in

[dpdk-dev] [PATCH v3 18/22] common/mlx5: optimize cache list object memory

Currently, hash list uses the cache list as bucket list. The list in the buckets have the same name, ctx and callbacks. This wastes the memory. This commit abstracts all the name, ctx and callback members in the list to a constant struct and others to the inconstant struct, uses the wrapper functi

[dpdk-dev] [PATCH v3 17/22] net/mlx5: optimize hash list table allocate on demand

Currently, all the hash list tables are allocated during start up. Since different applications may only use dedicated limited actions, optimized the hash list table allocate on demand will save initial memory. This commit optimizes hash list table allocate on demand. Signed-off-by: Suanming Mou

[dpdk-dev] [PATCH v3 16/22] net/mlx5: enable index pool per-core cache

This commit enables the tag and header modify action index pool per-core cache in non-reclaim memory mode. Signed-off-by: Suanming Mou Acked-by: Matan Azrad --- drivers/net/mlx5/mlx5.c | 4 +++- drivers/net/mlx5/mlx5.h | 1 + drivers/net/mlx5/mlx5_flow_dv.c | 3 ++- 3 files chan

[dpdk-dev] [PATCH v3 15/22] common/mlx5: allocate cache list memory individually

Currently, the list's local cache instance memory is allocated with the list. As the local cache instance array size is RTE_MAX_LCORE, most of the cases the system will only have very limited cores. allocate the instance memory individually per core will be more economic to the memory. This commit

[dpdk-dev] [PATCH v3 12/22] common/mlx5: add per-lcore cache to hash list utility

From: Matan Azrad Using the mlx5 list utility object in the hlist buckets. This patch moves the list utility object to the common utility, creates all the clone operations for all the hlist instances in the driver. Also adjust all the utility callbacks to be generic for both list and hlist. Si

[dpdk-dev] [PATCH v3 14/22] net/mlx5: adjust the hash bucket size

With the new per core optimization to the list, the hash bucket size can be tuned to a more accurate number. This commit adjusts the hash bucket size. Signed-off-by: Suanming Mou --- drivers/net/mlx5/linux/mlx5_os.c | 2 +- drivers/net/mlx5/mlx5.c | 2 +- drivers/net/mlx5/mlx5_defs.h

[dpdk-dev] [PATCH v3 13/22] net/mlx5: move modify header allocator to ipool

From: Matan Azrad Modify header actions are allocated by mlx5_malloc which has a big overhead of memory and allocation time. One of the action types under the modify header object is SET_TAG, The SET_TAG action is commonly not reused by the flows and each flow has its own value. Hence, the mlx

[dpdk-dev] [PATCH v3 10/22] net/mlx5: relax the list utility atomic operations

From: Matan Azrad The atomic operation in the list utility no need a barriers because the critical part are managed by RW lock. Relax them. Signed-off-by: Matan Azrad Acked-by: Suanming Mou --- drivers/net/mlx5/mlx5_utils.c | 12 ++-- 1 file changed, 6 insertions(+), 6 deletions(-)

[dpdk-dev] [PATCH v3 11/22] net/mlx5: allocate list memory by the create API

From: Matan Azrad Currently, the list memory was allocated by the list API caller. Move it to be allocated by the create API in order to save consistence with the hlist utility. Signed-off-by: Matan Azrad Acked-by: Suanming Mou --- drivers/net/mlx5/linux/mlx5_os.c | 105 +++

[dpdk-dev] [PATCH v3 08/22] net/mlx5: minimize list critical sections

From: Matan Azrad The mlx5 internal list utility is thread safe. In order to synchronize list access between the threads, a RW lock is taken for the critical sections. The create\remove\clone\clone_free operations are in the critical sections. These operations are heavy and make the critical s

[dpdk-dev] [PATCH v3 09/22] net/mlx5: manage list cache entries release

From: Matan Azrad When a cache entry is allocated by lcore A and is released by lcore B, the driver should synchronize the cache list access of lcore A. The design decision is to manage a counter per lcore cache that will be increased atomically when the non-original lcore decreases the referenc

[dpdk-dev] [PATCH v3 06/22] net/mlx5: remove cache term from the list utility

From: Matan Azrad The internal mlx5 list tool is used mainly when the list objects need to be synchronized between multiple threads. The "cache" term is used in the internal mlx5 list API. Next enhancements on this tool will use the "cache" term for per thread cache management. To prevent conf

[dpdk-dev] [PATCH v3 07/22] net/mlx5: add per lcore cache to the list utility

From: Matan Azrad When mlx5 list object is accessed by multiple cores, the list lock counter is all the time written by all the cores what increases cache misses in the memory caches. In addition, when one thread accesses the list for add\remove\lookup operation, all the other threads coming to

[dpdk-dev] [PATCH v3 05/22] net/mlx5: optimize modify header action memory

From: Matan Azrad Define the types of the modify header action fields to be with the minimum size needed for the optional values range. Signed-off-by: Matan Azrad Acked-by: Suanming Mou --- drivers/common/mlx5/linux/mlx5_glue.h | 1 + drivers/net/mlx5/linux/mlx5_flow_os.h | 3 ++- drivers/n

[dpdk-dev] [PATCH v3 04/22] net/mlx5: replace flow list with index pool

The flow list is used to save the create flows and to be used only when port closes all the flows need to be flushed. This commit takes advantage of the index pool foreach operation to flush all the allocated flows. Signed-off-by: Suanming Mou Acked-by: Matan Azrad --- drivers/net/mlx5/linux/m

[dpdk-dev] [PATCH v3 03/22] net/mlx5: add index pool foreach define

In some cases, application may want to know all the allocated index in order to apply some operations to the allocated index. This commit adds the indexed pool functions to support foreach operation. Signed-off-by: Suanming Mou Acked-by: Matan Azrad --- drivers/net/mlx5/mlx5_utils.c | 86 +

[dpdk-dev] [PATCH v3 02/22] net/mlx5: add indexed pool local cache

For object which wants efficient index allocate and free, local cache will be very helpful. Two level cache is introduced to allocate and free the index more efficient. One as local and the other as global. The global cache is able to save all the allocated index. That means all the allocated inde

[dpdk-dev] [PATCH v3 01/22] net/mlx5: allow limiting the index pool maximum index

Some ipool instances in the driver are used as ID\index allocator and added other logic in order to work with limited index values. Add a new configuration for ipool specify the maximum index value. The ipool will ensure that no index bigger than the maximum value is provided. Use this configurat

[dpdk-dev] [PATCH v3 00/22] net/mlx5: insertion rate optimization

This patch series optimize the flow insertion rate with adding local cache to index pool and list. For object which wants efficient index allocate and free, local cache will be very helpful. For index pool, two level cache is added, one as local and another as global. The global cache is able to

[dpdk-dev] [PATCH v2] net/bnxt: fix rxq interrupt setting

Don't set rxq interrupt config Applications can set the rxq interrupt config to 1 or 0 as needed. If an application is not interested in handling Rx interrupts and prefers to poll Rx rings, there is no need for the PMD to set this config option to 1. Fixes: 1fe427fd08ee ("net/bnxt: support enable/

Re: [dpdk-dev] [PATCH] net/bnxt: set rxq interrupt config to 0

On Thu, Jul 1, 2021 at 12:44 PM David Marchand wrote: > On Thu, Jul 1, 2021 at 8:13 PM Ajit Khaparde > wrote: > > > > Set rxq interrupt config to 0 instead of 1. > > Applications can set the rxq interrupt config to 1 or 0 as needed. > > If an application is not interested in handling Rx interrup

Re: [dpdk-dev] [PATCH v5 0/7] Enable ETS-based Tx QoS for VF in DCF

> -Original Message- > From: Xu, Ting > Sent: Thursday, July 1, 2021 6:20 PM > To: dev@dpdk.org > Cc: Zhang, Qi Z ; Wu, Jingjing ; > Xing, Beilei ; Yang, Qiming ; > Xu, Ting > Subject: [PATCH v5 0/7] Enable ETS-based Tx QoS for VF in DCF > > This patch enables the ETS-based Tx QoS for

Re: [dpdk-dev] [PATCH v5 05/24] net/ngbe: add log type and error type

On July 1, 2021 9:57 PM, David Marchand wrote: > Hello, > > Currently looking at new drivers posted on the ml. > > > On Wed, Jun 2, 2021 at 11:40 AM Jiawen Wu > wrote: > > @@ -124,3 +131,12 @@ RTE_PMD_REGISTER_PCI(net_ngbe, > rte_ngbe_pmd); > > RTE_PMD_REGISTER_PCI_TABLE(net_ngbe, pci_id_ngbe_m

Re: [dpdk-dev] [PATCH v2] ifpga/base/meson: fix looking for librt

Hi Tianfei, To reproduce this, assume using Ubuntu 20.04. 1. Install libfdt (It will include ifpga driver from drivers/raw/ifpga/base) $ sudo apt-get install libfdt-dev 2. configure DPDK library $ meson build 3. check libdpdk.pc file in build/meson-private/libdpdk.pc. It will include path

[dpdk-dev] 回复: [PATCH] net/mlx5: fix incorrect r/w lock usage in DMA unmap

Hi, Slava That's OK. Thanks for your reviewing. Best Regards Feifei > -邮件原件- > 发件人: Slava Ovsiienko > 发送时间: 2021年7月1日 22:27 > 收件人: Feifei Wang ; Matan Azrad > ; Shahaf Shuler > 抄送: dev@dpdk.org; nd ; Shahaf Shuler > ; sta...@dpdk.org; Ruifeng Wang > ; nd > 主题: RE: [PATCH] net/mlx5: fi

Re: [dpdk-dev] [PATCH 0/2] provide thread unsafe async registration functions

> -Original Message- > From: Maxime Coquelin > Sent: Thursday, July 1, 2021 11:43 PM > To: Hu, Jiayu ; Maxime Coquelin > ; dev@dpdk.org > Cc: Xia, Chenbo ; Wang, Yinan > ; David Marchand > Subject: Re: [PATCH 0/2] provide thread unsafe async registration functions > > Hi Jiayu, > > On

[dpdk-dev] [PATCH] net/bnxt: set rxq interrupt config to 0

Set rxq interrupt config to 0 instead of 1. Applications can set the rxq interrupt config to 1 or 0 as needed. If an application is not interested in handling Rx interrupts and prefers to poll Rx rings, there is no need for the PMD to set this config option to 1. Signed-off-by: Ajit Khaparde Revi

Re: [dpdk-dev] [PATCH v1] lib/eal: enforce alarm APIs parameters check

On Fri, Jul 02, 2021 at 12:36:45AM +0300, Dmitry Kozlyuk wrote: > 2021-07-01 09:21 (UTC-0700), Tyler Retzlaff: > > On Thu, Jul 01, 2021 at 02:31:29AM +0300, Dmitry Kozlyuk wrote: > > > Hi Jie, > > > > > > 2021-06-23 17:36 (UTC-0700), Jie Zhou: > > > > From: Jie Zhou > > [...] > > > > + /*

Re: [dpdk-dev] [PATCH v1] lib/eal: enforce alarm APIs parameters check

2021-07-01 09:21 (UTC-0700), Tyler Retzlaff: > On Thu, Jul 01, 2021 at 02:31:29AM +0300, Dmitry Kozlyuk wrote: > > Hi Jie, > > > > 2021-06-23 17:36 (UTC-0700), Jie Zhou: > > > From: Jie Zhou > [...] > > > + /* Check if us is valid */ > > > + if (us < 1 || us >(UINT64_MAX - US_PER_S)) { > > >

Re: [dpdk-dev] [PATCH] devtools: recommend new logtype helpers

01/07/2021 16:11, David Marchand: > Following commit eeded2044af5 ("log: register with standardized names"), > the new helpers should be preferred so that we can maintain a consistent > naming for logtypes. > > Signed-off-by: David Marchand Applied, thanks

Re: [dpdk-dev] [PATCH] net/bnxt: set rxq interrupt config to 0

On Thu, Jul 1, 2021 at 8:13 PM Ajit Khaparde wrote: > > Set rxq interrupt config to 0 instead of 1. > Applications can set the rxq interrupt config to 1 or 0 as needed. > If an application is not interested in handling Rx interrupts and > prefers to poll Rx rings, there is no need for the PMD to s

Re: [dpdk-dev] [PATCH 1/2] net/tap: remove useless offload capa functions

> On Jul 1, 2021, at 12:18 PM, Stephen Hemminger > wrote: > > On Thu, 1 Jul 2021 17:16:21 +0300 > Andrew Rybchenko wrote: > >> On 6/16/21 7:15 AM, Stephen Hemminger wrote: >>> Since these always return 0, they were doing nothing useful. >>> >>> Signed-off-by: Stephen Hemminger >> >> I h

Re: [dpdk-dev] [PATCH 1/2] net/tap: remove useless offload capa functions

On Thu, 1 Jul 2021 17:16:21 +0300 Andrew Rybchenko wrote: > On 6/16/21 7:15 AM, Stephen Hemminger wrote: > > Since these always return 0, they were doing nothing useful. > > > > Signed-off-by: Stephen Hemminger > > I have no strong opinion on the patch, but sometimes even > empty functions w

[dpdk-dev] [PATCH v2 2/2] net/cxgbe: remove queue stats from basic stats

Remove queue stats from basic stats because they're now available via xstats API. Also remove RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS flag. Signed-off-by: Rahul Lakkireddy --- v2: - No change. drivers/net/cxgbe/cxgbe_ethdev.c | 18 +++--- 1 file changed, 3 insertions(+), 15 deletions(-)

[dpdk-dev] [PATCH v2 1/2] net/cxgbe: add support for xstats API

Add support to fetch port and queue stats via xstats API. Signed-off-by: Rahul Lakkireddy --- v2: - Perform explicit checks for NULL in conditions. - Use rte_strlcpy() instead of strcpy(). drivers/net/cxgbe/cxgbe_ethdev.c | 303 ++- 1 file changed, 294 insertions(+),

[dpdk-dev] [PATCH v2 0/2] net/cxgbe: add support for xstats API

This series of patches add support to fetch port and queue stats via xstats API. Patch 1 adds support to fetch port and queue stats via xstats API. Patch 2 removes queue stats from basic stats since they're available via xstats API. Also removes RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS flag. --- v2: -

Re: [dpdk-dev] dmadev discussion summary

On Thu, Jul 01, 2021 at 08:31:00PM +0530, Jerin Jacob wrote: > On Sat, Jun 26, 2021 at 9:29 AM fengchengwen wrote: > > > > Hi, all > > I analyzed the current DPAM DMA driver and drew this summary in > > conjunction > > with the previous discussion, and this will as a basis for the V2 > > imple

[dpdk-dev] DPDK Release Status Meeting 01/07/2021

Release status meeting minutes {Date} = :Date: 1 July 2021 :toc: .Agenda: * Release Dates * Subtrees * Roadmaps * LTS * Defects * Opens .Participants: * Broadcom * Canonical * Debian/Microsoft * Intel * Marvell * Nvidia * Red Hat Release Dates -

Re: [dpdk-dev] [PATCH] net/memif: replace memcpy() with rte_memcpy() to improve perf

On 6/23/21 8:03 PM, Stephen Hemminger wrote: > On Wed, 23 Jun 2021 09:49:35 -0700 > David Christensen wrote: > >> Replacing memcpy() with rte_memcpy() improved 64 byte packet >> performance by 33% on a POWER9 system and by 10% on an x86_64 >> system. > > I see rte_memcpy was already used in the

Re: [dpdk-dev] [PATCH v1] lib/eal: enforce alarm APIs parameters check

On Thu, Jul 01, 2021 at 02:31:29AM +0300, Dmitry Kozlyuk wrote: > Hi Jie, > > 2021-06-23 17:36 (UTC-0700), Jie Zhou: > > From: Jie Zhou > > > > lib/eal alarm APIs rte_eal_alarm_set and rte_eal_alarm_cancel > > on Windows do not check parameters to fail fast for invalid > > parameters, which capt

Re: [dpdk-dev] [PATCH] net/memif: replace memcpy() with rte_memcpy() to improve perf

On 6/24/21 7:15 PM, Tyler Retzlaff wrote: > On Wed, Jun 23, 2021 at 01:31:54PM -0700, David Christensen wrote: >> Replacing memcpy() with rte_memcpy() improved 64 byte packet >> performance by 33% on a POWER9 system and by 10% on an x86_64 >> system. >> >> Signed-off-by: David Christensen >> --- >

Re: [dpdk-dev] 20.11.2 patches review and test

Hi, > -Original Message- > From: Xueming(Steven) Li > Sent: Sunday, June 27, 2021 2:28 AM > To: sta...@dpdk.org > Cc: dev@dpdk.org; Abhishek Marathe ; > Akhil Goyal ; Ali Alnubani ; > benjamin.wal...@intel.com; David Christensen ; > hariprasad.govindhara...@intel.com; Hemant Agrawal > ; I

Re: [dpdk-dev] 19.11.9 patches review and test

Hi, > -Original Message- > From: Christian Ehrhardt > Sent: Thursday, June 17, 2021 9:38 AM > To: sta...@dpdk.org > Cc: dev@dpdk.org; Abhishek Marathe ; > Akhil Goyal ; Ali Alnubani ; > benjamin.wal...@intel.com; David Christensen ; > hariprasad.govindhara...@intel.com; Hemant Agrawal > ;

Re: [dpdk-dev] [PATCH 0/2] provide thread unsafe async registration functions

Hi Jiayu, On 6/29/21 7:36 AM, Hu, Jiayu wrote: > Hi Maxime, > >> -Original Message- >> From: Maxime Coquelin >> Sent: Monday, June 7, 2021 9:20 PM >> To: Hu, Jiayu ; dev@dpdk.org >> Cc: maxime.coque...@redhat.com; Xia, Chenbo ; >> Wang, Yinan >> Subject: Re: [PATCH 0/2] provide thread u

Re: [dpdk-dev] [PATCH 2/2] net/cxgbe: remove queue stats from basic stats

Hi Rahul, On 7/1/21 6:26 PM, Rahul Lakkireddy wrote: > Hi Andrew, > > On Thursday, July 07/01/21, 2021 at 17:50:19 +0300, Andrew Rybchenko wrote: >> On 6/3/21 6:30 PM, Rahul Lakkireddy wrote: >>> Remove queue stats from basic stats because they're now available >>> via xstats API. Also remove RTE

Re: [dpdk-dev] [PATCH 2/2] net/cxgbe: remove queue stats from basic stats

Hi Andrew, On Thursday, July 07/01/21, 2021 at 17:50:19 +0300, Andrew Rybchenko wrote: > On 6/3/21 6:30 PM, Rahul Lakkireddy wrote: > > Remove queue stats from basic stats because they're now available > > via xstats API. Also remove RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS flag. > > > > Signed-off-by:

Re: [dpdk-dev] [PATCH] ethdev: fix doc comment for struct rte_flow_action

On 6/25/21 9:58 AM, Singh, Aman Deep wrote: > > On 6/23/2021 7:43 PM, Andrew Rybchenko wrote: >> On 6/23/21 4:39 PM, Ori Kam wrote: >>> Hi Jan, >>> -Original Message- From: Jan Viktorin Sent: Wednesday, June 23, 2021 12:18 PM The struct rte_flow_action was missing

Re: [dpdk-dev] [PATCH v1] doc: policy on promotion of experimental APIs

On Thu, Jul 01, 2021 at 11:19:27AM +0100, Kinsella, Ray wrote: > > > On 30/06/2021 20:56, Tyler Retzlaff wrote: > > On Tue, Jun 29, 2021 at 07:38:05PM +0100, Kinsella, Ray wrote: > >> > >> > +Promotion to stable > +~~~ > + > +Ordinarily APIs marked as ``experim

Re: [dpdk-dev] [PATCH] net/hns3: fix traffic management

On 6/21/21 10:38 AM, Min Hu (Connor) wrote: > From: Huisong Li > > In a multi-TC scenario, if the length of packets destined for different > TCs is different, for example, 64B and 1500B packets destined for TC0 and > TC1 respectively. There is a problem that the bandwidth of the TC to which > lar

Re: [dpdk-dev] dmadev discussion summary

On Sat, Jun 26, 2021 at 9:29 AM fengchengwen wrote: > > Hi, all > I analyzed the current DPAM DMA driver and drew this summary in conjunction > with the previous discussion, and this will as a basis for the V2 > implementation. > Feedback is welcome, thanks Thanks for the write-up. > > dpaa

Re: [dpdk-dev] [PATCH] net/memif: fix abstract socket addr_len

On 6/17/21 7:22 PM, Nathan Skrzypczak wrote: > This fixes using abstract sockets with memifs. > we were not passing the exact addr_len, which > requires zeroing the remaining sun_path and > doesn't appear well in other utilities (e.g. > lsof -U) > > Signed-off-by: Nathan Skrzypczak > --- > drive

Re: [dpdk-dev] [PATCH] eal: create runtime dir even when shared data is not used

On Thu, Jul 01, 2021 at 04:43:48PM +0200, Morten Brørup wrote: > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Bruce Richardson > > Sent: Thursday, 1 July 2021 11.35 > > > > When multi-process is not wanted and DPDK is run with the "no-shconf" > > flag, the telemetry library still needs a

Re: [dpdk-dev] [PATCH 2/2] net/cxgbe: remove queue stats from basic stats

On 6/3/21 6:30 PM, Rahul Lakkireddy wrote: > Remove queue stats from basic stats because they're now available > via xstats API. Also remove RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS flag. > > Signed-off-by: Rahul Lakkireddy > --- > drivers/net/cxgbe/cxgbe_ethdev.c | 18 +++--- > 1 file chan

Re: [dpdk-dev] [PATCH 1/2] net/cxgbe: add support for xstats API

On 6/3/21 6:30 PM, Rahul Lakkireddy wrote: > Add support to fetch port and queue stats via xstats API. > > Signed-off-by: Rahul Lakkireddy [snip] > + count = 0; > + xstats_str = cxgbe_dev_port_stats_strings; > + for (i = 0; i < CXGBE_NB_PORT_STATS; i++, count++) { > + if

Re: [dpdk-dev] [PATCH v1] doc: policy on promotion of experimental APIs

On Thu, Jul 01, 2021 at 08:56:22AM +0100, Ferruh Yigit wrote: > On 6/30/2021 8:56 PM, Tyler Retzlaff wrote: > > On Tue, Jun 29, 2021 at 07:38:05PM +0100, Kinsella, Ray wrote: > >> > >> > +Promotion to stable > +~~~ > + > +Ordinarily APIs marked as ``experimental`

Re: [dpdk-dev] [PATCH] eal: create runtime dir even when shared data is not used

> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Bruce Richardson > Sent: Thursday, 1 July 2021 11.35 > > When multi-process is not wanted and DPDK is run with the "no-shconf" > flag, the telemetry library still needs a runtime directory to place > the > unix socket for telemetry connections

Re: [dpdk-dev] [PATCH v2 2/2] net/nfp: fix PF secondary process probing

On 6/9/21 5:08 PM, Heinrich Kuhn wrote: > This patch creates a new function for handling PF probing of a secondary > process. A CPP handle is obtained for the CPP bridge service and the > service itself is also registered during secondary process > initialization. DPDK services aren't shared betwee

Re: [dpdk-dev] [PATCH v2 1/2] net/nfp: improve PF probing logic

On 6/9/21 5:08 PM, Heinrich Kuhn wrote: > When using rte_eth_dev_pci_generic_probe() during probing a > rte_eth_dev will be created with the name field corresponding to the PCI > address of the device. NFP4000/6000 devices only have a single PF (but > potentially multiple physical ports). This mean

Re: [dpdk-dev] [PATCH] net/mlx5: fix incorrect r/w lock usage in DMA unmap

Hi, Feifei Sorry for the delayed review. I think it is a good catch, thank you for the patch. Acked-by: Viacheslav Ovsiienko With best regards, Slava > -Original Message- > From: Feifei Wang > Sent: Tuesday, June 22, 2021 4:54 > To: Feifei Wang ; Matan Azrad > ; Shahaf Shuler ; Slava

Re: [dpdk-dev] [PATCH v3] ethdev: add IPv4 and L4 checksum RSS offload types

On 6/15/21 11:19 AM, Alvin Zhang wrote: > This patch defines new RSS offload types for IPv4 and L4 checksum, > which are required when users want to distribute packets based on the > IPv4 or L4 checksum field. > > For example "flow create 0 ingress pattern eth / ipv4 / end > actions rss types ipv4

Re: [dpdk-dev] [PATCH 2/2] net/tap: replace offload_capa function with define

On 6/16/21 7:15 AM, Stephen Hemminger wrote: > Since the offload values are always the same, these can > just be data instead of code. > > Signed-off-by: Stephen Hemminger No strong opinion as well, but above reason is not strong enough to touch the code. So, I'll wait for the maintainer here as

Re: [dpdk-dev] [PATCH 1/2] net/tap: remove useless offload capa functions

On 6/16/21 7:15 AM, Stephen Hemminger wrote: > Since these always return 0, they were doing nothing useful. > > Signed-off-by: Stephen Hemminger I have no strong opinion on the patch, but sometimes even empty functions with comments add value. So, I see no point to touch it. So, I'll wait for ma

[dpdk-dev] [PATCH] devtools: recommend new logtype helpers

Following commit eeded2044af5 ("log: register with standardized names"), the new helpers should be preferred so that we can maintain a consistent naming for logtypes. Signed-off-by: David Marchand --- devtools/checkpatches.sh | 8 1 file changed, 8 insertions(+) diff --git a/devtools/c

Re: [dpdk-dev] [PATCH v6] ethdev: add new ext hdr for gtp psc

Hi Raslan, could you reply, please. Andrew. On 6/22/21 10:27 AM, Singh, Aman Deep wrote: > Hi Raslan, > > Can you please provide link to this RFC 38415-g30 > I just had some doubt on byte-order conversion as per RFC 1700 > > > Regards > Aman

Re: [dpdk-dev] [PATCH] app/testpmd: fix offloads for the newly attached port

On 6/19/21 6:40 PM, Viacheslav Ovsiienko wrote: > For the newly attached ports (with "port attach" command) the > default offloads settings, configured from application command > line, were not applied, causing port start failure following > the attach. For example, if scattering offload was config

Re: [dpdk-dev] [PATCH v5 01/15] drivers: introduce mlx5 crypto PMD

On Thu, Jul 1, 2021 at 3:27 PM Shiri Kuzin wrote: [snip] > +RTE_INIT(rte_mlx5_crypto_init) > +{ > + mlx5_common_init(); > + if (mlx5_glue != NULL) > + mlx5_pci_driver_register(&mlx5_crypto_driver); > +} > + > +RTE_PMD_REGISTER_CRYPTO_DRIVER(mlx5_cryptodev_driver, mlx5_d

Re: [dpdk-dev] [PATCH v5 05/24] net/ngbe: add log type and error type

Hello, Currently looking at new drivers posted on the ml. On Wed, Jun 2, 2021 at 11:40 AM Jiawen Wu wrote: > @@ -124,3 +131,12 @@ RTE_PMD_REGISTER_PCI(net_ngbe, rte_ngbe_pmd); > RTE_PMD_REGISTER_PCI_TABLE(net_ngbe, pci_id_ngbe_map); > RTE_PMD_REGISTER_KMOD_DEP(net_ngbe, "* igb_uio | uio_pci_g

Re: [dpdk-dev] [PATCH v16 0/9] app/testpmd: enable testpmd on Windows

On 6/29/21 11:50 PM, Jie Zhou wrote: > 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

Re: [dpdk-dev] [PATCH v16 5/9] app/testpmd: resolve name collisions

On 6/29/21 11:50 PM, Jie Zhou wrote: > Resolve name collisions with Windows types > > Signed-off-by: Jie Zhou > Acked-by: Tal Shnaiderman > Acked-by: Dmitry Kozlyuk I'll fix long lines checkpatches.sh warnings on apply.

[dpdk-dev] basicfwd

Hello, I tried to run a sample program (Skeleton) after installing DPDK according to the Getting Started Guide for Linux document, but failed, "Detected shared linkage of DPDK" is missing in the output of EAL initialization, compared with the correct output result. And the rte_eth_dev_count_ava

Re: [dpdk-dev] [PATCH v3 19/20] net/sfc: support flow action COUNT in transfer rules

On Thu, Jul 01, 2021 at 04:05:56PM +0300, Andrew Rybchenko wrote: > @Bruce, see below. > > On 7/1/21 3:34 PM, David Marchand wrote: > > On Thu, Jul 1, 2021 at 11:22 AM Andrew Rybchenko > > wrote: > >> The build works fine for me on FC34, but it has > >> libatomic-11.1.1-3.fc34.x86_64 installed. >

Re: [dpdk-dev] [PATCH v16 6/9] app/testpmd: fix parse_fec_mode return type name

On 6/29/21 11:50 PM, Jie Zhou wrote: > Passing an uint32_t pointer to an enum pointer parameter causes > pointer-sign warning on Windows (converts between pointers to > integer types with different sign), since enum is implicitly > converted to int on Windows. > > And the current enum pointer para

[dpdk-dev] [PATCH v5 14/15] test/crypto: add mlx5 crypto driver

In order to test the new mlx5 crypto PMD, the driver is added to the crypto test application. Added mlx5 vectors that include 4 testing vectors with length 512 and 4096. Added mlx5 encryption function and decryption function that will both use the mlx5 vectors that will also set the dataunit_len

[dpdk-dev] [PATCH v5 13/15] crypto/mlx5: add statistic get and reset operations

From: Suanming Mou This commit adds mlx5 crypto statistic get and reset operations. Signed-off-by: Suanming Mou Signed-off-by: Matan Azrad --- drivers/crypto/mlx5/mlx5_crypto.c | 40 --- 1 file changed, 37 insertions(+), 3 deletions(-) diff --git a/drivers/crypto/

[dpdk-dev] [PATCH v5 15/15] test/crypto: add mlx5 multi segment tests

The crypto mlx5 driver supports multi segment encryption and decryption operations. Added mlx5 multi segment encryption function and multi segment decryption function that will both use the mlx5 vectors. The added tests will test both data integrity and correct stat values. Signed-off-by: Shiri

[dpdk-dev] [PATCH v5 12/15] crypto/mlx5: add enqueue and dequeue operations

From: Suanming Mou The crypto operations are done with the WQE set which contains one UMR WQE and one rdma write WQE. Most segments of the WQE set are initialized properly during queue setup, only limited segments are initialized according to the crypto detail in the datapath process. This commi

[dpdk-dev] [PATCH v5 11/15] crypto/mlx5: add WQE set initialization

From: Suanming Mou Currently, HW handles the WQEs much faster than the software, Using the constant WQE set layout can initialize most of the WQE segments in advanced, and software only needs to configure very limited segments in datapath. This accelerates the software WQE organize in datapath.

[dpdk-dev] [PATCH v5 10/15] crypto/mlx5: add maximum segments devarg

From: Suanming Mou The mlx5 HW crypto operations are done by attaching crypto property to a memory region. Once done, every access to the memory via the crypto-enabled memory region will result with in-line encryption or decryption of the data. As a result, the design choice is to provide two ty

[dpdk-dev] [PATCH v5 09/15] crypto/mlx5: add keytag devarg

From: Suanming Mou A keytag is a piece of data encrypted together with a DEK. When a DEK is referenced by an MKEY.bsf through its index, the keytag is also supplied in the BSF as plaintext. The HW will decrypt the DEK (and the attached keytag) and will fail the operation if the keytags don't mat

[dpdk-dev] [PATCH v5 08/15] crypto/mlx5: create login object using DevX

To work with crypto engines that are marked with wrapped_import_method, a login session is required. A crypto login object needs to be created using DevX. The crypto login object contains: - The credential pointer. - The import_KEK pointer to be used for all secured information

[dpdk-dev] [PATCH v5 07/15] crypto/mlx5: add memory region management

Mellanox user space drivers don't deal with physical addresses as part of a memory protection mechanism. The device translates the given virtual address to a physical address using the given memory key as an address space identifier. That's why any mbuf virtual address is moved directly to the HW d

[dpdk-dev] [PATCH v5 06/15] crypto/mlx5: add dev stop and start operations

Add the dev_start function that is used to start a configured device. Add the dev_stop function that is used to stop a configured device. Both functions set the dev parameter as used and return 0. Signed-off-by: Shiri Kuzin Acked-by: Matan Azrad --- drivers/crypto/mlx5/mlx5_crypto.c | 17 +

[dpdk-dev] [PATCH v5 05/15] crypto/mlx5: add queue pairs operations

The HW queue pairs are a pair of send queue and receive queue of independent work queues packed together in one object for the purpose of transferring data between nodes of a network. Completion Queue is a FIFO queue of completed work requests. In crypto driver we use one QP in loopback in order

[dpdk-dev] [PATCH v5 04/15] crypto/mlx5: add basic operations

The basic dev control operations are configure, close and get info. Extended the existing support of configure and close: -mlx5_crypto_dev_configure- function used to configure device. -mlx5_crypto_dev_close- function used to close a configured device. Added config struc

[dpdk-dev] [PATCH v5 03/15] crypto/mlx5: add session operations

Sessions are used in symmetric transformations in order to prepare objects and data for packet processing stage. A mlx5 session includes iv_offset, pointer to mlx5_crypto_dek struct, bsf_size, bsf_p_type, block size index, encryption_order and encryption standard. Implement the next session opera

[dpdk-dev] [PATCH v5 02/15] crypto/mlx5: add DEK object management

A DEK(Data encryption Key) is an mlx5 HW object which represents the cipher algorithm key. The DEKs are used during data encryption/decryption operations. In symmetric algorithms like AES-STS, we use the same DEK for both encryption and decryption. Use the mlx5 hash-list tool to manage the DEK ob

[dpdk-dev] [PATCH v5 01/15] drivers: introduce mlx5 crypto PMD

Add a new PMD for Mellanox devices- crypto PMD. The crypto PMD will be supported starting Nvidia ConnectX6 and BlueField2. The crypto PMD will add the support of encryption and decryption using the AES-XTS symmetric algorithm. The crypto PMD requires rdma-core and uses mlx5 DevX. This patch add

[dpdk-dev] [PATCH v5 00/15] drivers: introduce mlx5 crypto PMD

Add a new PMD for Nvidia devices- crypto PMD. The crypto PMD will be supported on Nvidia ConnectX6. The crypto PMD will add the support of encryption and decryption using the AES-XTS symmetric algorithm. The crypto PMD requires rdma-core and uses mlx5 DevX. v2: Add data-path part. v3: Rebase. v4

Re: [dpdk-dev] [PATCH] kni: fix compilation on SLES15-SP3

On Thu, Jul 1, 2021 at 10:23 AM Christian Ehrhardt wrote: > > On Thu, Jun 17, 2021 at 10:25 AM Marco Varlese wrote: > > > > Hello, > > > > On 6/17/21 8:41 AM, Thomas Monjalon wrote: > > > 17/06/2021 08:14, Christian Ehrhardt: > > >> On Thu, Jun 10, 2021 at 12:30 PM Christian Ehrhardt > > >> wrot

Re: [dpdk-dev] [PATCH v2] ifpga/base/meson: fix looking for librt

> -Original Message- > From: Xu, Rosen > Sent: 2021年6月30日 18:03 > To: Hussin, Mohamad Noor Alim ; > Zhang, Tianfei > Cc: dev@dpdk.org; Huang, Wei ; sta...@dpdk.org > Subject: RE: [PATCH v2] ifpga/base/meson: fix looking for librt > > CC Tianfei, who is maintainer. > > > -Original

Re: [dpdk-dev] [PATCH v3 19/20] net/sfc: support flow action COUNT in transfer rules

@Bruce, see below. On 7/1/21 3:34 PM, David Marchand wrote: > On Thu, Jul 1, 2021 at 11:22 AM Andrew Rybchenko > wrote: >> The build works fine for me on FC34, but it has >> libatomic-11.1.1-3.fc34.x86_64 installed. > > I first produced the issue on my "old" FC32. > Afaics, for FC33 and later, g

Re: [dpdk-dev] [PATCH] net/mlx5: remove unwanted barrier

Hi, Honnappa The rte_io_rmb() was inserted not to prevent the extra access to cqe->op_own (the volatile qualifier is quite enough, if we had some doubts, we would insert rte_compiler_barrier), but the real intention of io_rmw was to isolate cqe->op_own loads on hardware level. cqe points to the

[dpdk-dev] [pull-request] dpdk-next-eventdev - v21.08 - RC1

https://patches.dpdk.org/user/todo/dpdk/?series=17514 series pushed to RC2 The following changes since commit 568d97c09ca00e52fd7805ef0cab522250dac18b:   common/mlx5: fix Netlink port name padding in probing (2021-06-24 13:19:54 +0200) are available in the Git repository at:   http://dpdk.org

Re: [dpdk-dev] [PATCH v2] net/i40e: fix data path corrupt on secondary process

> -Original Message- > From: Yu, DapengX > Sent: Monday, June 21, 2021 3:24 PM > To: Xing, Beilei > Cc: dev@dpdk.org; Zhang, Qi Z ; Yu, DapengX > ; sta...@dpdk.org > Subject: [PATCH v2] net/i40e: fix data path corrupt on secondary process > > From: Dapeng Yu > > The rte_eth_devices

  1   2   >