[PATCH] vhost: support CPU copy for small packets

Offloading small packets to DMA degrades throughput 10%~20%, and this is because DMA offloading is not free and DMA is not good at processing small packets. In addition, control plane packets are usually small, and assign those packets to DMA will significantly increase latency, which may cause tim

RE: [PATCH v1] net/iavf: fix pattern check for flow director parser

> -Original Message- > From: Steve Yang > Sent: Wednesday, August 10, 2022 2:48 PM > To: dev@dpdk.org > Cc: Wu, Jingjing ; Xing, Beilei > ; Yang, SteveX ; > sta...@dpdk.org > Subject: [PATCH v1] net/iavf: fix pattern check for flow director parser > > FDIR rules with masks are not supp

RE: [PATCH] net/ice: support disabling ACL engine in DCF via devargs

> -Original Message- > From: Zeng, ZhichaoX > Sent: Monday, July 25, 2022 11:15 AM > To: dev@dpdk.org > Cc: Yang, Qiming ; Zeng, ZhichaoX > ; Zhang, Qi Z > Subject: [PATCH] net/ice: support disabling ACL engine in DCF via devargs > > From: Zhichao Zeng > > Support disabling DCF ACL

[PATCH v2] ethdev: remove header split Rx offload

From: Xuan Ding As announced in the deprecation note, this patch removes the Rx offload flag 'RTE_ETH_RX_OFFLOAD_HEADER_SPLIT' and 'split_hdr_size' field from the structure 'rte_eth_rxmode'. Meanwhile, the place where the examples and apps initialize the 'split_hdr_size' field, and where the driv

[PATCH v1] net/ice/base: fix switch rules not cleared on warm reset

When users killed app forcely (e.g.: kill -9 pid), the driver reset couldn't make all registers of NIC recovery to initial status. For example, the switch filter rules, which involved the vlan tag, couldn't be added. Tell the Firmware to shut down the AdminQ to avoid possible error when process wa

RE: [PATCH] net/igc: add support for Ethernet Controller I225-IT

> -Original Message- > From: Guo, Junfeng > Sent: Wednesday, August 10, 2022 2:07 PM > To: Zhang, Qi Z ; Yang, Qiming > ; Su, Simei > Cc: dev@dpdk.org; Guo, Junfeng > Subject: [PATCH] net/igc: add support for Ethernet Controller I225-IT > > Add device id for Ethernet Controller (2) I

RE: [PATCH] net/iavf: fix VLAN insertion

> -Original Message- > From: Yiding Zhou > Sent: Friday, August 12, 2022 10:53 AM > To: dev@dpdk.org > Cc: Wu, Jingjing ; Xing, Beilei > ; sta...@dpdk.org; Zhou, YidingX > > Subject: [PATCH] net/iavf: fix VLAN insertion > > When the PF driver tells the VF to insert VLAN tag using the

[PATCH] net/iavf: fix VLAN insertion

When the PF driver tells the VF to insert VLAN tag using the L2TAG2 field, vector Tx path does not use Tx context descriptor and would cause VLAN tag inserted into the wrong location. This commit is to fix the issue by using normal Tx path to handle L2TAG2 case. Fixes: 3aa957338503 ("net/iavf: fi

RE: [PATCH] net/ice: remove deprecated VF flow action

> -Original Message- > From: Zeng, ZhichaoX > Sent: Wednesday, August 10, 2022 2:50 PM > To: dev@dpdk.org > Cc: Yang, Qiming ; Zhou, YidingX > ; Zeng, ZhichaoX ; > Zhang, Qi Z > Subject: [PATCH] net/ice: remove deprecated VF flow action > > From: Zhichao Zeng > > According to the AB

RE: [PATCH v2] net/iavf: fix Tx L3 checksum offload flag

> -Original Message- > From: Ke Zhang > Sent: Wednesday, August 10, 2022 5:57 PM > To: Wu, Jingjing ; Xing, Beilei > ; dev@dpdk.org > Cc: Zhang, Ke1X ; sta...@dpdk.org > Subject: [PATCH v2] net/iavf: fix Tx L3 checksum offload flag > > When ol_flag is only RTE_MBUF_F_TX_IPV4, the Tx L3

[PATCH] net/iavf: fix VLAN insertion

When the driver tells the VF to insert VLAN tag using the L2TAG2 field, vector Tx path does not use Tx context descriptor and would cause VLAN tag inserted into the wrong location. This commit is to fix issue by using normal Tx path to handle L2TAG2 case. Fixes: 3aa957338503 ("net/iavf: fix VLAN

RE: [RFC v2] non-temporal memcpy

> > > >> > >>> From: Mattias Rönnblom [mailto:hof...@lysator.liu.se] > >>> Sent: Wednesday, 10 August 2022 13.56 > >>> > >>> On 2022-08-09 17:26, Stephen Hemminger wrote: > >> > >> [...] > >> > >>> > >>> Alignment seems like a non-issue to me. A NT-store memcpy() can be > >>> made free of alignme

Re: [dpdk-kmods] windows/netuio: fix bar parsing

On 8/9/2022 2:15 AM, Dmitry Kozlyuk wrote: 2022-08-08 17:33 (UTC-0700), Kadam, Pallavi: [...] Hi Pallavi, In the first place, it was wrong to always test `prev_bar & PCI_TYPE_64BIT` because only the first BAR slot of a 64-bit BAR contains flags. The current code has a state to track (curr_bar,

[dpdk-kmods v2] windows/netuio: fix BAR parsing

Current code was always checking the 'prev_bar & PCI_TYPE_64BIT' though only the first BAR slot of a 64-bit BAR contains flags. Also for certain PCIe devices, BAR values were not continuous. This patch fixes this incorrectness and maps the BAR addresses correctly. Reported-by: Qiao Liu Suggested-

RE: [PATCH v3 1/4] eventdev/timer: add periodic event timer support

Hi Harish, > -Original Message- > From: Naga Harish K, S V > Sent: Thursday, August 11, 2022 10:37 AM > To: Carrillo, Erik G ; jer...@marvell.com > Cc: pbhagavat...@marvell.com; sthot...@marvell.com; dev@dpdk.org > Subject: [PATCH v3 1/4] eventdev/timer: add periodic event timer support >

RE: [RFC v2] non-temporal memcpy

> >> > >> +TO: @Honnappa, we need input from ARM > >> > >>> From: Konstantin Ananyev [mailto:konstantin.anan...@huawei.com] > >>> Sent: Friday, 29 July 2022 21.49 > > > From: Konstantin Ananyev [mailto:konstantin.anan...@huawei.com] > > Sent: Friday, 29 July 2022 14.14 > > >

RE: [PATCH v2 2/4] event/sw: report periodic event timer capability

Hi Gabe, > -Original Message- > From: Carrillo, Erik G > Sent: Thursday, August 11, 2022 2:22 AM > To: Naga Harish K, S V ; jer...@marvell.com; > Van Haaren, Harry > Cc: dev@dpdk.org > Subject: RE: [PATCH v2 2/4] event/sw: report periodic event timer capability > > Hi Harish, > > > ---

RE: [PATCH v2 1/4] eventdev/timer: add periodic event timer support

Hi Gabe, > -Original Message- > From: Carrillo, Erik G > Sent: Thursday, August 11, 2022 1:25 AM > To: Naga Harish K, S V ; jer...@marvell.com > Cc: pbhagavat...@marvell.com; sthot...@marvell.com; dev@dpdk.org > Subject: RE: [PATCH v2 1/4] eventdev/timer: add periodic event timer > suppor

RE: [PATCH v2 3/4] timer: fix function to stop all timers

Hi Gabe, > -Original Message- > From: Carrillo, Erik G > Sent: Thursday, August 11, 2022 1:00 AM > To: Naga Harish K, S V > Cc: dev@dpdk.org; sta...@dpdk.org > Subject: RE: [PATCH v2 3/4] timer: fix function to stop all timers > > Hi Harish, > > > -Original Message- > > From: N

[PATCH v3 3/4] timer: fix function to stop all timers

There is a possibility of deadlock in this API, as same spinlock is tried to be acquired in nested manner. If the lcore that is stopping the timer is different from the lcore that owns the timer, the timer list lock is acquired in timer_del(), even if local_is_locked is true. Because the same lock

[PATCH v3 2/4] event/sw: report periodic event timer capability

update the software eventdev pmd timer_adapter_caps_get callback function to report the support of periodic event timer capability Signed-off-by: Naga Harish K S V --- drivers/event/sw/sw_evdev.c | 2 +- lib/eventdev/eventdev_pmd.h | 3 +++ 2 files changed, 4 insertions(+), 1 deletion(-) diff -

[PATCH v3 4/4] test/event: update periodic event timer tests

This patch updates the software timer adapter tests to configure and use periodic event timers. Signed-off-by: Naga Harish K S V --- app/test/test_event_timer_adapter.c | 41 ++--- 1 file changed, 37 insertions(+), 4 deletions(-) diff --git a/app/test/test_event_timer_ad

[PATCH v3 1/4] eventdev/timer: add periodic event timer support

This patch adds support to configure and use periodic event timers in software timer adapter. The structure ``rte_event_timer_adapter_stats`` is extended by adding a new field, ``evtim_drop_count``. This stat represents the number of times an event_timer expiry event is dropped by the event timer

RE: [PATCH v2 2/4] event/sw: report periodic event timer capability

> -Original Message- > From: Carrillo, Erik G > Sent: Wednesday, August 10, 2022 9:52 PM > To: Naga Harish K, S V ; jer...@marvell.com; Van > Haaren, Harry > Cc: dev@dpdk.org > Subject: RE: [PATCH v2 2/4] event/sw: report periodic event timer capability > > Hi Harish, > > > -Origina

Re: [PATCH v5 07/12] net/nfp: add flower ctrl VNIC related logics

On Thu, 11 Aug 2022 06:31:31 + Chaoyong He wrote: > > -Original Message- > > From: Stephen Hemminger > > Sent: Thursday, August 11, 2022 12:25 PM > > To: Chaoyong He > > Cc: Andrew Rybchenko ; Niklas > > Soderlund ; dev@dpdk.org > > Subject: Re: [PATCH v5 07/12] net/nfp: add flower

RE: [PATCH v11 1/7] eventdev/eth_rx: add adapter instance get API

Hi Jerin, Could you please review this? Thanks, Ganapati > -Original Message- > From: Ganapati Kundapura > Sent: 19 July 2022 13:56 > To: jer...@marvell.com; Jayatheerthan, Jay ; > Naga Harish K, S V ; dev@dpdk.org > Subject: [PATCH v11 1/7] eventdev/eth_rx: add adapter instance get A

Re: [PATCH] multi_proc_support.rst: updated file location for config file in documentation

Hi, I've submitted a patch on 17th March, 2022. One of the tests is failing and it is not related to the patch. Following test is failing: - ci/iol-broadcom-Functional Can you please rerun the tests so that the patch can be submitted. Regards, -Jawad On Thu, Mar 17, 2022 at 11:19 AM Muhammad Ja

[PATCH v2] examples/ethtool: update rxmode to increase functionality

previously rxmode functionality did not allow the user to choose between different vfs nor did it allow the user to choose rxmode settings by default it was set to vf = 0, rxmode = AUPE and the on/off state toggled without letting the user know what state it is in also there were no error messages

[RFC] ethdev: introduce hairpin memory capabilities

This RFC introduces new hairpin queue configuration options through rte_eth_hairpin_conf struct, allowing to tune Rx and Tx hairpin queues memory configuration. Hairpin configuration is extended with the following fields: - use_locked_device_memory - If set, PMD will use specialized on-device me

Re: [RFC v2] non-temporal memcpy

On 2022-08-10 23:20, Honnappa Nagarahalli wrote: From: Mattias Rönnblom [mailto:hof...@lysator.liu.se] Sent: Wednesday, 10 August 2022 13.56 On 2022-08-09 17:26, Stephen Hemminger wrote: [...] Alignment seems like a non-issue to me. A NT-store memcpy() can be made free of alignment req

Re: [RFC v2] non-temporal memcpy

On 2022-08-10 23:05, Honnappa Nagarahalli wrote: +TO: @Honnappa, we need input from ARM From: Konstantin Ananyev [mailto:konstantin.anan...@huawei.com] Sent: Friday, 29 July 2022 21.49 From: Konstantin Ananyev [mailto:konstantin.anan...@huawei.com] Sent: Friday, 29 July 2022 14.14 Sorr

[RFC] ethdev: add send to kernel action

In some cases application may receive a packet that should have been received by the kernel. In this case application uses KNI or other means to transfer the packet to the kernel. This commit introduces rte flow action that the application may use to route the packet to the kernel while still in th

[PATCH v2] examples/ethtool: adds promiscuous mode functionality

ethtool did not have promiscuous mode functioality previously which is needed for viewing broadcast and multicast packets. This patch allows user to turn on/off promiscuous mode on each port through command line. Signed-off-by: Muhammad Jawad Hussain --- doc/guides/sample_app_ug/ethtool.rst |

[PATCH] ethdev: remove header split Rx offload

From: Xuan Ding As announced in the deprecation note, this patch removes the Rx offload flag 'RTE_ETH_RX_OFFLOAD_HEADER_SPLIT' and 'split_hdr_size' field from the structure 'rte_eth_rxmode'. User can still use `RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT` for per-queue packet split offload, which is configur

[PATCH] common/cnxk: add CPT LF reset sequence

Adds code to reset CPT LF as part of cpt_lf_fini. Signed-off-by: Srujana Challa --- drivers/common/cnxk/roc_cpt.c | 82 ++ drivers/common/cnxk/roc_mbox.h | 6 +++ 2 files changed, 88 insertions(+) diff --git a/drivers/common/cnxk/roc_cpt.c b/drivers/common/cnxk

Re: [PATCH] usertools: fix bind failure from dpdk to kernel

On 09/08/2022 20:58, Stephen Hemminger wrote: >> >> However, due to the patch[1] merged into 5.19 kernel, 'driver_override' >> in the pci_dev is no longer NULL by writing '\00' to driver_override file. >> This causes PCI match device failure and the device will never be bound to >> their kernel dri

[PATCH] bus/pci: 'RTE_PMD_REGISTER_PCI' support expand 'nm'

Before, 'RTE_PMD_REGISTER_PCI' can not expand nm. So, it can not work in this situation: #define TEST(a, b) a##_##b RTE_PMD_REGISTER_PCI(TEST(a, b), test_pmd_driver) Signed-off-by: Lichao Liu --- drivers/bus/pci/rte_bus_pci.h | 7 +-- 1 file changed, 5 insertions(+), 2 deletions(-)

[PATCH] net/bnxt: fix null pointer dereference in bnxt_hwrm_port_led_cfg()

From: maoyingming VFs's "bp->leds" is allways null, check bp->leds is not null before use bp->leds->num_leds. segfault backtrace in trex program when use VF: 11: bnxt_hwrm_port_led_cfg (bp=0x23ffb2140, led_on=true) 10: bnxt_dev_led_on_op (dev=0x22d7780 ) 9: rte_eth_led_on (port_id=0) 8: DpdkTR

Re: [PATCH] usertools: fix bind failure from dpdk to kernel

On 09/08/2022 14:44, lihuisong (C) wrote: > > 在 2022/8/5 23:35, Stephen Hemminger 写道: >> On Fri, 5 Aug 2022 11:10:22 +0800 >> Huisong Li wrote: >> >>> Currently, the steps for binding device from dpdk driver to kernel >>> driver is as follows: >>> echo $BDF > /sys/bus/pci/drivers/vfio-pci/unbind

[PATCH] net/bnxt: fix null pointer dereference in bnxt_hwrm_port_led_cfg()

From: maoyingming VFs's "bp->leds" is allways null, check bp->leds is not null before use bp->leds->num_leds. segfault backtrace in trex program when use VF: 11: bnxt_hwrm_port_led_cfg (bp=0x23ffb2140, led_on=true) 10: bnxt_dev_led_on_op (dev=0x22d7780 ) 9: rte_eth_led_on (port_id=0) 8: DpdkTR

Reason to alway to build both static and shared libs

Hi all, Could I know the reason for always building both static and shared libs of DPDK? I can find the patch to enable this behavior, but it seems that it didn't mention the reason behind it. Also, if I propose a change to use "both" as th

[PATCH] net/bnxt: fix null pointer dereference in bnxt_hwrm_port_led_cfg()

From: maoyingming VFs's "bp->leds" is allways null, check bp->leds is not null before use bp->leds->num_leds. segfault backtrace in trex program when use VF: 11: bnxt_hwrm_port_led_cfg (bp=0x23ffb2140, led_on=true) 10: bnxt_dev_led_on_op (dev=0x22d7780 ) 9: rte_eth_led_on (port_id=0) 8: DpdkTR

Re: 19.11.13 patches review and test

On Thu, Aug 11, 2022 at 9:37 AM Jiang, YuX wrote: > > > -Original Message- > > From: christian.ehrha...@canonical.com > > Sent: Thursday, August 4, 2022 4:22 PM > > To: sta...@dpdk.org > > Cc: dev@dpdk.org; Abhishek Marathe ; > > Ali Alnubani ; Walker, Benjamin > > ; David Christensen > >

RE: 19.11.13 patches review and test

> -Original Message- > From: christian.ehrha...@canonical.com > Sent: Thursday, August 4, 2022 4:22 PM > To: sta...@dpdk.org > Cc: dev@dpdk.org; Abhishek Marathe ; > Ali Alnubani ; Walker, Benjamin > ; David Christensen > ; Hemant Agrawal ; > Stokes, Ian ; Jerin Jacob ; > Mcnamara, John ;

[RFC] ethdev: add direction info when creating the transfer table

The transfer domain rule is able to match traffic wire/vf origin and it means two directions' underlayer resource. In customer deployments, they usually match only one direction traffic in single flow table: either from wire or from vf. Introduce one new member transfer_mode into rte_flow_attr to