Currently, default MAC address obtained from firmware in PF driver is
directly used by .mac_addr_set ops implementation function when the
rte_eth_dev_start API function is executed. At this moment, if the
default MAC addr isn't an unicast address, it will fail to set default
MAC addr to hardware.
So this patch adds the validity check of default MAC addr in PF driver.
We will use a random unicast address, if the default MAC address
obtained from firmware is not a valid unicast address.
In addition, this patch also adjusts the location of processing default
MAC addr in VF driver so as to increase relevance and readability of the
code.
Fixes: eab21776717e ("net/hns3: support setting VF MAC address by PF driver")
Fixes: d51867db65c1 ("net/hns3: add initialization")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
This patch uses RTE_MAX function in DPDK lib to replace the private
macro named max_t in driver.
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
The FDIR counter was used to count the number of FDIR hit, the maximum
number of the counter is 128 based on kunpeng 920, and it was 256 based
on kunpeng 930.
The firmware is responsible to allocate counters for different PF
devices, so the available counter number of one PF may be bigger than
128.
Currently, there are two places using the counter in driver:
1. Configure the counter. Driver uses the command whose opcode is
HNS3_OPC_FD_AD_OP, now we extend one bit to hold the high bit of
counter-id in the command format.
2. Query the statistic information of the counter. Driver uses the
command whose opcode is HNS3_OPC_FD_COUNTER_OP, now the command
already support 16-bit counter-id.
Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
The new firmware adds the hardware error types reported by MSI-x mode.
These errors are defined as RAS errors in hardware and belong to a
different type from the MSI-x errors processed by the driver.
When hardware detects an error which must be handled by the driver for
device to run properly it reports the error information through the
MSI-x interrupt. After receiving the interrupt reported by the hardware,
the driver queries the error information and identifies the error level,
then rectifies the error. All errors will be logged.
In addition, the hardware may be reset at the function or global level
based on the error level. After the reset is complete, the hardware will
recover to the normal status.
Signed-off-by: Hongbo Zheng <zhenghongbo3@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
There are difference about padding ultra-short frame in Tx procession
for different versions of hardware network engine.
If packet length is less than minimum packet length supported by
hardware in Tx direction, driver need to pad it to avoid error. The
minimum packet length in Tx direction is 33 based on kunpeng 920, and 9
based on kunpeng 930.
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
There are difference about queue's interrupt configurations for
different versions of hardware network engine, such as queue's interrupt
mapping mode, coalesce configuration, etc.
The following uses the configuration differences of the interrupt
mapping mode as an example.
1) For some versions of hardware network engine, such as kunpeng 920,
because of the hardware constraint, we need implement unmmapping
relationship configurations by binding all queues to the last
interrupt vector and reserving the last interrupt vector. This
results in a decrease of the maximum queues when upper applications
call the rte_eth_dev_configure API function to enable Rx interrupt.
2) And for another versions, such as kunpeng 930, hns3 PMD driver can
map/unmmap all interrupt vectors with queues when Rx interrupt is
enabled.
This patch resolves configuration differences about Rx interrupts based
on kunpeng 920 and kunpeng 930.
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
This patch adds getting device capabilities from firmware, so driver can
supply different capabilities and specifications to upper level
applications base on different versions of hardware network engine.
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
The "command mismatch" warning shouldn't be triggered by
VIRTCHNL_OP_EVENT opcode, because the VIRTCHNL_OP_EVENT
opcode is used by PF notifies status change events to VF.
This patch fixed the issue.
Fixes: 837c2ed86e4c ("net/iavf: return error if opcode is mismatched")
Cc: stable@dpdk.org
Signed-off-by: Junyu Jiang <junyux.jiang@intel.com>
Reviewed-by: Qiming Yang <qiming.yang@intel.com>
When PF event VIRTCHNL_EVENT_LINK_CHANGE received, i40evf need update
the link status and issue RTE_ETH_EVENT_INTR_LSC via rte ether device
callback function.
Signed-off-by: Steve Yang <stevex.yang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
On BCM957508-N2100 adapters, FW will not allow any user other
than BMC to shutdown the port. As a result, bnxt_get_hwrm_link_config()
always returns link up.
Because of this, driver will not update the new port configurations
such as speed, autoneg during a port start.
Fixed the condition to invoke bnxt_set_hwrm_link_config() in
bnxt_init_chip().
Fixes: 7bc8e9a227cc ("net/bnxt: support async link notification")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
DPDK does not support RoCE and XDP. The driver should set the
bit 5:4 of the flag to 1 and set bit 6 of the flag in the
HWRM_CFA_L2_FILTER_ALLOC command to disable RoCE and XDP features.
This change will greatly reduce the CFA resource consumption.
Fixes: f92735db1e4c ("net/bnxt: add L2 filter alloc/init/free")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Venkat Duvvuru <venkatkumar.duvvuru@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
rte_pktmbuf_prefree_seg() can return NULL if the mbuf still
has remaining references on it.
Adding a NULL check to prevent segfault.
Fixes: bc4a000f2f53 ("net/bnxt: implement SSE vector mode")
Fixes: 398358341419 ("net/bnxt: support NEON")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
During port start if bnxt_alloc_all_hwrm_stat_ctxs() fails,
in the cleanup path we do see errors like below:
bnxt_hwrm_ring_free(): hwrm_ring_free cp failed. rc:2
bnxt_hwrm_ring_free(): hwrm_ring_free rx failed. rc:2
The reason for this is in bnxt_free_all_hwrm_rings(), the check
is made against "ring->fw_ring_id != INVALID_HW_RING_ID" which
always return true as ring->fw_ring_id is not set to INVALID_HW_RING_ID
while initialising the ring structs.
Fixes: 6eb3cc2294fd ("net/bnxt: add initial Tx code")
Fixes: 2eb53b134aae ("net/bnxt: add initial Rx code")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
The maximum number of TCP segments that can be aggregated & the
maximum number of aggregations the VNIC supports are configured
incorrectly during LRO configuration.
This patch fixes these values.
Fixes: b150a7e7ee66 ("net/bnxt: support LRO on Thor adapters")
Cc: stable@dpdk.org
Signed-off-by: Venkat Duvvuru <venkatkumar.duvvuru@broadcom.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Use "req.tunnel_dst_port_val" in bnxt_hwrm_tunnel_dst_port_alloc()
as big endian since hwrm spec mandates this field in network byte order.
Also, fixed the endianness while parsing the command output.
Fixes: 10d074b2022d ("net/bnxt: support tunneling")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Fix VF representor port add when it's endpoint interface is down.
While adding vf representor port to a bridge, vnic & svif information of
vf representors endpoint(VF) would be needed to program default flow
rules.
However, if the endpoint interface is down when vf representor port is
added, firmware will return invalid vnic & svif information.
This patch fixes the problem by registering to DEFAULT_VNIC_CHANGE
async event and once the async event is received, use the endpoint
information(VF's fid) to fetch it's vnic & svif information and
program the default flow rules.
Fixes: 322bd6e70272 ("net/bnxt: add port representor infrastructure")
Cc: stable@dpdk.org
Signed-off-by: Venkat Duvvuru <venkatkumar.duvvuru@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Set ulp_ctx explicitly to NULL in ulp_ctx_deinit() so that representor
init is aborted if parent ulp context is not initialized.
Also check for the same before creation of port default rules.
Additional checks added in VF rep dev ops for proper parent dev
initialization, to avoid null pointer dereference.
Fixes: 322bd6e70272 ("net/bnxt: add port representor infrastructure")
Fixes: 313ac35ac701 ("net/bnxt: support ULP session manager init")
Cc: stable@dpdk.org
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Venkat Duvvuru <venkatkumar.duvvuru@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
The lookup default action record parif table is updated to catch
the miss path for the entries in the exact match table.
PARIF is handler to a partition of the physical port. The lookup
parif table contains entries for each incoming interface the default
action for the miss entries that do not match the configured rules in
the exact match table. This fix configures those entries in that table.
Fixes: fe82f3e02701 ("net/bnxt: support exact match templates")
Cc: stable@dpdk.org
Signed-off-by: Kishore Padmanabha <kishore.padmanabha@broadcom.com>
Reviewed-by: Shahaji Bhosle <sbhosle@broadcom.com>
Reviewed-by: Mike Baucom <michael.baucom@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Configure loopback parif for full offload egress flows.
PARIF is handler to a partition of the physical port.
The full offload egress flows for the VF rep interface must
use loopback parif to offload missed flows. The miss
flow path for the VF rep interface has to be loopback interface
parif entry and for the non-VF rep interface it has to be the
interface's parif entry.
Fixes: fe82f3e02701 ("net/bnxt: support exact match templates")
Cc: stable@dpdk.org
Signed-off-by: Kishore Padmanabha <kishore.padmanabha@broadcom.com>
Reviewed-by: Shahaji Bhosle <sbhosle@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
In bnxt_validate_and_parse_flow_type
vxlan_spec cannot be NULL since its already being accessed
before. Remove the redundant NULL check.
Signed-off-by: Gaurav Singh <gaurav1086@gmail.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
The function rte_malloc() could return NULL, the return
value need to be checked.
Fixes: b7778e8a1c00 ("net/bnxt: refactor to properly allocate resources for PF/VF")
Cc: stable@dpdk.org
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
When freeing a vf_info, we should free the 'vlan_as_table'
and 'vlan_table' for the vf_info.
Fixes: b7778e8a1c00 ("net/bnxt: refactor to properly allocate resources for PF/VF")
Cc: stable@dpdk.org
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
The af_xdp rx function was returning a negative value on error, when an
unsigned value is expected. Fix this.
Fixes: d8a210774e1d ("net/af_xdp: support unaligned umem chunks")
Cc: stable@dpdk.org
Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Offensive terms should be replaced with more inclusive engineering
terms.
Deprecated terms include "master/slave" and "blacklist/whitelist".
Signed-off-by: Yuying Zhang <yuying.zhang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
For GTPU rule without extend header the training packet
for FDIR is different. This patch enable these case.
Fixes: 934fd00c9389 ("net/ice/base: fix GTPU IP hash")
Cc: stable@dpdk.org
Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Tested-by: Zhimin Huang <zhiminx.huang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
GTPU TEID hash should only be enabled when ETH_RSS_GTPU is required.
And the hash parser should not restrict the combined usage of protocol.
Fixes: e7cc68c70736 ("net/ice: fix GTPU TEID hash")
Cc: stable@dpdk.org
Signed-off-by: Jeff Guo <jia.guo@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Add RXDID #25 to support Auxiliary IP Offset Rx descriptor, including
FlexiMD.4: Outer/Single IPv4 Header offset
FlexiMD.5: Outer/Single IPv6 Header offset
And parse the valid IP Offset into mbuf by flexible descriptor
section via devargs "proto_xtr" with "proto_xtr=ip_offset".
Signed-off-by: Junfeng Guo <junfeng.guo@intel.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
According to ABI policy, the internal functions should have the new tag
__rte_internal and linked into INTERNAL sessions.
Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
The vf index in rte_flow command need to check in order that not larger
than actual vf number. Also change some error log for wrong action.
Fixes: 829c3106812d ("net/ice: enable switch flow on DCF")
Cc: stable@dpdk.org
Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Tested-by: Nannan Lu <nannan.lu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Add AQ command "NVM update in process"
to replace the original AQ command "NVM progress".
Signed-off-by: Jaroslaw Ilgiewicz <jaroslaw.ilgiewicz@intel.com>
Signed-off-by: Guinan Sun <guinanx.sun@intel.com>
Acked-by: Jeff Guo <jia.guo@intel.com>
This patch adds support for disabling unused ports.
Signed-off-by: Damian Milosek <damian.milosek@intel.com>
Signed-off-by: Guinan Sun <guinanx.sun@intel.com>
Acked-by: Jeff Guo <jia.guo@intel.com>
Fix possible uninitialized variable in i40e in the i40e_get_lpi_counters
function.
Fixes: 429bdc0cd967 ("net/i40e/base: add function to read LPI counters")
Cc: stable@dpdk.org
Signed-off-by: Adam Ludkiewicz <adam.ludkiewicz@intel.com>
Signed-off-by: Guinan Sun <guinanx.sun@intel.com>
Acked-by: Jeff Guo <jia.guo@intel.com>
Upcoming FW increment API version to 1.11 due to new bit and new
fields in the Replace Cloud Filters AQ command.
Signed-off-by: Jacek Naczyk <jacek.naczyk@intel.com>
Signed-off-by: Guinan Sun <guinanx.sun@intel.com>
Acked-by: Jeff Guo <jia.guo@intel.com>
This patch adds the new filter types needed for custom cloud filters.
These custom cloud filters will route traffic to VFs based on the
dst IP for both tunneled and non-tunneled packets.
Signed-off-by: Harshitha Ramamurthy <harshitha.ramamurthy@intel.com>
Signed-off-by: Guinan Sun <guinanx.sun@intel.com>
Acked-by: Jeff Guo <jia.guo@intel.com>
Add I40E_RX_PTYPE_PARSER_ABORTED definition, so i40e driver will know
opcode for parser aborted packets.
Without this definition driver would have to rely on magic numbers.
Signed-off-by: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>
Signed-off-by: Guinan Sun <guinanx.sun@intel.com>
Acked-by: Jeff Guo <jia.guo@intel.com>
Starting with API version 1.10 firmware for X722 devices has ability
to change FEC settings in PHY. Code added in this patch
checks API version and sets appropriate capability flag.
Signed-off-by: Dawid Lukwinski <dawid.lukwinski@intel.com>
Signed-off-by: Guinan Sun <guinanx.sun@intel.com>
Acked-by: Jeff Guo <jia.guo@intel.com>
The Flexible Metadata #4 and #5 in the Rx Flex Descriptor are defined to
extract the protocol specified fields or its offset. Its function relays
on the DDP package support in hardware.
Optimize to only check the hardware support when the user specifies the
'proto_xtr' devargs. And not require all the types need to be support in
hardware, otherwise if new protocol extraction type is introduced, it is
hard to maintain the compatibility. Just check the type support in need.
And the protocol IDs are 8 bits length, so the uint8_t is the right type
to be used. Also introduce the 'opcode' variable to specify the metadata
extraction type: it can be protocol fields or offset report.
Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Set RTE_ETH_DEV_CLOSE_REMOVE upon probe so all the private resources
for the port can be freed by rte_eth_dev_close().
Signed-off-by: Steve Yang <stevex.yang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
File mlx5_vlan.c contains Netlink APIs (Linux dependent) as part of VM
workaround implementation. Move this implementation to file
linux/mlx5_vlan_os.c. To remove Netlink dependency in header files
change pointer of type 'struct mlx5_nl_vlan_vmwa_context *' to 'void *'.
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
When updating a queue vlan stripping offload - either the WQ is modified
in Verbs or the RQ is modified in DevX. Add a vlan stripping modify
callback to 'struct mlx5_obj_ops' and assign it with the specific Verbs
and DevX implementations: 'rxq_obj_modify_wq_vlan_strip' and
'rxq_obj_modify_rq_vlan_strip' respectively.
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Several DV-based structs of type 'struct mlx5dv_devx_XXX' are replaced
with 'void *' to enable compilation under non-Linux operating systems.
New getter functions were added to retrieve the specific fields that
were previously accessed directly.
Replaced structs:
'struct mlx5dv_pp *'
'struct mlx5dv_devx_event_channel *'
'struct mlx5dv_devx_umem *'
'struct mlx5dv_devx_uar *'
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Flow metering is supported only in direct rules (DR). Currently the APIs
of meter actions create and modify are under #ifdef
HAVE_MLX5_DR_CREATE_ACTION_FLOW_METER, while detaching the meter action
is executed unconditionally. This commit adds the same ifdef to API
mlx5_flow_meter_detach().
This commit avoids compilation failure of non-Linux operating systems
which do not support DR.
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Remove utility macros INFO, WARN, ERROR. They are not in use and
conflict with identical definitions when compiled under Windows.
Fixes: 80f2d0ed7ff9 ("net/mlx5: add hardware flow debug dump")
Cc: stable@dpdk.org
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
This commit adds Linux implementation of routine mlx5_os_mac_addr_flush
as wrapper to Netlink API to avoid direct calls under non-Linux
operating systems.
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
