The variable build is already initialized as true in
drivers/meson.build. Duplicate initializations can be removed from mlx.
When the variable build is set to false, it is easier to call
subdir_done() than branch the rest of the code on build condition.
Suggested-by: Bruce Richardson <bruce.richardson@intel.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Rosen Xu <rosen.xu@intel.com>
The maximum MTU for af_xdp zero copy is equal to the page size less the
frame overhead introduced by AF_XDP (XDP HR = 256) and DPDK (frame
headroom = 320). The patch updates this value to reflect this.
This change also makes it possible to remove unneeded constants for both
zero-copy and copy mode.
Fixes: d8a210774e1d ("net/af_xdp: support unaligned umem chunks")
Cc: stable@dpdk.org
Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
The fill queue addresses should start at the beginning of the mempool
object instead of the beginning of the mbuf. This is because the umem
frame headroom includes the mp hdrobj size. Starting at this point
ensures AF_XDP doesn't write past the available room in the frame, in
the case of larger packets which are close to the size of the mbuf.
Fixes: d8a210774e1d ("net/af_xdp: support unaligned umem chunks")
Cc: stable@dpdk.org
Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
The previous frame size calculation incorrectly used
mb_pool->private_data_size and didn't include mb_pool->header_size.
Instead of performing a manual calculation, use the
rte_mempool_calc_obj_size API to determine the frame size.
The previous frame headroom calculation also incorrectly used
mb_pool->private_data_size and didn't include mb_pool->header_size or
the mbuf priv size. Fix this.
Fixes: d8a210774e1d ("net/af_xdp: support unaligned umem chunks")
Cc: stable@dpdk.org
Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
The meter flows are split into three subflows each, the prefix subflow
with meter action color the packet, the meter subflow filters out the
colored packets, the suffix subflow applies all the remaining actions
to the passed packets. The tag header modify action is added to the
prefix subflow to make the suffix subflow to match the packets from the
prefix subflow.
Currently, the tag header modify action is added at the beginning in the
prefix subflow even before decap action. The header modify action does
not make sense to the later decap action, so the flow create will be
validated as incorrect flow rule and failed.
Move the tag header modify action just before meter action in the prefix
subflow to make the flow with decap action to do the decap first, then
do the tag and meter to fix that issue.
Fixes: 9ea9b049a960 ("net/mlx5: split meter flow")
Cc: stable@dpdk.org
Reported-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Signed-off-by: Suanming Mou <suanmingm@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Currently MLX5 PMD can't match on untagged packets specifically.
Tagged traffic still hits the flows intended for untagged packets.
If the flow has ETH, it will catch all matching packets, tagged
and untagged.
The solution is to use cvlan_tag bit.
If mask=1 and value=0 it matches on untagged traffic.
If mask=1 and value=1 it matches on tagged traffic.
This is the kernel implementation.
This patch updated MLX5 PMD to set cvlan_tag mask and value according
to flow rule contents.
This update is relevant when using DV flow engine (dv_flow_en=1).
See example at https://doc.dpdk.org/guides/nics/mlx5.html#limitations.
Fixes: fc2c498ccb94 ("net/mlx5: add Direct Verbs translate items")
Cc: stable@dpdk.org
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
The blocker should take FDB into consideration, since FDB all directions
have transfer ingress in it.
Fixes: 55060e62e4d2 ("net/mlx5: block push VLAN action on Rx")
Cc: stable@dpdk.org
Signed-off-by: Wisam Jaddo <wisamm@mellanox.com>
Acked-by: Raslan Darawsheh <rasland@mellanox.com>
This patch fixes (Logically dead code) coverity issue.
Coverity issue: 353624
Fixes: ba7b12dd64e4 ("net/ixgbe: fix link up in FreeBSD")
Cc: stable@dpdk.org
Signed-off-by: Shougang Wang <shougangx.wang@intel.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
Replace checking against 65535 limit,
with a simpler form using RTE_MIN and UINT16_MAX macros.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Matan Azrad <matan@mellanox.com>
To provide the better PCIe bandwidth utilization the ConnectX-4LX
NIC supports the multi-packet write (MPW) sessions allowing to
pack multiple packets into one descriptor (WQE). This is legacy
feature and it has some limitations on the packets and data
description segments. To provide the best performance all inline
packets must be put into shared data segment and the total length
of MPW session must be limited. The limit is controlled with
txq_inline_mpw devarg.
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>
The encapsulation and decapsulation actions are divided into 2 types:
L2 and L3.
In order to configure L3 xcapsulation actions the user should use both
RAW_DECAP and RAW_ENCAP and setting the appropriated data sizes in
their action configuration structures.
The PMD flow validation wrongly didn't detect the RAW_DECAP
and RAW_ENCAP combination to distinguish between L3_DECAP and L3_ENCAP.
Thus, some xcapsulation related validation failed.
For example, when configuring modify header action before L3_DECAP.
Simplify the xcapsulation defines and fix the L3 xcapsulation detection
using the action configuration data sizes.
By the way, add the hairpin validation in this area.
Fixes: d85c7b5ea59f ("net/mlx5: split hairpin flows")
Fixes: 8ba9eee4ce32 ("net/mlx5: add raw data encap/decap to Direct Verbs")
Cc: stable@dpdk.org
Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
If application is using link state interrupt, the correct link state
needs to be filled in when device is started. This is similar to
how virtio updates link information.
Fixes: 4e9c73e96e83 ("net/netvsc: add Hyper-V network device")
Cc: stable@dpdk.org
Reported-by: Mohammed Gamal <mgamal@redhat.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Tested-by: Mohammed Gamal <mgamal@redhat.com>
Merge of ESP and L2TP code in i40e_fdir.c added checks on
cus_pctype->index which caused flow creation for ESP to fail.
Added fill_ipv4_function()
Refactored code to have one path for customized and non customized
pctype's.
Fixes: c5f8365bc85d ("net/i40e: support flow director for L2TPv3 over IP")
Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com>
Acked-by: Rory Sexton <rory.sexton@intel.com>
This patch fixes (Logically dead code) coverity issue.
Coverity issue: 353613
Fixes: 3c4270187518 ("net/ixgbe: support VF MAC address add/remove")
Signed-off-by: Guinan Sun <guinanx.sun@intel.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
The original PASSTHRU implementation is forwarding to queue 0. The
corrected implementation is forwarding to the next stage filter.
Fixes: f5cafa961fae ("net/ice: add flow director create and destroy")
Cc: stable@dpdk.org
Signed-off-by: Yahui Cao <yahui.cao@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
In case of a buffer allocation failure, we reattempt buffer allocation
before the Rx handler exits. We were not attempting this when producer
index is greater than the number of buffers to allocate. Fixed it with
correct checks.
Fixes: d9dd0b29ed31 ("net/bnxt: fix Rx handling and buffer allocation logic")
Cc: stable@dpdk.org
Signed-off-by: Santoshkumar Karanappa Rastapur <santosh.rastapur@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
bnxt_rss_ctxts() function is declared in header file bnxt.h,
implemented in bnxt_ethdev.c, and called only in bnxt_ethdev.c.
Also many functions are declared in header file bnxt_hwrm.h,
implemented in bnxt_hwrm.c, and called only in bnxt_hwrm.c.
This patch moves these function declarations into bnxt_ethdev.c
and bnxt_hwrm.c, as static functions.
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
During live FW upgrade or error recovery, if restoring the filter
settings fail after port start, driver invokes bnxt_uninit_resources()
only. Fix it to invoke bnxt_dev_stop_op() first before calling
bnxt_uninit_resources().
Fixes: df6cd7c1f73a ("net/bnxt: handle reset notify async event from FW")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Rahul Gupta <rahul.gupta@broadcom.com>
PF driver has to register for the debug notification async event
with firmware in the HWRM_FUNC_DRV_RGTR command.
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
The debug notifications are not functional in nature, they should
only have diagnostic value. Other than logging to system log,
drivers shall not take any other functional action based on this
async event.
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Initially when driver is loading, there is no HWRM timeout configured
by FW, the VER_GET command needs use default timeout as 500ms and
while recovering from fatal/non-fatal FW error, it should use timeout
as 50ms.
Fixes: 458f0360e8dc ("net/bnxt: get default HWRM command timeout from FW")
Cc: stable@dpdk.org
Signed-off-by: Rahul Gupta <rahul.gupta@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>
VFs and multifunction PFs do not have the privilege to change
link configuration. We force the physical link down as a part
of device stop only for single physical function(SPF).
This change also helps in eliminating the logs when a VF port
is stopped:
"Port 0: link state change event"
"bnxt_print_link_info(): Port 0 Link Up - speed 25000 Mbps - full-duplex"
Fixes: 316e412299fd ("net/bnxt: fix crash when closing")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
The header file rte_config.h is always included by make or meson.
If required in an exported API header file, it must be included
in the public header file for external applications.
In the internal files, explicit include of rte_config.h is useless,
and can be removed.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Matan Azrad <matan@mellanox.com>
Acked-by: David Marchand <david.marchand@redhat.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Rather than assuming all compilers support the -flax-vector-extensions
flag, we should test this before using it, thereby potentially avoiding
warnings.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Three warnings are commonly disabled in DPDK with make and meson:
* address-of-packed-member
always disabled
* missing-field-initializers
disabled with meson
disabled with make + clang or make + gcc < 4.7
disabled with make + gcc <= 5 for test files and event drivers
* packed-not-aligned
disabled with meson
This change is removing exceptions for missing-field-initializers.
As it is always disabled, some redundant configs are cleaned up.
Now the situation is:
* address-of-packed-member
always disabled
* missing-field-initializers
always disabled
* packed-not-aligned
disabled with meson
It could alternatively be decided to disable missing-field-initializers
only for old gcc (< 6).
The warning packed-not-aligned is not modified in this change.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Tag type configuration for the inline processed packets is set during
ethdev configuration, it might conflict with tag type configuration
done during Rx adapter configuration which would be setup later.
This conflict is fixed as part of flow rule creation by updating
tag type config of inline same as Rx adapter configured tag type.
Signed-off-by: Ankur Dwivedi <adwivedi@marvell.com>
Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Signed-off-by: Archana Muniganti <marchana@marvell.com>
Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Added new flag for SECURITY in compiler optimized Tx fastpath
framework. With this, compiler autogenerates functions which
have security enabled.
Signed-off-by: Ankur Dwivedi <adwivedi@marvell.com>
Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Signed-off-by: Archana Muniganti <marchana@marvell.com>
Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com>
lookup_mem provides fast accessing of data path fields.
Storing sa indices in lookup_mem which are required in
inline rx data path.
Signed-off-by: Ankur Dwivedi <adwivedi@marvell.com>
Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Signed-off-by: Archana Muniganti <marchana@marvell.com>
Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Adding the infrastructure to save one opaque pointer in idev and
implement the consumer-producer in the PMDs which uses it accordingly.
Signed-off-by: Ankur Dwivedi <adwivedi@marvell.com>
Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Signed-off-by: Archana Muniganti <marchana@marvell.com>
Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Previous patch changed the format of struct
mlx5_flow_dv_modify_hdr_resource, to use a flexible array for
modification actions.
In __flow_dv_translate() a union was defined with item of this struct,
and an array of maximal possible size.
Array elements are filled in several functions.
In function flow_dv_convert_action_set_reg(), array element is filled
partially, while the other fields of this array element are left
uninitialized.
This may cause failure of flow_dv_modify_hdr_resource_register()
when calling driver function with the 'dirty' array.
This patch updates flow_dv_convert_action_set_reg(), setting the
selected array element fields while clearing the other fields.
Other functions that fill the same array elements are also updated
for clarity and proofing future use.
Fixes: 024e95759c16 ("net/mlx5: fix modify actions support limitation")
Cc: stable@dpdk.org
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
The mpx5 PMD maintains the list of devices for those the memory
operation callback routines must be invoked to keep the device MRs (MR
is the entity backing the hardware DMA transactions) consistent with the
mapped memory.
Each device context in the list is protected with dedicated lock on per
device basis, which might be taken inside the callback routine.
When device is closing the PMD frees all MRs by calling
mlx5_mr_release(), that might call rte_free() under the taken device
lock. If this rte_free call triggers the entire memory segment freeing
it, in its turn, invokes the callback routine and attempt to take the
lock inside this one causes the deadlock.
The patch proposes the remove the device from the callback list first
and then call mlx5_mr_release() and free the remaining device MRs
explicitly.
Fixes: 0e3d0525b2f2 ("net/mlx5: fix memory event callback list")
Cc: stable@dpdk.org
Signed-off-by: Michael Baum <michaelba@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
This information is useful or needed for user applications as t-rex.
Signed-off-by: Július Milan <jmilan.dev@gmail.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Get a burst mode information for Rx/Tx queues in mlx5.
Provide callback functions to show this information in
a "show rxq info" and "show txq info" output.
Signed-off-by: Alexander Kozyrev <akozyrev@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Earlier after a successful mac_addr_add operation, index was returned
by underlying layer which was unused but same as provided by DPDK API.
So API is enhanced to use application provided index location to add
MAC address entry.
Fixes: e4373bf1b3f5 ("net/octeontx: add unicast MAC filter")
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Acked-by: Harman Kalra <hkalra@marvell.com>
MAC address table is allocated during octeontx device create and
same is used to maintain list of MAC address associated to port.
This table is not getting freed niether in case of error nor during
graceful shutdown of port.
Patch fixes memory required memory for both the cases as mentioned.
Fixes: f18b146c498d ("net/octeontx: create ethdev ports")
Cc: stable@dpdk.org
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Acked-by: Harman Kalra <hkalra@marvell.com>
Tx flow controlled is disabled in the Ax silicon version due to an errata.
This errata is not applicable for HIGIG Tx flow control, therefore
not enabling in HIGIG case.
Fixes: 602009ee2dfb ("net/octeontx2: support HIGIG2")
Cc: stable@dpdk.org
Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
The interrupt vector which bind to queues should not be larger than
the max available vector. It will cause port start failed. This patch
changed the judgement condition of the limited vector id. It can
effectively avoid vector id out of range.
Fixes: 6a6cf5f88b4a ("net/i40e: enable multi-queue Rx interrupt for VF")
Signed-off-by: Lunyuan Cui <lunyuanx.cui@intel.com>
Reviewed-by: Qiming Yang <qiming.yang@intel.com>
The patch distinguishes fdir rules for GTPU with or without
extend header, so flow to match below patterns can be created
correctly.
1. eth / ipv4 / udp / gtpu teid is 10 / ...
2. eth / ipv4 / udp / gtpu teid is 10 / gtp_psc / ...
3. eth / ipv4 / udp / gtpu / gtp_psc qfi is 10 / ...
4. eth / ipv4 / udp / gtpu teid is 10 / gtp_psc is 10 / ...
Fixes: efc16c621415 ("net/ice: support flow director GTPU tunnel")
Cc: stable@dpdk.org
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
Use PCI_PRI_FMT instead of "%04d:%02d:%02d:%d" print format.
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
Currently, when using VXLAN-GPE or VXLAN item in the flow
both are being treated the same with flags 0x8 in VXLAN
header. Which mean the matching of the item VXLAN-GPE
will match any VXLAN packet.
This fixes the translation of VXLAN GPE item into PMD flow
item. Which will by default set the flags to VXLAN-GPE
to be 0xc.
Fixes: 3d69434113d1 ("net/mlx5: add Direct Verbs validation function")
Cc: stable@dpdk.org
Signed-off-by: Raslan Darawsheh <rasland@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
