net/hns3: add free threshold in Rx

This patch optimizes the Rx performance by adding the rx_free_thresh
related process in the '.rx_pkt_burst' ops implementation function named
hns3_recv_pkts. The related change as follows:
1. Adding the rx_free_thresh related process to reduce the number of
   writing the HNS3_RING_RX_HEAD_REG register.
2. Adjusting the internal macro named DEFAULT_RX_FREE_THRESH to 32 and
   adjusting HNS3_MIN_RING_DESC to 64 to make the effect of the thresh
   more obvious.

Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>

net/hns3: remove one IO barrier in Rx

When receiving a packet, hns3 hardware network engine firstly writes the
packet content to the memory pointed by the 'addr' field of the Rx
Buffer Descriptor, secondly fills the result of parsing the packet
include the valid field into the Rx Buffer Descriptor in one write
operation, and thirdly writes the number of the Buffer Descriptor not
processed by the driver to the HNS3_RING_RX_FBDNUM_REG register.

This patch optimizes the Rx performance by removing one rte_io_rmb call
in the '.rx_pkt_burst' ops implementation function named hns3_recv_pkts.
The change as follows:
1. Driver no longer read HNS3_RING_RX_FBDNUM_REG register, so remove one
   rte_io_rmb call, and directly read the valid flag of Rx Buffer
   Descriptor to check whether the BD is ready.
2. Delete the non_vld_descs field from the statistic information of the
   hns3 driver because now it has become a common case that the valid
   flag of Rx Buffer Descriptor read by the driver is invalid.

Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>

net/hns3: reduce judgements of free Tx ring space

This patch reduces the number of the judgement of the free Tx ring space
in the 'tx_pkt_burst' ops implementation function to avoid performance
loss. According to hardware constraints, we need to reserve a Tx Buffer
Descriptor in the TX ring in hns3 network engine.

Signed-off-by: Yisen Zhuang <yisen.zhuang@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>

net/hns3: support setting VF MAC address by PF driver

Currently, we only support VF device is bound to vfio_pci or igb_uio and
then driven by DPDK driver when PF is driven by kernel mode hns3 ethdev
driver, VF is not supported when PF is driven by hns3 DPDK driver.

This patch adds support setting VF MAC address by hns3 PF kernel ethdev
driver on the host by "ip link set ..." command.
1) If the hns3 PF kernel ethdev driver sets the MAC address for VF
   device before the initialization of the related VF device, hns3 VF
   PMD driver should get the MAC address from PF driver through mailbox
   and configure hardware using this MAC address in the initialization.
   The hns3 VF PMD driver get the MAC address form PF driver, if
   obtaining a non-zero MAC address from mailbox, VF driver will
   configure hardware using it, otherwise using a random MAC address in
   the initialization of VF device.
2) If the hns3 PF kernel ethdev driver sets the MAC address for VF
   device after the initialization of the related VF device, the PF
   driver will notify VF driver to reset VF device to make the new MAC
   address effective immediately. The hns3 VF PMD driver should check
   whether the MAC address has been changed by the PF kernel netdevice
   driver, if changed VF driver should configure hardware using the new
   MAC address in the recovering hardware configuration stage of the
   reset process.
3) When user has configured a mac address for VF device by "ip link set
   ..." command based on the PF device, the hns3 PF kernel ethdev driver
   does not allow VF driver to request reconfiguring a different default
   mac address and return -EPREM to VF driver through mailbox.

Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Signed-off-by: Hongbo Zheng <zhenghongbo3@huawei.com>

net/hns3: support different numbers of Rx and Tx queues

Hardware does not support individually enable/disable/reset the Tx or Rx
queue in hns3 network engine, driver must enable/disable/reset Tx and Rx
queues at the same time.

Currently, hns3 PMD driver does not support the scenarios as below:
1) When calling the following function, the input parameter nb_rx_q and
   nb_tx_q are not equal.
     rte_eth_dev_configure(uint16_t port_id, uint16_t nb_rx_q,
                      uint16_t nb_tx_q,
      const struct rte_eth_conf *dev_conf);
2) When calling the following functions to setup queues, the
   cumulatively setup Rx queues are not the same as the setup Tx queues.
     rte_eth_rx_queue_setup(uint16_t port_id, uint16_t rx_queue_id,,,);
     rte_eth_tx_queue_setup(uint16_t port_id, uint16_t tx_queue_id,,,);
However, these are common usage scenarios in some applications, such as,
l3fwd, ip_ressmbly and OVS-DPDK, etc.

This patch adds support for this usage of these functions by setup fake
Tx or Rx queues to adjust numbers of Tx/Rx queues. But these fake queues
are imperceptible, and can not be used by upper applications.

Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>

net/i40e: support flow director for L2TPv3 over IP

Adding FDIR support for L2TPv3 over IP header matching and adding
a new customized pctype for l2tpv3 over IP.

Signed-off-by: Rory Sexton <rory.sexton@intel.com>
Signed-off-by: Dariusz Jagus <dariuszx.jagus@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

ethdev: add L2TPv3 over IP header to flow API

This patch adds the new flow item RTE_FLOW_ITEM_TYPE_L2TPV3OIP to
flow API to match a L2TPv3 over IP header. This patch supports only
L2TPv3 over IP header format which is different to L2TPv2/L2TPv3
over UDP. The difference in header formats between L2TPv3 over IP
and L2TP over UDP require a separate implementation for each.

Signed-off-by: Rory Sexton <rory.sexton@intel.com>
Signed-off-by: Dariusz Jagus <dariuszx.jagus@intel.com>
Acked-by: Ori Kam <orika@mellanox.com>

net/af_xdp: use single producer/consumer ring

The ring is used only by af_xdp PMD itself, so no need to support
multi-producer and multi-consumer mode. This patch changes the ring
to single-producer and single-consumer mode, which could yield better
performance for addr enqueue and dequeue.

Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Tested-by: Ciara Loftus <ciara.loftus@intel.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>

app/testpmd: fix GENEVE flow item

Previous patch added support of GENEVE pattern item in flow rule.
Function flow_item_default_mask() was not updated, so using it with
GENEVE item returns null.
Using testpmd command "set raw_decap" or "set raw_encap" with
GENEVE item, without specifying any parameters, results in
segmentation fault.

This patch updates function flow_item_default_mask(),
adding case to handle GENEVE item.

Fixes: 0f4203fe9d18 ("app/testpmd: support GENEVE pattern item in flow rules")
Cc: stable@dpdk.org
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>

ethdev: fix callback unregister with wildcard argument list

The function was checking -1 against the callback data instead of
the given cb_arg parameter.

Fixes: af75078fece3 ("first public release")
Cc: stable@dpdk.org
Signed-off-by: Ricardo Roldan <rroldan@bequant.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/mlx5: fix Tx burst routines set

The tx_burst routine supporting multi-segment packets with
legacy MPW and without inline was missed, and there was no
valid selection for these options, patch adds the missing
routine.

Fixes: 82e75f8323bf ("net/mlx5: fix legacy multi-packet Tx descriptors")
Cc: stable@dpdk.org
Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>

net/ice: use ethernet copy API to do MAC assignment

Use the API rte_ether_addr_copy to do MAC assignment, instead of
calling rte_memcpy function directly.

Fixes: 84dc7a95a2d3 ("net/ice: enable flow director engine")
Cc: stable@dpdk.org
Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>

net/iavf/base: move to drivers common directory

Change the iavf base code as driver common library, it is used by iavf
PMD now, and it can be used by other Intel SR-IOV PMDs in the future.

Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>

net/i40e/base: add new link speed constants

This patch fixes 'NIC Link is Up, Unknown bps' message in dmesg
for 2.5Gb/5Gb speeds. This problem is fixed by adding constants
for VIRTCHNL_LINK_SPEED_2_5GB and VIRTCHNL_LINK_SPEED_5GB.

Fixes: a58860f68929 ("net/i40e/base: use new virtchnl header file")

Signed-off-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: enable LED blinking flow

Add X710T*L device specific operations (in port LED detection and
handling of GLGEN_GPIO_CTL.PIN_FUNC field) to enable LED blinking.

Signed-off-by: Damian Milosek <damian.milosek@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: add CRC stripping capability

Some VFs may want to disable CRC stripping on incoming packets
so create an offload for that. The VF already sends information
about configuring it's RX queues so use that structure to
indicate that the CRC stripping should be enabled or not.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: introduce firmware EMP reset register

Introduce offsets of firmware registers holding a number of EMP resets.
Implement offsets as a preprocessor's constants.
Required by drivers to read a number of EMP resets reported by FW.

Signed-off-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: expose LED functionality

The previous code had the LED code wrapped such that, it was
static to the file and not usable outside of it. It is now available.

Signed-off-by: Mateusz Rusinski <mateusz.rusinski@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: add function to read LPI statistics

There was no time duration statistics for LPI (Low Power Idle)
in EEE (Energy-Efficient Ethernet).
Added new procedure to get duration values from FW.
Otherwise there is no available procedure to get tx_lpi_duration
and rx_lpi_duration values.

Signed-off-by: Jaroslaw Gawin <jaroslawx.gawin@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: set PHY access flag on X722

The X722 FW API version 1.9 adds support for accessing PHY
registers with Admin Queue Command. This enables reading
EEPROM data from (Q)SFP+ transceivers, what was previously
possible only on X710 devices.

Signed-off-by: Adam Ludkiewicz <adam.ludkiewicz@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: remove unreachable code

The result of the check is always FALSE and the code within "if"
is never executed.

Signed-off-by: Dzmitry Sautsa <dzmitryx.sautsa@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: add flow director constants

Add defines for creating Flow Director flows as defined in datasheet
section 7.1.5.4 for field vectors.

Signed-off-by: Doug Dziggel <douglas.a.dziggel@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: fix display of FEC settings

Fix display of FEC settings for ethtool.
This patch change variable 'phy.link_info.req_fec_info' sent to ethtool.
Without this patch wrong FEC settings can be shown.

Fixes: 1216fb9f0c8c ("net/i40e/base: store the requested FEC information")

Signed-off-by: Jaroslaw Gawin <jaroslawx.gawin@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: mark additional AQ bits as reserved

Mark bits 0xD through 0xF for the command flags of a cloud filter as
reserved. These bits are not yet defined and are considered as reserved
in the data sheet.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: fix retrying logic

Fixed a bug where driver was breaking out of the loop and
reporting an error without retrying first.

Fixes: 466eec7d6b1a ("net/i40e/base: retry AQC to overcome IRCRead hangs")

Signed-off-by: Marcin Formela <marcin.formela@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: add function to set MAC type

Make i40e_set_mac_type() public. i40e driver requires
i40e_set_mac_type() to be public. It is required for recovery mode
handling. Without this patch recovery mode could not be detected at
probe() function.

Signed-off-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: support persistent LLDP

This patch adds a function to read NVM module data and uses it to
read current LLDP agent configuration from NVM API version 1.8.

Signed-off-by: Sylwia Wnuczko <sylwia.wnuczko@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: update FW API version

Update FW API version to 1.9.

Signed-off-by: Piotr Azarewicz <piotr.azarewicz@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: add FWS1B register masks

Function check_recovery_mode had wrong if statement. Now we check proper
FWS1B register values, which are responsible for the recovery mode.
Recovery mode has 6 values in total. That's why we need 6 different
flags.  Now in if statement we recognize type of mac and register value.
Without those changes driver could show wrong state.

Signed-off-by: Adrian Podlawski <adrian.podlawski@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: add drop mode parameter to set MAC config

This patch adds "drop mode" parameter to set mac config AQ command.
This bit controls the behavior when a no-drop packet is blocking a TC
queue.
0 – The PF driver is notified.
1 – The blocking packet is dropped and then the PF driver is notified.

Signed-off-by: Sylwia Wnuczko <sylwia.wnuczko@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: extend PHY access AQ command

Currently FW use MDIO I/F number corresponded with current PF for PHY
access. This code allow to specify used MDIO I/F number.

Signed-off-by: Piotr Azarewicz <piotr.azarewicz@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: fix Tx descriptors number

The existing driver allows setting the number of TX descriptors
to the value that is indivisible by 32. This is not properly
supported by x710 hardware. The patch limits the number of TX
descriptors to the whole value of 32.

Fixes: 8db9e2a1b232 ("i40e: base driver")
Cc: stable@dpdk.org
Signed-off-by: Doug Dziggel <douglas.a.dziggel@intel.com>
Signed-off-by: Dariusz Chaberski <dariuszx.chaberski@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: update status codes

Add a few new status codes and rename a few to make them more
consistent.  Error code are mapped to similar values as in
i40e_status.h, so as to be compatible with older VF drivers not using
this status enum.

Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: backport style changes from upstream Linux

The upstream virtchnl.h has a few minor style differences to what is
out-of-tree, and there is no real reason we should prefer the current
style over whats upstream.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: add function to read LPI counters

This change introduces new function i40e_get_lpi_counters() reading
Low Power Idle (LPI) mode counters from Energy Efficient Ethernet (EEE)
statistics. Since reading EEE statistics requires running DNL script,
function i40e_aq_run_phy_activity() able to run it, was implemented.

i40e_lpi_stat_update() was introduced as helper function for retrieving
LPI counters' values relative to given offsets.

This change also fixes order of fields in struct
i40e_aqc_run_phy_activity and introduces union in the struct
providing more descriptive names of fields depending on usage
(command or response).

Signed-off-by: Piotr Pietruszewski <piotr.pietruszewski@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: fix missing link modes

Fix for missing "Supported link modes" and "Advertised link modes"
info in ethtool after changed speed on X722 devices with BASE-T PHY
with FW API version >= 1.7.
The same FW API version on X710 and X722 does not mean the same
feature set so the change was needed as mac type of the device
should also be checked instead of FW API version only.

Fixes: e8228f1a16b7 ("net/i40e/base: report supported link modes")

Signed-off-by: Martyna Szapar <martyna.szapar@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: read LPI status from external PHY

When external PHY is used, MAC register may not reflect actual
state of LPI. Add function reading it directly from PHY or MAC,
depending on what is supported by the device.

Signed-off-by: Krzysztof Galazka <krzysztof.galazka@intel.com>
Signed-off-by: Piotr Pietruszewski <piotr.pietruszewski@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: support Energy Efficient Ethernet

This change adds i40e_enable_eee() function controlling advertisement
of Energy Efficient Ethernet mode (EEE).

Signed-off-by: Piotr Pietruszewski <piotr.pietruszewski@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: add function to get FW LLDP agent status

This change introduces i40e_get_fw_lldp_status() function capable of
reporting status of FW Link Layer Discovery Protocol (LLDP) agent.

Since reading LLDP configuration from NVM only gives information what
is the default state of FW LLDP agent after POR, this change introduces
more reliable method for checking if agent is enabled.

Current state of FW LLDP agent may be different if user disabled the
agent, so i40e_get_fw_lldp_status() uses LLDP MIB checking to determine
status of the agent.

Signed-off-by: Piotr Pietruszewski <piotr.pietruszewski@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: improve AQ log granularity

This patch makes it possible to log only AQ descriptors, without the
entire AQ message buffers being dumped too. It should greatly reduce
kernel log size in cases where a full AQ dump is not needed.
Selection is made by setting flags in hw->debug_mask.

Additionally, some debug messages that preceded an AQ dump have been
moved to I40E_DEBUG_AQ_COMMAND class, which seems more appropriate.

Signed-off-by: Doug Dziggel <douglas.a.dziggel@intel.com>
Signed-off-by: Robert Konklewski <robertx.konklewski@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: add flags for PHY types

Adding bit flags to enable/disable EEE capability for 2.5GBase-T and
5GBase-T PHY types as well as 'Set autonomous EEE for relevant enabled
PHY types' flag.

Signed-off-by: Doug Dziggel <douglas.a.dziggel@intel.com>
Signed-off-by: Michal Litwicki <michalx.litwicki@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: check MAC type

Some features were introduced in different FW API version on XL710 and
X722 MACs. Others are available only on specific MAC type. Make sure
that they are properly assigned. Also fix the style issues reported by
current check-patch.

Signed-off-by: Krzysztof Galazka <krzysztof.galazka@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: add new device IDs

New device ids are created to support SFP+ and backplane connections.

Signed-off-by: Piotr Azarewicz <piotr.azarewicz@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: support LLDP agent

This code implements changes necessary for LLDP Agent support.

1. Modified i40e_aq_start_lldp and i40e_aq_stop_lldp. Now Stop and Start
can also be persistent across power cycles.
2. Added new function i40e_aq_restore_lldp which restores factory
setting for LLDP Agent or gets its status.

Signed-off-by: Jaroslaw Ilgiewicz <jaroslaw.ilgiewicz@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: fix error message

This patch changes an error code for an admin queue head overrun to use
I40E_ERR_ADMIN_QUEUE_FULL instead of I40E_ERR_QUEUE_EMPTY.

Fixes: 8db9e2a1b232 ("i40e: base driver")
Cc: stable@dpdk.org
Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: decouple MIB change event from DCB init

There is a need to enable MIB change event, not at the same time as
init_dcb. This patch will serve this requirement.

Modify the i40e_init_dcb to return the correct error when LLDP or DCBX
is not in operational state.

Signed-off-by: Chinh T Cao <chinh.t.cao@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: fix buffer address

The high 32-bits were being set incorrectly in the 'Set Local LLDP MIB'
AQ command (0x0A08). Change it to use the right macro to get the correct
bits.

Fixes: 1fa6324ad30b ("i40e/base: add new interfaces")
Cc: stable@dpdk.org
Signed-off-by: Piotr Azarewicz <piotr.azarewicz@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: increase max VSI count for VFs

For historical reasons, we allowed 3 VSIs per VF, but never used more
than one. Now with ADq, we can use up to 4 VSIs per VF, so change this
value to match.

Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/i40e/base: support NVM feature flags

Extend NVM Update API to support reporting of features
available for the tools.

This change is needed by NVM Update to determine if driver
supports changing NVM layout from structured to flat.

Signed-off-by: Piotr Pietruszewski <piotr.pietruszewski@intel.com>
Signed-off-by: Krzysztof Galazka <krzysztof.galazka@intel.com>
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>

net/octeontx2: fix getting supported packet types

Setting up supported ptypes using testpmd command is failing
because supported_ptypes_get callback will never return ptypes
supported, as NIX_RX_OFFLOAD_PTYPE_F bit in rx_offload_flags
is zero by default in testpmd.

Fixes: d2706e15e6fb ("net/octeontx2: support reduced set of packet types")
Cc: stable@dpdk.org
Signed-off-by: Harman Kalra <hkalra@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

net/qede: remove local bool type

Replace 'typedef int bool' with 'stdbool.h' to avoid possible
multiple definitions of 'bool'.

(Not sure if the previous code is checking for true/false condition.
Recommend careful review on this patch.)

Signed-off-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/fm10k: remove local bool type

Replace 'typedef int bool' with 'stdbool.h' to avoid possible
multiple definitions of 'bool'.

Signed-off-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Acked-by: Xiao Wang <xiao.w.wang@intel.com>

net/e1000: remove local bool type

Replace 'typedef int bool' with 'stdbool.h' to avoid possible
multiple definitions of 'bool'.

Signed-off-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/bnx2x: remove local bool type

Replace 'typedef int bool' with 'stdbool.h' to avoid possible
multiple definitions of 'bool'.

Signed-off-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/vmxnet3: remove local bool type

Replace 'typedef int bool' with 'stdbool.h' to avoid possible
multiple definitions of 'bool'.
Remove 'typedef char Bool' and use 'bool' instead.

Signed-off-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Acked-by: Yong Wang <yongwang@vmware.com>

net/cxgbe: remove local bool type

Replace 'typedef int bool' with 'stdbool.h' to avoid possible
multiple definitions of 'bool'.

Signed-off-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ixgbe: avoid multiple definitions of bool

Compilation issue arises due to multiple definitions of 'bool'
in 'ixgbe_ethdev.h'.

Error:
'/dpdk/drivers/net/ixgbe/ixgbe_ethdev.c: In function
‘ixgbe_dev_setup_link_alarm_handler’:
/dpdk/drivers/net/ixgbe/ixgbe_ethdev.c:4075:43:
error: passing argument 3 of ‘ixgbe_get_link_capabilities’ from
incompatible pointer type [-Werror=incompatible-pointer-types]
   ixgbe_get_link_capabilities(hw, &speed, &autoneg);
                                           ^
In file included from /dpdk/drivers/net/ixgbe/ixgbe_ethdev.c:41:0:
/dpdk/drivers/net/ixgbe/base/ixgbe_api.h:63:5: note: expected
‘bool * {aka int *}’ but argument is of type ‘_Bool *’'

Signed-off-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/octeontx2: improve performance in vector Rx

Use scalar loads instead of vector loads for fields
that don't need any vector operations.

Signed-off-by: Andrew Pinski <apinski@marvell.com>
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>

net/octeontx2: fix VF configuration

Returning -ENOTSUP only in case loopback mode is enabled and
device is VF or SDP.

Fixes: c2c0aa75cd01 ("net/octeontx2: fix loopback config return for VF")
Cc: stable@dpdk.org
Signed-off-by: Harman Kalra <hkalra@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

net/mlx5: support IPv4/IPv6 DSCP rewrite action

This commit add the IPv4/IPv6 DSCP rewrite actions to the PMD code.

Supported actions:
RTE_FLOW_ACTION_TYPE_SET_IPV4_DSCP
RTE_FLOW_ACTION_TYPE_SET_IPV6_DSCP

Signed-off-by: Suanming Mou <suanmingm@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>

ethdev: add IPv4/IPv6 DSCP rewrite action

For some overlay network, such as VXLAN, the DSCP field in the new outer
IP header after VXLAN decapsulation may need to be updated accordingly.

This commit introduce the DSCP modify action for IPv4 and IPv6.

Signed-off-by: Suanming Mou <suanmingm@mellanox.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Ori Kam <orika@mellanox.com>

vhost: fix socket initial value

By default, a vhost socket is created without attaching VDPA device,
this patch fixes the initial value of vdpa_dev_id.

Fixes: b4953225cea4 ("vhost: add APIs for datapath configuration")
Cc: stable@dpdk.org
Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>

net/virtio-user: check file descriptor before closing

Valgrind complains that virtio_user is calling close(-1).
Fix this by adding check in virtio that is similar to existing code.

Fixes: 37a7eb2ae816 ("net/virtio-user: add device emulation layer")
Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>

net/vhost: check creation failure

The function eth_dev_vhost_create() could return errors,
the return value need to be checked.

Fixes: ee584e9710b9 ("vhost: add driver on top of the library")
Cc: stable@dpdk.org
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>

vhost: add dynamic logging system

Currently there are a couple of limitations on the logging system: Most
of the logs are compiled out and both datapath and controlpath logs
share the same loglevel.

This patch tries to help fix that situation by:
- Splitting control plane and data plane logs
- Making control plane logs dynamic while keeping data plane logs
  compiled out by default for log levels lower than the INFO.

As a result, two macros are introduced:
- VHOST_LOG_CONFIG(LEVEL, ...): Config path logging. Level can be
  dynamically controlled by "lib.vhost.config"

- VHOST_LOG_DATA(LEVEL, ...): Data path logging. Level can be
  dynamically controlled by "lib.vhost.data". Every log macro with a
  level lower than RTE_LOG_DP_LEVEL (which defaults to RTE_LOG_INFO)
  will be compiled out.

Signed-off-by: Adrian Moreno <amorenoz@redhat.com>
Acked-by: Tiwei Bie <tiwei.bie@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>

examples/vhost_blk: check unused value on init

Add the assert to handle error.

Coverity issue: 350592
Fixes: c19beb3f38cd ("examples/vhost_blk: introduce vhost storage sample")
Cc: stable@dpdk.org
Signed-off-by: Jin Yu <jin.yu@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>

net/af_xdp: fix redundant check for wakeup need

Function kick_tx() has built-in detection on NEED_WAKEUP flag, so just
call it directly, like elsewhere in the driver.

Fixes: d8a210774e1d ("net/af_xdp: support unaligned umem chunks")
Cc: stable@dpdk.org
Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Tested-by: Ciara Loftus <ciara.loftus@intel.com>

net/axgbe: add xstats

Implements eth dev ops xstats_get, xstats_reset, xstats_get_names,
xstats_get_names_by_id, xstats_get_by_id

Signed-off-by: Chandu Babu N <chandu@amd.com>
Acked-by: Ravi Kumar <ravi1.kumar@amd.com>

net/ena: support Rx offset

Allow the data in the first buffer to be shifted by `pkt_offset` after
the header room. The offset value is provided by `ena_rx_ctx`.

As part of this update, the version of the ENA was upgraded to v2.0.3.

Signed-off-by: Maciej Bielski <mba@semihalf.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>

net/ena/base: upgrade HAL for new HW features

This version of the HAL allows to use the latest HW features, like
rx offsets.

Driver was adjusted to the new version to fix the build.

Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Signed-off-by: Maciej Bielski <mba@semihalf.com>

maintainers: update for flow API

I volunteer to be co-maintainer for the rte_flow lib.

Signed-off-by: Ori Kam <orika@mellanox.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/ice/base: clean up

Couple minor code clean include:
1. Improve debug message format.
2. Add missing macro and comment.
3. Remove unnecessary compile options.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>

net/ice: support 1/10G device IDs

Add support for 1/10G devices.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>

net/ice/base: support adding MAC rules on specific port

ice_add_mac_rule allow user to add rule to port based on
hw->port_info->lport number. Function in some case should
allow user to add filter rule on different port, write
another function which implemented that behaviour.
The same situation is which removing mac function.

Add new API function which allow user to choose port on which
rule going to be added. Leave add mac rule function that always
add rule on hw->port_info->lport to avoid changes in components
which don't need to choose different port. Also add function to
remove rule from specific port.

Alloc more switch_info structs to track separately rules for each
port. Choose switch_info struct basing on logic port number
because in FW added rules are connected with port.

Signed-off-by: Michal Swiatkowski <michal.swiatkowski@intel.com>
Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>

net/ice/base: rework flow director desc preparation

Change internal implementation of how FD filter programming desc
is prepared. This is to minimize the amount of code needed to prep
the FD filter programming desc (avoid memcpy, etc...) and just use
predefined shifts and mask. This type of change are needed to expedite
FD setup during data path (ADQ uses this codepath during initial
flow setup) and it will also be useful when adding side-band
flow-director filter.

Signed-off-by: Kiran Patil <kiran.patil@intel.com>
Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>

net/ice/base: increase PF reset wait timeout

Increase the maximum time that the driver will wait for a PF reset
from 200 milliseconds to 300 milliseconds, to account for possibility
of a slightly longer than expected PF reset.

Fixes: 453d087ccaff ("net/ice/base: add common functions")
Cc: stable@dpdk.org
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>

net/ice/base: fix loop limit

In ice_prot_type_to_id routine, correct the loop limit check
to use ARRAY_SIZE instead of looking for the array element to
have a specific value.

Fixes: fed0c5ca5f19 ("net/ice/base: support programming a new switch recipe")
Cc: stable@dpdk.org
Signed-off-by: Dan Nowlin <dan.nowlin@intel.com>
Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>

net/ice/base: enable clearing of HW tables

Enable the code to clear the HW tables.

Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>

net/ice/base: allow flexbytes matching on header

Change the extraction sequence generated by flow director flexbytes to
use package mac protocol. Without this change data in packet headers
cannot be used for flexbyte matching. The old extraction for flex bytes
started at the beginning of the payload which is after the header.

Signed-off-by: Henry Tieman <henry.w.tieman@intel.com>
Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>

net/ice/base: cleanup format of static const declarations

Use a format consistent with the rest of the code.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>

net/ice/base: do not wait for PE unit to load

When RDMA is not enabled, when checking for completion of a CORER or
GLOBR do not wait for the PE unit to be loaded (indicated by GLNVM_ULD
register's PE_DONE bit being set) since that does not happen and will
cause issues such as failing to initialize the device.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>

net/ice/base: support MAC/VLAN with TCP/UDP in switch

Add a feature to allow user to add switch filter using input like
MAC + VLAN (C-tag only) + L4 (TCP/UDP) port. API "ice_add_adv_rule"
is extended to handle this filter type.

Signed-off-by: Kiran Patil <kiran.patil@intel.com>
Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>

net/ice/base: whitelist register for NVM access

Allow tools to access register offset 0xB8188 (GLGEN_RSTAT) for
NVMUpdate operations.  This is a read-only register, so risk of other
issues stemming from this change is low. Even so, update the write
command to prevent and reject any commands which attempt to write to
this register, just like we do for GL_HICR_EN.

Signed-off-by: Jeb Cramer <jeb.j.cramer@intel.com>
Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Qiming Yang <qiming.yang@intel.com>

app/testpmd: call cleanup on exit

The rte_eal_cleanup code is not exercised by testpmd which
is the most used DPDK test tool. Add a call at end of program.

This helps exercise free and close paths which can
be checked with tools like valgrind.

Fixes: af75078fece3 ("first public release")
Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>

net/hns3: remove unused constants

This patch removed some unused macros defined in hns3_rss.h,
these macros are used to set tuples for abandoned RTE_ETH_FILTER_HASH
in hns3_dev_filter_ctrl().

Signed-off-by: Hao Chen <chenhao164@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>

net/hns3: remove unnecessary assignments in Tx

This patch removes the unnecessary assignment in the '.tx_pkt_burst' ops
implementation function to avoid performance loss.

Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Signed-off-by: Yisen Zhuang <yisen.zhuang@huawei.com>

net/hns3: remove useless variable initializations

This patch removes the redundant initialization of the variable
named ret.

Signed-off-by: Hongbo Zheng <zhenghongbo3@huawei.com>
Signed-off-by: Hao Chen <chenhao164@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>

net/hns3: remove redundant stats reset

This patch removes the redundant statement calling hns3_stats_reset()
to clear statistical information explicitly in the initialization of
VF device, because hardware has been reset by FLR in the initialization
and the initial hardware and software statistics values are 0.

Signed-off-by: Hao Chen <chenhao164@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>

net/hns3: optimize default RSS algorithm

This patch changed the default algorithm of RSS from simle_xor to
toeplitz because toeplitz is used more frequently by upper applications
such as ceph.

Signed-off-by: Hao Chen <chenhao164@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>

net/hns3: remove custom macro for minimum length

This patch replaces custom macro named HNS3_MIN_FRAME_LEN for ethernet
minimum frame length with the macro named RTE_ETHER_MIN_LEN that defined
in DPDK framework.

Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>

net/hns3: modify return value of enable MSI-X

This patch replaces the return value "-1" with "-ENXIO".

Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>

net/hns3: get link state change through mailbox

Currently, firmware adds the function of sending message to PF driver
through mailbox when the link status is changed, hns3 PMD driver can
usually recognize link state change faster through the message.

And in some extreme cases, this way is not faster than existing method
regularly updating link status by issuing the command every second in PF
driver, because of the parallel processing of mailbox and command
messages in firmware. So we reserve updating link status using timers in
PF driver, and add querying link status by issuing command to the
firmware in '.link_update' ops implementation function named
hns3_dev_link_update to solve the out of date link status.

Signed-off-by: Hongbo Zheng <zhenghongbo3@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Signed-off-by: Huisong Li <lihuisong@huawei.com>

net/hns3: support Rx interrupt

This patch adds supports of receive packets through interrupt mode for
hns3 PF/VF driver. The following ops functions should be implemented
defined in the struct eth_dev_ops:
rx_queue_intr_enable
rx_queue_intr_disable

Signed-off-by: Hao Chen <chenhao164@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>

net/mlx5: fix crash when meter action conf is null

When offloading the meter, should check the action
conf and make sure it is valid.

Fixes: f46bf7488705 ("net/mlx5: support meter flow action")
Cc: stable@dpdk.org
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Suanming Mou <suanmingm@mellanox.com>

net/bnxt: fix overwriting error message

In some cases when flow creation fails, we overwrite the specific
error message with a generic error message. This patch fixes it.

Fixes: d24610f7bfda ("net/bnxt: allow flow creation when RSS is enabled")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Lance Richardson <lance.richardson@broadcom.com>

net/bnxt: support flow mark action

Add support for RTE_FLOW_ACTION_TYPE_MARK.
Use the flow_id provided by FW during flow creation to lookup the
mark id provided by the application.

Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Lance Richardson <lance.richardson@broadcom.com>

net/bnxt: fix reusing L2 filter

The software L2 filter was being released back to the free pool,
though it was created in HW and the filter corresponding to an actual
'flow' would have reference to the HW L2 filter.
But if this 'flow were to be deleted, then this HW L2 filter also
would be gone.
Fix this by storing the L2 filter created originally either for an
n-tuple flow or otherwise as part of the vnic's filter list.
This would require the filter_info struct to have a backptr to the
vnic which it came from.
Now that L2 filters can be re-used for an n-tuple filter(s), delete
L2 filter as well so the reference count of an L2 filter (if reused)
can be decremented appropriately.

Fixes: 5c1171c97216 ("net/bnxt: refactor filter/flow")
Cc: stable@dpdk.org
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>

net/bnxt: fix non matching flow hitting filter rule

As part of ntuple filter, we were creating L2 filter with the ntuple
redirect queue resulting in any L2 matching flow getting steered to
this queue. For ntuple filters, we need to create the L2 filter with
default queue. The user specified redirect queue will be set while
creating the ntuple filter in hardware.

Fixes: 5c1171c97216 ("net/bnxt: refactor filter/flow")
Cc: stable@dpdk.org
Signed-off-by: Santoshkumar Karanappa Rastapur <santosh.rastapur@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>

net/bnxt: fix flow flush to sync with flow destroy

Sync flow flush routine with flow destroy so that the operations
performed per flow during a flush are the same as that are done for an
individual flow destroy by having a common function to call for both.
One of the things that was missed in the flow flush routine was the
deletion of the L2 filter that would have been created as part of an
n-tuple filter.
Also, decrement the l2_ref_cnt for a filter in the case of a filter
update as it would've bumped up previously in validate_and_parse_flow()

Fixes: 89278c59d97c ("net/bnxt: fix flow flush handling")
Cc: stable@dpdk.org
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Santoshkumar Karanappa Rastapur <santosh.rastapur@broadcom.com>

net/bnxt: fix Tx queue profile selection

In the case when CoS classification is disabled, driver was iterating
looking for only lossy profiles as that is what is expected to be used
for regular NIC operations. But in certain custom profiles, there were
no lossy profiles configured, only lossless profiles instead.
To handle such cases, it is better to fallback to using the first valid
profile.

Fixes: 698aa7e95325 ("net/bnxt: add code to determine the Tx COS queue")
Cc: stable@dpdk.org
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>