Re: [dpdk-dev] [PATCH v3] lib/rte_rib6: fix stack buffer overflow

On Mon, Jun 21, 2021 at 3:28 PM wrote: > > From: Owen Hilyard Hi Owen, Vladimir, Owen, two comments on the patch title. - We (try to) never prefix with lib/, as it gives no additional info. The prefix should be the library name. There were some transgressions to this rule, but this was Thomas

Re: [dpdk-dev] [PATCH] lib/flow_classify: fix leaking rules on delete

On Wed, Jun 16, 2021 at 9:57 PM wrote: > > From: Owen Hilyard > > Rules in a classify table were not freed if the table > had a delete function. > > Fixes: be41ac2a3 ("flow_classify: introduce flow classify library") Cc: sta...@dpdk.org > > Signed-off-by: Owen Hilyard > --- > lib/flow_classify

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

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] kni: fix wrong mbuf alloc count in kni_allocate_mbufs

> -Original Message- > From: Ferruh Yigit [mailto:ferruh.yi...@intel.com] > Sent: Monday, June 21, 2021 7:26 PM > To: wangyunjian ; dev@dpdk.org > Cc: liucheng (J) ; dingxiaoxiong > > Subject: Re: [dpdk-dev] [PATCH] kni: fix wrong mbuf alloc count in > kni_allocate_mbufs > > On 6/21/2021

Re: [dpdk-dev] [RFC PATCH v3 1/3] sched: add PIE based congestion management

> -Original Message- > From: Stephen Hemminger > Sent: Monday, June 21, 2021 8:18 PM > To: Liguzinski, WojciechX > Cc: dev@dpdk.org; Singh, Jasvinder ; Dumitrescu, > Cristian ; Dharmappa, Savinay > ; Ajmera, Megha > Subject: Re: [dpdk-dev] [RFC PATCH v3 1/3] sched: add PIE based con

Re: [dpdk-dev] [PATCH] kni: fix wrong mbuf alloc count in kni_allocate_mbufs

On 6/22/2021 8:32 AM, wangyunjian wrote: >> -Original Message- >> From: Ferruh Yigit [mailto:ferruh.yi...@intel.com] >> Sent: Monday, June 21, 2021 7:26 PM >> To: wangyunjian ; dev@dpdk.org >> Cc: liucheng (J) ; dingxiaoxiong >> >> Subject: Re: [dpdk-dev] [PATCH] kni: fix wrong mbuf alloc

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

Acked-by: Aman Deep Singh

Re: [dpdk-dev] [PATCH] parray: introduce internal API for dynamic arrays

> > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ananyev, > > Konstantin > > > > > > > > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ananyev, > > > > Konstantin > > > > > > > > > > How can we hide the callbacks since they are used by inline > > burst > > > > functions. > > > >

Re: [dpdk-dev] [PATCH v1 4/7] power: remove thread safety from PMD power API's

> Currently, we expect that only one callback can be active at any given > moment, for a particular queue configuration, which is relatively easy > to implement in a thread-safe way. However, we're about to add support > for multiple queues per lcore, which will greatly increase the > possibility

Re: [dpdk-dev] [PATCH v1 1/2] devtools: add relative path support for ABI compatibility check

On Tue, Jun 01, 2021 at 09:56:52AM +0800, Feifei Wang wrote: > From: Phil Yang > > Because dpdk guide does not limit the relative path for ABI > compatibility check, users maybe set 'DPDK_ABI_REF_DIR' as a relative > path: > > ~/dpdk/devtools$ DPDK_ABI_REF_VERSION=v19.11 DPDK_ABI_REF_DIR=build-g

[dpdk-dev] [PATCH 0/3] net/bonding: make dedicated queues work with mlx5

This patchset fixes the inability to use dedicated queues on mlx5 PMD due to RTE Flow rule attempted creation prior to starting the device. Missing return value check and copy paste error near the rule creation have also been fixed. Cc: Chas Williams Cc: "Min Hu (Connor)" Cc: Declan Doherty Cc:

[dpdk-dev] [PATCH 1/3] net/bonding: fix proper return value check and correct log message

Return value is now saved to errval and log message on error reports correct function name, doesn't use q_id which was out of context, and uses up-to-date errval. Fixes: 112891cd27e5 ("net/bonding: add dedicated HW queues for LACP control") Cc: tomaszx.kula...@intel.com Signed-off-by: Martin Havl

[dpdk-dev] [PATCH 2/3] net/bonding: fix not checked return value

Return value from bond_ethdev_8023ad_flow_set() is now checked and appropriate message is logged on error. Fixes: 112891cd27e5 ("net/bonding: add dedicated HW queues for LACP control") Cc: tomaszx.kula...@intel.com Signed-off-by: Martin Havlik Cc: Jan Viktorin --- drivers/net/bonding/rte_eth_b

[dpdk-dev] [PATCH 3/3] net/bonding: start ethdev prior to setting 8023ad flow

When dedicated queues are enabled, mlx5 PMD fails to install RTE Flows if the underlying ethdev is not started: bond_ethdev_8023ad_flow_set(267) - bond_ethdev_8023ad_flow_set: port not started (slave_port=0 queue_id=1) Signed-off-by: Martin Havlik Cc: Jan Viktorin --- drivers/net/bonding/rte_e

Re: [dpdk-dev] [PATCH v1 5/7] power: support callbacks for multiple Rx queues

> Currently, there is a hard limitation on the PMD power management > support that only allows it to support a single queue per lcore. This is > not ideal as most DPDK use cases will poll multiple queues per core. > > The PMD power management mechanism relies on ethdev Rx callbacks, so it > is v

[dpdk-dev] 回复: [PATCH v1 1/2] net/i40e: improve performance for scalar Tx

Hi, Beilei Thanks for your comments, please see below. > -邮件原件- > 发件人: Xing, Beilei > 发送时间: 2021年6月22日 14:08 > 收件人: Feifei Wang > 抄送: dev@dpdk.org; nd ; Ruifeng Wang > > 主题: RE: [PATCH v1 1/2] net/i40e: improve performance for scalar Tx > > > > > -Original Message- > > From:

Re: [dpdk-dev] [PATCH] parray: introduce internal API for dynamic arrays

> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ananyev, > Konstantin > > > > > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ananyev, > > > Konstantin > > > > > > > > > > > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ananyev, > > > > > Konstantin > > > > > > > > > > > >

[dpdk-dev] 回复: [PATCH v1 1/2] net/i40e: improve performance for scalar Tx

Sorry for a mistake for the code, it should be: int n = txq->tx_rs_thresh; int32_t i = 0, j = 0; const int32_t k = RTE_ALIGN_FLOOR(n, RTE_I40E_TX_MAX_FREE_BUF_SZ); const int32_t m = n % RTE_I40E_TX_MAX

[dpdk-dev] [PATCH v5] devtools: script to track map symbols

Script to track growth of stable and experimental symbols over releases since v19.11. Signed-off-by: Ray Kinsella --- v2: reworked to fix pylint errors v3: sent with the correct in-reply-to v4: fix typos picked up by the CI v5: fix terminal_size & directory args devtools/count_symbols.py | 262

Re: [dpdk-dev] [PATCH v3] lib/rte_rib6: fix stack buffer overflow

Hi Owen, David, Apart from David's comments looks good to me. On 22/06/2021 10:10, David Marchand wrote: On Mon, Jun 21, 2021 at 3:28 PM wrote: From: Owen Hilyard Hi Owen, Vladimir, Owen, two comments on the patch title. - We (try to) never prefix with lib/, as it gives no additional

[dpdk-dev] [PATCH v2] kni: fix wrong mbuf alloc count in kni_allocate_mbufs

From: Yunjian Wang In kni_allocate_mbufs(), we alloc mbuf for alloc_q as this code. allocq_free = (kni->alloc_q->read - kni->alloc_q->write - 1) \ & (MAX_MBUF_BURST_NUM - 1); The value of allocq_free maybe zero, for example : The ring size is 1024. After init, write = read = 0. Th

Re: [dpdk-dev] [PATCH] lib/flow_classify: fix leaking rules on delete

Hi David, Owen, > -Original Message- > From: David Marchand > Sent: Tuesday, June 22, 2021 8:24 AM > To: Owen Hilyard ; Iremonger, Bernard > ; Yigit, Ferruh > Cc: dev > Subject: Re: [PATCH] lib/flow_classify: fix leaking rules on delete > > On Wed, Jun 16, 2021 at 9:57 PM wrote: > > >

Re: [dpdk-dev] [PATCH] parray: introduce internal API for dynamic arrays

> > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ananyev, > > Konstantin > > > > > > > > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ananyev, > > > > Konstantin > > > > > > > > > > > > > > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ananyev, > > > > > > Konstantin

Re: [dpdk-dev] [PATCH v2] kni: fix wrong mbuf alloc count in kni_allocate_mbufs

On 6/22/2021 11:57 AM, wangyunjian wrote: > From: Yunjian Wang > > In kni_allocate_mbufs(), we alloc mbuf for alloc_q as this code. > allocq_free = (kni->alloc_q->read - kni->alloc_q->write - 1) \ > & (MAX_MBUF_BURST_NUM - 1); > The value of allocq_free maybe zero, for example : > T

Re: [dpdk-dev] [PATCH v2] kni: fix wrong mbuf alloc count in kni_allocate_mbufs

> -Original Message- > From: Ferruh Yigit [mailto:ferruh.yi...@intel.com] > Sent: Tuesday, June 22, 2021 8:28 PM > To: wangyunjian ; dev@dpdk.org > Cc: tho...@monjalon.net; gowrishanka...@linux.vnet.ibm.com; > dingxiaoxiong ; sta...@dpdk.org; liucheng (J) > > Subject: Re: [PATCH v2] kni: f

[dpdk-dev] [PATCH] kni: fix crash on userspace VA for segmented packets

When IOVA=VA, address translation for segmented packets is wrong, it assumes the address in the mbuf->next is physical address, not VA address. Fixing the address translation to work both PA & VA mode. Fixes: e73831dc6c26 ("kni: support userspace VA") Cc: sta...@dpdk.org Signed-off-by: Ferruh Yi

[dpdk-dev] [PATCH v2 1/2] power: don't use rte prefix in internal code

From: Anatoly Burakov Currently, ACPI code uses rte_power_info as the struct name, which gives the appearance that this is an externally visible API. Fix to use internal namespace. Signed-off-by: Anatoly Burakov Acked-by: David Hunt --- lib/power/power_acpi_cpufreq.c | 34 +---

[dpdk-dev] [PATCH v2 2/2] power: refactor pstate and acpi code

From: Anatoly Burakov Currently, ACPI and PSTATE modes have lots of code duplication, confusing logic, and a bunch of other issues that can, and have, led to various bugs and resource leaks. This commit factors out the common parts of sysfs reading/writing for ACPI and PSTATE drivers. Signed-of

[dpdk-dev] [PATCH v3] kni: fix mbuf allocation for alloc FIFO

From: Yunjian Wang In kni_allocate_mbufs(), we alloc mbuf for alloc_q as this code. allocq_free = (kni->alloc_q->read - kni->alloc_q->write - 1) \ & (MAX_MBUF_BURST_NUM - 1); The value of allocq_free maybe zero, for example : The ring size is 1024. After init, write = read = 0. Th

[dpdk-dev] [PATCH v5 1/2] power: don't use rte prefix in internal code

From: Anatoly Burakov Currently, ACPI code uses rte_power_info as the struct name, which gives the appearance that this is an externally visible API. Fix to use internal namespace. Signed-off-by: Anatoly Burakov Acked-by: David Hunt --- lib/power/power_acpi_cpufreq.c | 34 +---

[dpdk-dev] [PATCH v5 2/2] power: refactor pstate and acpi code

From: Anatoly Burakov Currently, ACPI and PSTATE modes have lots of code duplication, confusing logic, and a bunch of other issues that can, and have, led to various bugs and resource leaks. This commit factors out the common parts of sysfs reading/writing for ACPI and PSTATE drivers. Signed-of

Re: [dpdk-dev] [PATCH v2 1/2] power: don't use rte prefix in internal code

On 22/6/2021 1:43 PM, David Hunt wrote: From: Anatoly Burakov Currently, ACPI code uses rte_power_info as the struct name, which gives the appearance that this is an externally visible API. Fix to use internal namespace. Signed-off-by: Anatoly Burakov Acked-by: David Hunt --- lib/power/p

Re: [dpdk-dev] [PATCH v2 2/2] power: refactor pstate and acpi code

On 22/6/2021 1:43 PM, David Hunt wrote: From: Anatoly Burakov Currently, ACPI and PSTATE modes have lots of code duplication, confusing logic, and a bunch of other issues that can, and have, led to various bugs and resource leaks. This commit factors out the common parts of sysfs reading/wri

Re: [dpdk-dev] [PATCH] parray: introduce internal API for dynamic arrays

> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ananyev, > Konstantin > > > > > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ananyev, > > > Konstantin > > > > > > > > > > > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ananyev, > > > > > Konstantin > > > > > > > > > > > >

Re: [dpdk-dev] [PATCH v5 2/2] power: refactor pstate and acpi code

Adding people to the CC list that were on v4 of this patch set, and Richael who raised some issues in v4. On 22/6/2021 1:58 PM, David Hunt wrote: From: Anatoly Burakov Currently, ACPI and PSTATE modes have lots of code duplication, confusing logic, and a bunch of other issues that can, and ha

[dpdk-dev] [PATCH 1/1] net/i40e: fix compilation failure on core-avx-i

i40e_rxtx_vec_sse.c fails to compile with below configuration: - #define RTE_LIBRTE_I40E_16BYTE_RX_DESC 1 in config/rte_config.h - cpu=core-axv-i - gcc which supports -mavx2 (e.g. gcc 4.8.5) This is because commit 0604b1f2208f ("net/i40e: fix crash in AVX512") added i40e_rxq_rearm_common() to i40e

Re: [dpdk-dev] [PATCH] net/e1000: fix nic ops function was no initialized in secondary process

On 2021/6/22 上午10:16, Wang, Haiyue wrote: From: 张 杨 Sent: Monday, June 21, 2021 16:35 To: Wang, Haiyue Cc: dev@dpdk.org; Zhang, Qi Z ; Lin, Xueqin Subject: Re: [PATCH] net/e1000: fix nic ops function was no initialized in secondary process 发件人: Wang, Haiyue 发送

[dpdk-dev] [PATCH v2] examples/power: add baseline mode to PMD power

The PMD Power Management scheme currently has 3 modes, scale, monitor and pause. However, it would be nice to have a baseline mode for easy comparison of power savings with and without these modes. This patch adds a 'baseline' mode were the pmd power management is not enabled. Use --pmg-mgmt=basel

[dpdk-dev] [PATCH v3] net/mlx5: add TCP and IPv6 to supported flow items list in Windows

WINOF2 2.70 Windows kernel driver allows DevX rule creation of types TCP and IPv6. Added the types to the supported items in mlx5_flow_os_item_supported to allow them to be created in the PMD. Added description of new rules support in Windows kernel driver WINOF2 2.70 to the mlx5 driver guide. S

[dpdk-dev] [PATCH v1] doc: update ABI in MAINTAINERS file

Update to ABI MAINTAINERS. Signed-off-by: Ray Kinsella --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 5877a16971..dab8883a4f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -117,7 +117,6 @@ F: .ci/ ABI Policy & Versioning M: Ray Kinsella -M:

[dpdk-dev] [PATCH] net/mlx5: fix multi-segment inline for the first segment

If the first segment in the multi-segment packet is short and below the inline threshold it should be inline into the WQE to improve the performance. For example, the T-Rex traffic generator might use small leading segments to handle packet headers and performance was affected. Fixes: cacb44a09962

[dpdk-dev] [PATCH 0/2] OCTEONTX crypto adapter support

Below patches add crypto adapter OP_FORWARD support for OCTEON TX PMD. Shijith Thotton (2): drivers: add octeontx crypto adapter framework drivers: add octeontx crypto adapter data path doc/guides/rel_notes/release_21_08.rst| 4 + drivers/common/cpt/cpt_common.h | 2

[dpdk-dev] [PATCH 1/2] drivers: add octeontx crypto adapter framework

Set crypto adapter event device slow-path call backs. Signed-off-by: Shijith Thotton --- drivers/crypto/octeontx/meson.build | 1 + drivers/crypto/octeontx/otx_cryptodev.c | 4 ++ drivers/crypto/octeontx/otx_cryptodev.h | 4 -- .../crypto/octeontx/otx_cryptodev_hw_access

[dpdk-dev] [PATCH 2/2] drivers: add octeontx crypto adapter data path

Added support for crypto adapter OP_FORWARD mode. As OcteonTx CPT crypto completions could be out of order, each crypto op is enqueued to CPT, dequeued from CPT and enqueued to SSO one-by-one. Signed-off-by: Shijith Thotton --- doc/guides/rel_notes/release_21_08.rst| 4 + drivers/comm

Re: [dpdk-dev] [RFC PATCH] dmadev: introduce DMA device library

On Fri, Jun 18, 2021 at 3:11 PM fengchengwen wrote: > > On 2021/6/18 13:52, Jerin Jacob wrote: > > On Thu, Jun 17, 2021 at 2:46 PM Bruce Richardson > > wrote: > >> > >> On Wed, Jun 16, 2021 at 08:07:26PM +0530, Jerin Jacob wrote: > >>> On Wed, Jun 16, 2021 at 3:47 PM fengchengwen > >>> wrote: >

Re: [dpdk-dev] [RFC PATCH] dmadev: introduce DMA device library

On Fri, Jun 18, 2021 at 3:25 PM Bruce Richardson wrote: > > On Fri, Jun 18, 2021 at 11:22:28AM +0530, Jerin Jacob wrote: > > On Thu, Jun 17, 2021 at 2:46 PM Bruce Richardson > > wrote: > > > > > > On Wed, Jun 16, 2021 at 08:07:26PM +0530, Jerin Jacob wrote: > > > > On Wed, Jun 16, 2021 at 3:47 PM

Re: [dpdk-dev] [RFC PATCH] dmadev: introduce DMA device library

On Fri, Jun 18, 2021 at 3:34 PM Bruce Richardson wrote: > > On Fri, Jun 18, 2021 at 10:46:08AM +0530, Jerin Jacob wrote: > > On Thu, Jun 17, 2021 at 1:30 PM Bruce Richardson > > wrote: > > > > > > On Thu, Jun 17, 2021 at 01:12:22PM +0530, Jerin Jacob wrote: > > > > On Thu, Jun 17, 2021 at 12:43 A

Re: [dpdk-dev] [RFC PATCH] dmadev: introduce DMA device library

On Fri, Jun 18, 2021 at 2:22 PM fengchengwen wrote: > > On 2021/6/17 22:18, Bruce Richardson wrote: > > On Thu, Jun 17, 2021 at 12:02:00PM +0100, Bruce Richardson wrote: > >> On Thu, Jun 17, 2021 at 05:48:05PM +0800, fengchengwen wrote: > >>> On 2021/6/17 1:31, Bruce Richardson wrote: > On We

Re: [dpdk-dev] [RFC PATCH] dmadev: introduce DMA device library

On Tue, Jun 22, 2021 at 11:01:47PM +0530, Jerin Jacob wrote: > On Fri, Jun 18, 2021 at 3:25 PM Bruce Richardson > wrote: > > > > > > > Taking the case of a simple copy op, the parameters we need are: > > > > * src > > * dst > > * length > > OK. Is it the case where no other attribute that support

[dpdk-dev] [PATCH 1/2] common/octeontx2: send link event to VF

Currently link event is only sent to the PF by AF as soon as it comes up, or in case of any physical change in link. PF will broadcast these link events to all its VFs as soon as it receives it. But no event is sent when a new VF comes up, hence it will not have the link status. Adding support for

[dpdk-dev] [PATCH 2/2] net/octeontx2: callback for getting link status

Adding a new callback for reading the link status. PF can read it's link status and can forward the same to VF once it comes up. Signed-off-by: Harman Kalra --- drivers/net/octeontx2/otx2_ethdev.c | 8 +++- drivers/net/octeontx2/otx2_ethdev.h | 2 ++ drivers/net/octeontx2/otx2_link.c | 2

Re: [dpdk-dev] [dpdk-stable] [PATCH v3] kni: fix mbuf allocation for alloc FIFO

22/06/2021 14:44, wangyunjian: > From: Yunjian Wang > > In kni_allocate_mbufs(), we alloc mbuf for alloc_q as this code. > allocq_free = (kni->alloc_q->read - kni->alloc_q->write - 1) \ > & (MAX_MBUF_BURST_NUM - 1); > The value of allocq_free maybe zero, for example : > The ring siz

Re: [dpdk-dev] [PATCH] raw/ioat: fix memory leak in device configure

17/06/2021 16:20, Bruce Richardson: > On Thu, Jun 17, 2021 at 02:17:52PM +, Kevin Laatz wrote: > > During device configure, memory is allocated for "hdl_ring_flags". In the > > event of another call to the device configure function (reconfigure), a > > memory leak would occur. This patch fixes

Re: [dpdk-dev] [dpdk-stable] [PATCH] raw/ioat: fix missing ring pointer reset

17/06/2021 16:21, Bruce Richardson: > On Thu, Jun 17, 2021 at 02:18:15PM +, Kevin Laatz wrote: > > In the event of a device reconfigure, "hdls_avail" is not being reset. This > > can lead to miscalculations in rte_ioat_completed_ops(), causing the > > function to report an incorrect amount of c

Re: [dpdk-dev] rte_memcpy - fence and stream

> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Morten Brørup > Sent: Thursday, 27 May 2021 20.15 > > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Bruce Richardson > > Sent: Thursday, 27 May 2021 19.22 > > > > On Thu, May 27, 2021 at 10:39:59PM +0530, Manish Sharma wrote: > > >

Re: [dpdk-dev] [PATCH v4 2/2] bus/auxiliary: introduce auxiliary bus

> -Original Message- > From: Thomas Monjalon > Sent: Tuesday, June 22, 2021 12:11 AM > To: Parav Pandit ; Xueming(Steven) Li > Cc: dev@dpdk.org; Wang Haiyue ; Kinsella Ray > ; david.march...@redhat.com; > ferruh.yi...@intel.com > Subject: Re: [dpdk-dev] [PATCH v4 2/2] bus/auxiliary: i

[dpdk-dev] [PATCH v5 1/2] devargs: add common key definition

Adds common devargs key definition for "bus", "class" and "driver". Cc: Thomas Monjalon Signed-off-by: Xueming Li --- drivers/common/mlx5/mlx5_common.h | 2 -- drivers/common/mlx5/mlx5_common_pci.c | 2 +- drivers/common/sfc_efx/sfc_efx.c| 7 +++ drivers/common/sfc_efx/sf

[dpdk-dev] [PATCH v5 2/2] bus/auxiliary: introduce auxiliary bus

Auxiliary bus [1] provides a way to split function into child-devices representing sub-domains of functionality. Each auxiliary device represents a part of its parent functionality. Auxiliary device is identified by unique device name, sysfs path: /sys/bus/auxiliary/devices/ Devargs legacy synt

[dpdk-dev] [PATCH v3 0/4] vhost: support async dequeue for split ring

This patch implements asynchronous dequeue data path for split ring. A new asynchronous dequeue function is introduced. With this function, the application can try to receive packets from the guest with offloading large copies to the DMA engine, thus saving precious CPU cycles. v3: - Fix compilati

[dpdk-dev] [PATCH v3 1/4] examples/vhost: refactor vhost enqueue and dequeue datapaths.

Previously, by judging the flag, we call different enqueue/dequeue functions in data path. Now, we use an ops that was initialized when Vhost was created, so that we can call ops directly in Vhost data path without any more flag judgment. Signed-off-by: Wenwu Ma --- examples/vhost/main.c

[dpdk-dev] [PATCH v3 2/4] examples/vhost: use a new API to query remaining ring space

A new API for querying the remaining descriptor ring capacity is available, so we use the new one instead of the old one. Signed-off-by: Wenwu Ma --- examples/vhost/ioat.c | 6 +- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/examples/vhost/ioat.c b/examples/vhost/ioat.c index

[dpdk-dev] [PATCH v3 3/4] vhost: support async dequeue for split ring

From: Yuan Wang This patch implements asynchronous dequeue data path for split ring. A new asynchronous dequeue function is introduced. With this function, the application can try to receive packets from the guest with offloading large copies to the DMA engine, thus saving precious CPU cycles. S

[dpdk-dev] [PATCH v3 4/4] examples/vhost: support vhost async dequeue data path

This patch is to add vhost async dequeue data-path in vhost sample. vswitch can leverage IOAT to accelerate vhost async dequeue data-path. Signed-off-by: Wenwu Ma --- doc/guides/sample_app_ug/vhost.rst | 9 +- examples/vhost/ioat.c | 61 ++--- examples/vhost/ioat.h

Re: [dpdk-dev] [RFC PATCH] dmadev: introduce DMA device library

On 2021/6/23 1:25, Jerin Jacob wrote: > On Fri, Jun 18, 2021 at 3:11 PM fengchengwen wrote: >> >> On 2021/6/18 13:52, Jerin Jacob wrote: >>> On Thu, Jun 17, 2021 at 2:46 PM Bruce Richardson >>> wrote: On Wed, Jun 16, 2021 at 08:07:26PM +0530, Jerin Jacob wrote: > On Wed, Jun 16, 202

Re: [dpdk-dev] [RFC PATCH] dmadev: introduce DMA device library

On 2021/6/23 1:51, Jerin Jacob wrote: > On Fri, Jun 18, 2021 at 2:22 PM fengchengwen wrote: >> >> On 2021/6/17 22:18, Bruce Richardson wrote: >>> On Thu, Jun 17, 2021 at 12:02:00PM +0100, Bruce Richardson wrote: On Thu, Jun 17, 2021 at 05:48:05PM +0800, fengchengwen wrote: > On 2021/6/17

[dpdk-dev] [PATCH v4 0/2] power: add support for cppc cpufreq driver

v4: rebase on Anatoly's patch: http://dpdk.org/patch/94676 Richael Zhuang (2): power: add support for cppc cpufreq test/power: round cpuinfo cur freq only when using CPPC cpufreq app/test/test_power.c | 3 +- app/test/test_power_cpufreq.c | 26 +- lib/power/meson.build

[dpdk-dev] [PATCH v4 1/2] power: add support for cppc cpufreq

Currently in DPDK only acpi_cpufreq and pstate_cpufreq drivers are supported, which are both not available on arm64 platforms. Add support for cppc_cpufreq driver which works on most arm64 platforms. Signed-off-by: Richael Zhuang --- app/test/test_power.c | 3 +- app/test/test_power_c

[dpdk-dev] [PATCH v4 2/2] test/power: round cpuinfo cur freq only when using CPPC cpufreq

On arm platform, the value in "/sys/.../cpuinfo_cur_freq" may not be exactly the same as what was set when using CPPC cpufreq driver. For other cpufreq driver, no need to round it currently, or else this check will fail with turbo enabled. For example, with acpi_cpufreq, cpuinfo_cur_freq can be 240

[dpdk-dev] [PATCH v4 00/62] Marvell CNXK Ethdev Driver

This patchset adds support for Marvell CN106XX SoC based on 'common/cnxk' driver. In future, CN9K a.k.a octeontx2 will also be supported by same driver when code is ready and 'net/octeontx2' will be deprecated. Harman Kalra (1): common/cnxk: allocate lmt region in userspace Jerin Jacob (7): c

[dpdk-dev] [PATCH v4 01/62] common/cnxk: add support to lock NIX RQ contexts

From: Satha Rao This patch will consider device argument to lock rss table in NIX. This patch also adds few misc fixes such as disabling NIX Tx vlan insertion conf in SMQ, enabling SSO in NIX Tx SQ for Tx completions and TM related stats API. Signed-off-by: Satha Rao --- drivers/common/cnxk/r

[dpdk-dev] [PATCH v4 02/62] common/cnxk: fix batch alloc completion poll logic

From: Jerin Jacob The instruction generation was not correct due to fact that volatile suppose to use with ccode variable as well. Change the logic to use gcc atomic builtin to simplify and avoid explicit volatile from the code. Fixes: 81af26789316 ("common/cnxk: support NPA batch alloc/free")

[dpdk-dev] [PATCH v4 03/62] common/cnxk: add support to dump flow entries

From: Satheesh Paul Add NPC support API to dump created flow entries. Signed-off-by: Satheesh Paul --- drivers/common/cnxk/hw/npc.h| 2 + drivers/common/cnxk/meson.build | 1 + drivers/common/cnxk/roc_npc.c | 20 ++ drivers/common/cnxk/roc_npc.h |

[dpdk-dev] [PATCH v4 04/62] common/cnxk: support for mark and flag flow actions

From: Satheesh Paul Add roc API to get mark action. Signed-off-by: Satheesh Paul --- drivers/common/cnxk/roc_npc.c | 17 + drivers/common/cnxk/roc_npc.h | 3 +++ drivers/common/cnxk/version.map | 2 ++ 3 files changed, 22 insertions(+) diff --git a/drivers/common/cnxk/ro

[dpdk-dev] [PATCH v4 05/62] common/cnxk: allocate lmt region in userspace

From: Harman Kalra As per the new LMTST design, userspace shall allocate lmt region, setup the DMA translation and share the IOVA with kernel via MBOX. Kernel will convert this IOVA to physical memory and update the LMT table entry with the same. With this new design also shared mode (i.e. all pc

[dpdk-dev] [PATCH v4 06/62] common/cnxk: add provision to enable RED on RQ

From: Satha Rao Send RED pass/drop levels based on rq configurations to kernel. Fixed the aura and pool shift value calculation. Signed-off-by: Satha Rao --- drivers/common/cnxk/roc_nix.h | 8 ++ drivers/common/cnxk/roc_nix_queue.c | 50 + drivers

[dpdk-dev] [PATCH v4 07/62] common/cnxk: change model API to not use camel case

Change model check API's to not use Camel case in function names. Signed-off-by: Nithin Dabilpuram --- drivers/common/cnxk/roc_model.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/common/cnxk/roc_model.h b/drivers/common/cnxk/roc_model.h index fb774ac..5aaad5

[dpdk-dev] [PATCH v4 08/62] common/cnxk: support for VLAN push and pop flow actions

From: Satheesh Paul Add roc API to configure VLAN tag addition and removal. This patch also adds 98xx support for increased MCAM entries for rte flow. Signed-off-by: Satheesh Paul --- drivers/common/cnxk/roc_model.h| 6 + drivers/common/cnxk/roc_npc.c | 257

[dpdk-dev] [PATCH v4 09/62] net/cnxk: add build infra and common probe

Add build infrastructure and common probe and remove for cnxk driver which is used by both CN10K and CN9K SoC. Signed-off-by: Nithin Dabilpuram --- MAINTAINERS| 5 +- doc/guides/nics/cnxk.rst | 29 + doc/guides/nics/features/cnxk.ini | 9 ++

[dpdk-dev] [PATCH v4 10/62] net/cnxk: add platform specific probe and remove

Add platform specific probe and remove callbacks for CN9K and CN10K which use common probe and remove functions. Register ethdev driver for CN9K and CN10K. Signed-off-by: Nithin Dabilpuram --- drivers/net/cnxk/cn10k_ethdev.c | 64 drivers/net/cnxk/cn10k_ethdev.h

[dpdk-dev] [PATCH v4 11/62] net/cnxk: add common devargs parsing function

Add various devargs parsing command line arguments parsing functions supported by CN9K and CN10K. Signed-off-by: Nithin Dabilpuram --- doc/guides/nics/cnxk.rst | 94 +++ drivers/net/cnxk/cnxk_ethdev.c | 7 ++ drivers/net/cnxk/cnxk_ethdev.h | 9 +

[dpdk-dev] [PATCH v4 12/62] net/cnxk: support common dev infos get

Add support to retrieve dev infos get for CN9K and CN10K. Signed-off-by: Nithin Dabilpuram --- doc/guides/nics/cnxk.rst | 3 ++ doc/guides/nics/features/cnxk.ini | 4 ++ doc/guides/nics/features/cnxk_vec.ini | 4 ++ doc/guides/nics/features/cnxk_vf.ini | 3 ++ drivers/net/c

[dpdk-dev] [PATCH v4 13/62] net/cnxk: add device configuration operation

Add device configuration op for CN9K and CN10K. Most of the device configuration is common between two platforms except for some supported offloads. Signed-off-by: Nithin Dabilpuram --- doc/guides/nics/cnxk.rst | 2 + doc/guides/nics/features/cnxk.ini | 2 + doc/guides/nics/

[dpdk-dev] [PATCH v4 14/62] net/cnxk: support link status update

Add link status update callback to get current link status. Signed-off-by: Nithin Dabilpuram --- doc/guides/nics/cnxk.rst | 1 + doc/guides/nics/features/cnxk.ini | 2 + doc/guides/nics/features/cnxk_vec.ini | 2 + doc/guides/nics/features/cnxk_vf.ini | 2 + drivers/net

[dpdk-dev] [PATCH v4 15/62] net/cnxk: add Rx queue setup and release

Add Rx queue setup and release op for CN9K and CN10K SoC. Release is completely common while setup is platform dependent due to fast path Rx queue structure variation. Fastpath is platform dependent partly due to core cacheline size difference. Signed-off-by: Nithin Dabilpuram --- doc/guides/nic

[dpdk-dev] [PATCH v4 16/62] net/cnxk: add Tx queue setup and release

aDD tx queue setup and release for CN9K and CN10K. Release is common while setup is platform dependent due to differences in fast path Tx queue structures. Signed-off-by: Nithin Dabilpuram --- doc/guides/nics/features/cnxk.ini | 1 + doc/guides/nics/features/cnxk_vec.ini | 1 + doc/guides/

[dpdk-dev] [PATCH v4 17/62] net/cnxk: support packet type

Add support for packet type lookup on Rx to translate HW specific types to RTE_PTYPE_* defines Signed-off-by: Nithin Dabilpuram --- doc/guides/nics/cnxk.rst | 1 + doc/guides/nics/features/cnxk.ini | 1 + doc/guides/nics/features/cnxk_vec.ini | 1 + doc/guides/nics/featur

[dpdk-dev] [PATCH v4 18/62] net/cnxk: support queue start and stop

Add Rx/Tx queue start and stop callbacks for CN9K and CN10K. Signed-off-by: Nithin Dabilpuram --- doc/guides/nics/features/cnxk.ini | 1 + doc/guides/nics/features/cnxk_vec.ini | 1 + doc/guides/nics/features/cnxk_vf.ini | 1 + drivers/net/cnxk/cn10k_ethdev.c | 16 ++ drivers/n

[dpdk-dev] [PATCH v4 19/62] net/cnxk: add Rx burst for cn9k

From: Jerin Jacob Add Rx burst scalar version for CN9K. Signed-off-by: Jerin Jacob --- drivers/net/cnxk/cn9k_ethdev.h | 3 + drivers/net/cnxk/cn9k_rx.c | 46 drivers/net/cnxk/cn9k_rx.h | 237 + drivers/net/cnxk/cnxk_ethdev.h | 3 +

[dpdk-dev] [PATCH v4 20/62] net/cnxk: add Rx multi-segmented version for cn9k

Add Rx burst multi-segmented version for CN9K. Signed-off-by: Nithin Dabilpuram Signed-off-by: Pavan Nikhilesh --- drivers/net/cnxk/cn9k_rx.c | 17 drivers/net/cnxk/cn9k_rx.h | 60 ++--- drivers/net/cnxk/cn9k_rx_mseg.c | 17

[dpdk-dev] [PATCH v4 21/62] net/cnxk: add Rx vector version for cn9k

From: Jerin Jacob Add Rx burst vector version for CN9K. Signed-off-by: Jerin Jacob Signed-off-by: Nithin Dabilpuram --- drivers/net/cnxk/cn9k_rx.c | 13 ++- drivers/net/cnxk/cn9k_rx.h | 221 + drivers/net/cnxk/cn9k_rx_vec.c | 17 drivers/

[dpdk-dev] [PATCH v4 22/62] net/cnxk: add Tx burst for cn9k

From: Jerin Jacob Add Tx burst scalar version for CN9K. Signed-off-by: Jerin Jacob Signed-off-by: Nithin Dabilpuram Signed-off-by: Pavan Nikhilesh Signed-off-by: Harman Kalra --- drivers/net/cnxk/cn9k_ethdev.h | 1 + drivers/net/cnxk/cn9k_tx.c | 53 ++ drivers/net/cnxk/cn9k_tx.h

[dpdk-dev] [PATCH v4 23/62] net/cnxk: add Tx multi-segment version for cn9k

Add Tx burst multi-segment version for CN9K. Signed-off-by: Nithin Dabilpuram Signed-off-by: Pavan Nikhilesh --- drivers/net/cnxk/cn9k_tx.c | 14 drivers/net/cnxk/cn9k_tx.h | 150 drivers/net/cnxk/cn9k_tx_mseg.c | 25 +++ drivers/net

[dpdk-dev] [PATCH v4 24/62] net/cnxk: add Tx vector version for cn9k

Add Tx burst vector version for CN9K. Signed-off-by: Nithin Dabilpuram Signed-off-by: Pavan Nikhilesh --- drivers/net/cnxk/cn9k_tx.c | 16 +- drivers/net/cnxk/cn9k_tx.h | 743 + drivers/net/cnxk/cn9k_tx_vec.c | 25 ++ drivers/net/cnxk/meson.buil

[dpdk-dev] [PATCH v4 25/62] net/cnxk: add Rx burst for cn10k

From: Jerin Jacob Add Rx burst support for CN10K SoC. Signed-off-by: Jerin Jacob Signed-off-by: Nithin Dabilpuram Signed-off-by: Pavan Nikhilesh Signed-off-by: Harman Kalra --- drivers/net/cnxk/cn10k_ethdev.h | 3 + drivers/net/cnxk/cn10k_rx.c | 45 drivers/net/cnxk/cn10k_rx

[dpdk-dev] [PATCH v4 26/62] net/cnxk: add Rx multi-segment version for cn10k

Add Rx burst multi-segment version for CN10K. Signed-off-by: Nithin Dabilpuram Signed-off-by: Pavan Nikhilesh --- doc/guides/nics/cnxk.rst | 2 ++ doc/guides/nics/features/cnxk.ini | 2 ++ doc/guides/nics/features/cnxk_vec.ini | 1 + doc/guides/nics/features/cnxk_vf.ini |

[dpdk-dev] [PATCH v4 27/62] net/cnxk: add Rx vector version for cn10k

From: Jerin Jacob Add Rx burst vector version for CN10K. Signed-off-by: Jerin Jacob Signed-off-by: Nithin Dabilpuram --- doc/guides/nics/cnxk.rst| 1 + drivers/net/cnxk/cn10k_rx.c | 13 ++- drivers/net/cnxk/cn10k_rx.h | 222 drivers/

[dpdk-dev] [PATCH v4 28/62] net/cnxk: add Tx burst for cn10k

From: Jerin Jacob Add Tx burst scalar version for CN10K. Signed-off-by: Jerin Jacob Signed-off-by: Nithin Dabilpuram Signed-off-by: Pavan Nikhilesh Signed-off-by: Harman Kalra --- doc/guides/nics/cnxk.rst | 1 + doc/guides/nics/features/cnxk.ini | 7 + doc/guides/nics/f

[dpdk-dev] [PATCH v4 29/62] net/cnxk: add Tx multi-segment version for cn10k

Add Tx burst multi-segment version for CN10K. Signed-off-by: Nithin Dabilpuram Signed-off-by: Pavan Nikhilesh --- drivers/net/cnxk/cn10k_tx.c | 18 - drivers/net/cnxk/cn10k_tx.h | 171 +++ drivers/net/cnxk/cn10k_tx_mseg.c | 25 ++ drivers/

[dpdk-dev] [PATCH v4 30/62] net/cnxk: add Tx vector version for cn10k

Add Tx burst vector version for CN10K. Signed-off-by: Nithin Dabilpuram Signed-off-by: Pavan Nikhilesh --- drivers/net/cnxk/cn10k_tx.c | 10 + drivers/net/cnxk/cn10k_tx.h | 815 drivers/net/cnxk/cn10k_tx_vec.c | 25 ++ drivers/net/cnxk/meson.bu

[dpdk-dev] [PATCH v4 31/62] net/cnxk: add device start and stop operations

Add device start and stop operation callbacks for CN9K and CN10K. Device stop is common for both platforms while device start as some platform dependent portion where the platform specific offload flags are recomputed and the right Rx/Tx burst function is chosen. Signed-off-by: Nithin Dabilpuram

  1   2   >