mlx5_rxq_start() allocates rxq_ctrl->obj and frees it on failure,
but did not set it to NULL. Later mlx5_rxq_release() could not recognize
this object is already freed and attempted to release its resources,
resulting in a crash:
Configuring Port 0 (socket 0)
mlx5_common: Failed to create RQ using DevX
mlx5_common: Can't create DevX RQ object.
mlx5_net: Port 0 Rx queue 0 RQ creation failure.
Segmentation fault
Set rxq_ctrl->obj to NULL after it is freed to skip resource release.
Fixes: 1260a87b28 ("net/mlx5: share Rx control code")
Cc: stable@dpdk.org
Signed-off-by: Dmitry Kozlyuk <dkozlyuk@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
The RSS configuration in a policy action container was a pointer
inside a union, and the pointer area could be used as other fate
action. In the current implementation, the RSS of the green color
was prior to that of the yellow color. There was a high possibility
the pointer was considered as the RSS and result in a error flow
expansion when only the yellow color had the RSS action.
The check of the fate action type should also be done to get rid of
the misjudgment.
Fixes: b38a12272b ("net/mlx5: split meter color policy handling")
Cc: stable@dpdk.org
Signed-off-by: Bing Zhao <bingz@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Verbs API doesn't support device port number larger than 255 by design.
To support more VF or SubFunction port representors, forces DevX API
check when max Verbs device link ports larger than 255.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
Verbs API does not support Infiniband device port number larger 255 by
design. To support more representors on a single Infiniband device DevX
API should be engaged.
While creating Send Queue (SQ) object with Verbs API, the PMD assigned
IB device port attribute and kernel created the default miss flows in
FDB domain, to redirect egress traffic from the queue being created to
representor appropriate peer (wire, HPF, VF or SF).
With DevX API there is no IB-device port attribute (it is merely kernel
one, DevX operates in PRM terms) and PMD must create default miss flows
in FDB explicitly. PMD did not provide this and using DevX API for
E-Switch configurations was disabled.
The default miss FDB flow matches E-Switch manager vport (to make sure
the source is some representor) and SQn (Send Queue number - device
internal queue index). The root flow table managed by kernel/firmware
and it does not support vport redirect action, we have to split the
default miss flow into two ones:
- flow with lowest priority in the root table that matches E-Switch
manager vport ID and jump to group 1.
- flow in group 1 that matches E-Switch manager vport ID and SQn and
forwards packet to peer vport
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
When creating internal transfer flow on root table with lowest
priority, the flow was created with max UINT32_MAX priority. It is wrong
since the flow is created in kernel and max priority supported is 16.
This patch fixes this by adding internal flow check.
Fixes: 5f8ae44dd4 ("net/mlx5: enlarge maximal flow priority")
Cc: stable@dpdk.org
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
Extends txq flow pattern to support both hairpin and regular txq.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
For egress packet on representor, the vport ID in transport domain
is E-Switch manager vport ID since representor shares resources of
E-Switch manager. E-Switch manager vport ID and Tx queue internal device
index are used to match representor egress packet.
This patch adds flow item port ID match on E-Switch manager.
E-Switch manager vport ID is 0xfffe on BlueField, 0 otherwise.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
To detect number flow Verbs flow priorities, PMD try to create Verbs
flows in different priority. While Verbs is not designed to support
ports larger than 255.
When DevX supported by kernel driver, 16 Verbs priorities must be
supported, no need to create Verbs flows.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
IB spec doesn't allow 255 ports on a single HCA, port number of 256 was
cast to u8 value 0 which invalid to ibv_query_port()
This patch invokes Netlink API to query port state when port number
greater than 255.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
Add support for PPP over L2TPv2 over UDP protocol RSS Hash based
on inner IP src/dst address and TCP/UDP src/dst port.
Patterns are listed below:
eth/ipv4(6)/udp/l2tpv2/ppp/ipv4(6)
eth/ipv4(6)/udp/l2tpv2/ppp/ipv4(6)/udp
eth/ipv4(6)/udp/l2tpv2/ppp/ipv4(6)/tcp
Signed-off-by: Wenjun Wu <wenjun1.wu@intel.com>
Signed-off-by: Jie Wang <jie1x.wang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
CNF10KA does not differ it terms of RVU resources from
CN10KA platform hence add it to list of devices respective
drivers support.
Otherwise devices on CNF10KA are not probed even though
compatible drivers exist.
Signed-off-by: Tomasz Duszynski <tduszynski@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Expand the use of mempool registration to MR management for other
drivers.
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Since MR management has moved to the common area, there is no longer a
need for the DMA map and unmap function for each driver.
This patch share those functions. For most drivers it supports these
operations for the first time.
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Add global shared MR cache as a field of common device structure.
Move MR management to use this global cache for all drivers.
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Add function for global shared MR cache structure initialization.
This function include:
- btree initialization.
- set callbacks for reg and dereg MR.
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Add function for MR control structure initialization.
This function include:
- btree initialization.
- dev_gen_ptr initialization.
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
This patch remove two redundant things from MR file:
1. mr_find_contig_memsegs_data structure which is moved to common file
before.
2. External memory mechanism - mlx5_tx_update_ext_mp function.
Since commit [1] which added support for DMA map and unmap, external
mem must be configured by the user using rte_mem_map function and no
need to handle this in pmd.
[1]
commit 989e999d93
("net/mlx5: support PCI device DMA map and unmap")
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Add HCA attributes structure as a field of device config structure.
It query in common probing, and updates the timestamp format fields.
Each driver use HCA attributes from common device config structure,
instead of query it for itself.
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Create shared Protection Domain in common area and add it and its PDN as
fields of common device structure.
Use this Protection Domain in all drivers and remove the PD and PDN
fields from their private structure.
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Create shared context device in common area and add it as a field of
common device.
Use this context device in all drivers and remove the ctx field from
their private structure.
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Device configure structure has flag named devx as same as SH structure
with the same meaning.
Remove the flag from the configuration structure and move all the
usages to the SH flag.
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Add device configure structure and function to parse user device
arguments into it.
Move parsing and management of relevant device arguments to common.
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Create MACRO definitions file in the common driver as preparation for MR
and basic probe sharing.
Move relevant definitions from the net driver to the above file.
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Create common probing structure that includes, for now, basic probing
information detected by the common driver and share it with all the
internal drivers.
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
In device initialization, the driver registers to free hugepages events.
When hugepage is released, this callback frees all its related MRs.
In Windows initialization, this callback is not registered what may
cause to use invalid memory.
This patch adds memory event callback registration in Windows
initialization.
Fixes: 980826dc6f ("net/mlx5: probe on Windows")
Cc: stable@dpdk.org
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Sideband queue need to be initialized when device is initialized.
Otherwise the calling to function "ice_init_ctrlq" may fail.
This patch fixes it.
Fixes: 97f4f78bbd ("net/ice/base: add functions for device clock control")
Cc: stable@dpdk.org
Signed-off-by: Dapeng Yu <dapengx.yu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Meters are configured per flow using rte_flow_create API.
Patch adds support for destroy operation for meter action
applied on the flow.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Meters are configured per flow using rte_flow_create API.
Implement support for meter action applied on the flow.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Implement API to read and update stats corresponding to
given meter instance for CNXK platform.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Implement API to update DSCP table for pre-coloring for
incoming packet per nixlf for CNXK platform.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Implement API to enable or disable meter instance for
CNXK platform.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Implement API to delete meter instance for CNXK platform.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Implement API to create meter instance for CNXK platform.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Implement API to delete meter policy for CNXK platform.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Implement API to add meter policy for CNXK platform.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Implement API to validate meter policy for CNXK platform.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Implement API to delete meter profile for CNXK platform.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Implement API to add meter profile for CNXK platform.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Implement ethdev operation to get meter capabilities for
CNXK platform.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
To enable support for ingress meter, supported operations
are exposed for CNXK platform.
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Using fast metadata and userdata flags instead of
driver callbacks for set_pkt_metadata and
get_userdata in inline IPsec.
Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
Acked-by: Anoob Joseph <anoobj@marvell.com>
The RSS expansion algorithm is using a graph to find the possible
expansion paths. The current implementation does not differentiate
between standard (L2) VXLAN and L3 VXLAN. As result the flow is expanded
with all possible paths.
For example:
testpmd> flow create... / vxlan / end actions rss level 2 / end
It is currently expanded to the following paths:
ETH IPV4 UDP VXLAN END
ETH IPV4 UDP VXLAN ETH IPV4 END
ETH IPV4 UDP VXLAN ETH IPV6 END
ETH IPV4 UDP VXLAN IPV4 END
ETH IPV4 UDP VXLAN IPV6 END
The fix is to adjust the expansion according to the outer UDP destination
port. In case flow pattern defines a match on the standard udp port, 4789,
or does not define a match on the destination port, which also implies
setting the standard one, the expansion for the above example will be:
ETH IPV4 UDP VXLAN END
ETH IPV4 UDP VXLAN ETH IPV4 END
ETH IPV4 UDP VXLAN ETH IPV6 END
Otherwise, the expansion will be:
ETH IPV4 UDP VXLAN END
ETH IPV4 UDP VXLAN IPV4 END
ETH IPV4 UDP VXLAN IPV6 END
Fixes: f4f06e3615 ("net/mlx5: add flow VXLAN item")
Cc: stable@dpdk.org
Signed-off-by: Lior Margalit <lmargalit@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Rx descriptor is 16B/32B in size. If the DD bit is set, it indicates
that the rest of the descriptor words have valid values. Hence, the
word containing DD bit must be read first before reading the rest of
the descriptor words.
In NEON vector PMD, vector load loads two contiguous 8B of
descriptor data into vector register. Given vector load ensures no
16B atomicity, read of the word that includes DD field could be
reordered after read of other words. In this case, some words could
contain invalid data.
Read barrier is added after read of qword1 that includes DD field.
And qword0 is reloaded to update vector register. This ensures
that the fetched data is correct.
Testpmd single core test on N1SDP/ThunderX2 showed no performance drop.
Fixes: ae0eb310f2 ("net/i40e: implement vector PMD for ARM")
Cc: stable@dpdk.org
Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
To keep flow format uniform with ice, this patch adds support for
this RSS rule:
flow create 0 ingress pattern eth / ipv6 / ipv6_frag_ext / end \
actions rss types ipv6-frag end queues end queues end / end
Fixes: ef4c16fd91 ("net/i40e: refactor RSS flow")
Cc: stable@dpdk.org
Signed-off-by: Alvin Zhang <alvinx.zhang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
The common header file for vectorization is included in multiple files,
and so must use macros for the current compilation unit, rather than the
compiler-capability flag set for the whole driver. With the current,
incorrect, macro, the AVX512 or AVX2 flags may be set when compiling up
SSE code, leading to compilation errors. Changing from "CC_AVX*_SUPPORT"
to the compiler-defined "__AVX*__" macros fixes this issue. In addition,
splitting AVX-specific code into the new ice_rxtx_common_avx.h header
file to avoid such bugs.
Bugzilla ID: 788
Fixes: a4e480de26 ("net/ice: optimize Tx by using AVX512")
Fixes: 20daa1c978 ("net/ice: fix crash in AVX512")
Cc: stable@dpdk.org
Signed-off-by: Leyi Rong <leyi.rong@intel.com>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
The common header file for vectorization is included in multiple files,
and so must use macros for the current compilation unit, rather than the
compiler-capability flag set for the whole driver. With the current,
incorrect, macro, the AVX512 or AVX2 flags may be set when compiling up
SSE code, leading to compilation errors. Changing from "CC_AVX*_SUPPORT"
to the compiler-defined "__AVX*__" macros fixes this issue. In addition,
splitting AVX-specific code into the new i40e_rxtx_common_avx.h header
file to avoid such bugs.
Bugzilla ID: 788
Fixes: 0604b1f220 ("net/i40e: fix crash in AVX512")
Cc: stable@dpdk.org
Signed-off-by: Leyi Rong <leyi.rong@intel.com>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
After the meter policies are created, they are not freed on device
close.
This patch fixes it.
Fixes: 5f0d54f372 ("ethdev: add pre-defined meter policy API")
Cc: stable@dpdk.org
Signed-off-by: Dapeng Yu <dapengx.yu@intel.com>
Acked-by: Jasvinder Singh <jasvinder.singh@intel.com>
This version update contains:
* Fix for verification of the offload capabilities (especially for
IPv6 packets).
* Support for Tx and Rx free threshold values.
* Fixes for per-queue offload capabilities.
* Announce support of the scattered Rx offload.
* NUMA aware allocations.
* Check for the missing Tx completions.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
In some cases Tx descriptors may be uncompleted by the HW and as a
result they will never be released.
This patch adds checking for the missing Tx completions to the ENA timer
service, so in order to use this feature, the application must call the
function rte_timer_manage().
Missing Tx completion reset threshold is determined dynamically, by
taking into consideration ring size and the default value.
Tx cleanup is associated with the Tx burst function. As DPDK
applications can call Tx burst function dynamically, time when last
cleanup was called must be traced to avoid false detection of the
missing Tx completion.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shai Brandes <shaibran@amazon.com>
Only the IO rings memory was allocated with taking the socket ID into
the respect, while the other structures was allocated using the regular
rte_zmalloc() API.
Ring specific structures are now being allocated using the ring's
socket ID.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shai Brandes <shaibran@amazon.com>
ENA can't be forced to always pass single descriptor for the Rx packet.
Even if the passed buffer size is big enough to hold the data, we can't
make assumption that the HW won't use extra descriptor because of
internal optimizations. This assumption may be true, but only for some
of the FW revisions, which may differ depending on the used AWS instance
type.
As the scattered Rx support on the Rx path already exists, the driver
just needs to announce DEV_RX_OFFLOAD_SCATTER capability by turning on
the rte_eth_dev_data::scattered_rx option.
Fixes: 1173fca25a ("ena: add polling-mode driver")
Cc: stable@dpdk.org
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shai Brandes <shaibran@amazon.com>
As ENA currently doesn't support offloads which could be configured
per-queue, only per-port flags should be set.
In addition, to make the code cleaner, parsing appropriate offload
flags is encapsulated into helper functions, in a similar matter it's
done by the other PMDs.
[1] https://doc.dpdk.org/guides/prog_guide/
poll_mode_drv.html?highlight=offloads#hardware-offload
Fixes: 7369f88f88 ("net/ena: convert to new Rx offloads API")
Fixes: 56b8b9b7e5 ("net/ena: convert to new Tx offloads API")
Cc: stable@dpdk.org
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shai Brandes <shaibran@amazon.com>
The caller can pass Tx or Rx free threshold value to the configuration
structure for each ring. It determines when the Tx/Rx function should
start cleaning up/refilling the descriptors. ENA was ignoring this value
and doing it's own calculations.
Now the user can configure ENA's behavior using this parameter and if
this variable won't be set, the ENA will continue with the old behavior
and will use it's own threshold value.
The default value is not provided by the ENA in the ena_infos_get(), as
it's being determined dynamically, depending on the requested ring size.
Note that NULL check for Tx conf was removed from the function
ena_tx_queue_setup(), as at this place the configuration will be
either provided by the user or the default config will be used and it's
handled by the upper (rte_ethdev) layer.
Tx threshold shouldn't be used for the Tx cleanup budget as it can be
inadequate to the used burst. Now the PMD tries to release mbufs for the
ring until it will be depleted.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shai Brandes <shaibran@amazon.com>
ENA PMD has multiple checksum offload flags, which are more discrete
than the DPDK offload capabilities flags.
As the driver wasn't storing it's internal checksum offload capabilities
and was relying only on the DPDK capabilities, not all scenarios could
be properly covered (like when to prepare pseudo header checksum and
when not).
Moreover, the user could request offload capability, which isn't
supported by the HW and the PMD would quietly ignore the issue.
This commit reworks eth_ena_prep_pkts() function to perform additional
checks and to properly reflect the HW requirements. With the
RTE_LIBRTE_ETHDEV_DEBUG enabled, the function will do even more
verifications, to help the user find any issues with the mbuf
configuration.
Fixes: b3fc5a1ae1 ("net/ena: add Tx preparation")
Cc: stable@dpdk.org
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shai Brandes <shaibran@amazon.com>
In sfc, MAE admin serves as a transfer proxy. In order to track which
ethdev is privileged, augment every independent switch port structure
with information about its MAE privilege.
Signed-off-by: Viacheslav Galaktionov <viacheslav.galaktionov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Register unprivileged ports in the switch domain registry in order to
allow redirecting traffic to them.
Differentiate between different levels of MAE support, update all MAE
status checks.
Signed-off-by: Viacheslav Galaktionov <viacheslav.galaktionov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Removing 'DEV_RX_OFFLOAD_JUMBO_FRAME' offload flag.
Instead of drivers announce this capability, application can deduct the
capability by checking reported 'dev_info.max_mtu' or
'dev_info.max_rx_pktlen'.
And instead of application setting this flag explicitly to enable jumbo
frames, this can be deduced by driver by comparing requested 'mtu' to
'RTE_ETHER_MTU'.
Removing this additional configuration for simplification.
Suggested-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Rosen Xu <rosen.xu@intel.com>
Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Huisong Li <lihuisong@huawei.com>
Acked-by: Hyong Youb Kim <hyonkim@cisco.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>
Move requested MTU value check to the API to prevent the duplicated
code.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Rosen Xu <rosen.xu@intel.com>
Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Setting MTU bigger than RTE_ETHER_MTU requires the jumbo frame support,
and application should enable the jumbo frame offload support for it.
When jumbo frame offload is not enabled by application, but MTU bigger
than RTE_ETHER_MTU is requested there are two options, either fail or
enable jumbo frame offload implicitly.
Enabling jumbo frame offload implicitly is selected by many drivers
since setting a big MTU value already implies it, and this increases
usability.
This patch moves this logic from drivers to the library, both to reduce
the duplicated code in the drivers and to make behaviour more visible.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Rosen Xu <rosen.xu@intel.com>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Huisong Li <lihuisong@huawei.com>
There is a confusion on setting max Rx packet length, this patch aims to
clarify it.
'rte_eth_dev_configure()' API accepts max Rx packet size via
'uint32_t max_rx_pkt_len' field of the config struct 'struct
rte_eth_conf'.
Also 'rte_eth_dev_set_mtu()' API can be used to set the MTU, and result
stored into '(struct rte_eth_dev)->data->mtu'.
These two APIs are related but they work in a disconnected way, they
store the set values in different variables which makes hard to figure
out which one to use, also having two different method for a related
functionality is confusing for the users.
Other issues causing confusion is:
* maximum transmission unit (MTU) is payload of the Ethernet frame. And
'max_rx_pkt_len' is the size of the Ethernet frame. Difference is
Ethernet frame overhead, and this overhead may be different from
device to device based on what device supports, like VLAN and QinQ.
* 'max_rx_pkt_len' is only valid when application requested jumbo frame,
which adds additional confusion and some APIs and PMDs already
discards this documented behavior.
* For the jumbo frame enabled case, 'max_rx_pkt_len' is an mandatory
field, this adds configuration complexity for application.
As solution, both APIs gets MTU as parameter, and both saves the result
in same variable '(struct rte_eth_dev)->data->mtu'. For this
'max_rx_pkt_len' updated as 'mtu', and it is always valid independent
from jumbo frame.
For 'rte_eth_dev_configure()', 'dev->data->dev_conf.rxmode.mtu' is user
request and it should be used only within configure function and result
should be stored to '(struct rte_eth_dev)->data->mtu'. After that point
both application and PMD uses MTU from this variable.
When application doesn't provide an MTU during 'rte_eth_dev_configure()'
default 'RTE_ETHER_MTU' value is used.
Additional clarification done on scattered Rx configuration, in
relation to MTU and Rx buffer size.
MTU is used to configure the device for physical Rx/Tx size limitation,
Rx buffer is where to store Rx packets, many PMDs use mbuf data buffer
size as Rx buffer size.
PMDs compare MTU against Rx buffer size to decide enabling scattered Rx
or not. If scattered Rx is not supported by device, MTU bigger than Rx
buffer size should fail.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>
Acked-by: Huisong Li <lihuisong@huawei.com>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Rosen Xu <rosen.xu@intel.com>
Acked-by: Hyong Youb Kim <hyonkim@cisco.com>
Build error:
../drivers/net/enic/enic_fm_flow.c: In function 'enic_fm_flow_parse':
../drivers/net/enic/enic_fm_flow.c:1467:24:
error: 'dev' may be used uninitialized in this function
[-Werror=maybe-uninitialized]
struct rte_eth_dev *dev;
^~~
../drivers/net/enic/enic_fm_flow.c:1580:24:
error: 'dev' may be used uninitialized in this function
[-Werror=maybe-uninitialized]
struct rte_eth_dev *dev;
^~~
../drivers/net/enic/enic_fm_flow.c:1599:24:
error: 'dev' may be used uninitialized in this function
[-Werror=maybe-uninitialized]
struct rte_eth_dev *dev;
^~~
Build error looks like false positive, but to silence the compiler
initializing the pointer with NULL.
Bugzilla ID: 812
Fixes: 54bd4ebe8b ("net/enic: support meta flow actions to overrule destinations")
Reported-by: David Marchand <david.marchand@redhat.com>
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Since AARCH32 extension is not implemented on octeontx2 family, only
enable build for 64bit.
Due to Linux kernel AF(Admin Function) driver dependency, only enable
build for 64-bit Linux.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Since AARCH32 extension is not implemented on octeontx family, only
enable build for 64bit.
Due to Linux kernel AF(Admin function) driver dependency, only enable
build for 64-bit Linux.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Since AARCH32 extension is not implemented on thunderx family, only
enable build for 64bit.
Due to Linux kernel AF(Admin function) driver dependency, only enable
build for Linux.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
When release Tx queue, Rx queue data got freed because wrong Tx queue
data located.
This patch fixes the wrong Tx queue data location.
Fixes: 7483341ae5 ("ethdev: change queue release callback")
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Fate actions are different per domain.
When all the domains, ingress, egress and FDB (transfer),
can support all the policy actions, i.e. [SET_TAG],
the policy prepares resources for all the domains and
failure happens if one of the domains misses its fate action
in the policy action list.
Remove the domains missing their fate action
from the meter policy preparation.
Now, the policy will prepare a domain only when the domain supports
all the actions and when one of the domain fate actions is on the list.
Fixes: afb4aa4f12 ("net/mlx5: support meter policy operations")
Cc: stable@dpdk.org
Signed-off-by: Li Zhang <lizh@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
This patch fixes coverity issue by avoiding use of null pointer
in taking false branch.
Coverity issue: 373360
Fixes: 437dbd2fd4 ("net/ice: support 1PPS")
Signed-off-by: Simei Su <simei.su@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
In function ice_dcf_stop_queues(), RX queues and TX queues are actually
not freed, so their pointers shall not be set to NULL when queues are
stopped.
This patch adds function call to free queues on DCF device close,
which also set the RX and TX queues' pointers to NULL on freeing
queues, and avoids referring to the released resource when device is
started again.
Fixes: 1a86f4dbdf ("net/ice: support DCF device reset")
Cc: stable@dpdk.org
Signed-off-by: Dapeng Yu <dapengx.yu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
If flow redirect failed, the spinlock will not be unlocked.
This patch fixes it.
Fixes: bc9201388d ("net/ice: support flow redirect")
Cc: stable@dpdk.org
Signed-off-by: Dapeng Yu <dapengx.yu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
The Timestamp Overlay feature is available only in 32B Flex Descriptors.
This patch adds compile option when in 16B Flex Descriptors.
Fixes: 953e74e6b7 ("net/ice: enable Rx timestamp on flex descriptor")
Fixes: 646dcbe6c7 ("net/ice: support IEEE 1588 PTP")
Signed-off-by: Simei Su <simei.su@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Null-checking "p" suggests that it may be null, but it has already
been dereferenced on all paths leading to the check. Thus correct
the code lines and remove the redundant line.
Fixes: c84f8aa210 ("net/ice/base: add parser runtime skeleton")
Signed-off-by: Junfeng Guo <junfeng.guo@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Firmware 8.4+ will return I40E_AQ_RC_ENOENT when try to delete
non-existent MAC/VLAN addresses from the HW filtering, this should
not be considered as an Admin Queue error. But in i40e_asq_send_command,
it will return I40E_ERR_ADMIN_QUEUE_ERROR if the return value of Admin
Queue command processed by Firmware is not I40E_AQ_RC_OK or
I40E_AQ_RC_EBUSY.
Use i40e_aq_remove_macvlan_v2 instead so that we can get the
corresponding Admin Queue status, and not report as an error in DPDK
when Firmware return I40E_AQ_RC_ENOENT, and this also not break with an
old firmware.
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Adjust the code line order of the parser runtime reset, since the
struct rt->psr is used in function _rt_flag_set before assignment.
Fixes: c84f8aa210 ("net/ice/base: add parser runtime skeleton")
Cc: stable@dpdk.org
Signed-off-by: Junfeng Guo <junfeng.guo@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Some drivers do not provide per-queue statistics. So, there is no point
to have these misleading zeros in xstats.
Fixes: f30e69b41f ("ethdev: add device flag to bypass auto-filled queue xstats")
Cc: stable@dpdk.org
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Parse inner L2 length to set correct packet type, and ensure that
hardware can compute the checksum successfully.
Fixes: b950203be7 ("net/txgbe: support VXLAN-GPE")
Cc: stable@dpdk.org
Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
Setting exact link speed makes sense if auto-negotiation is
disabled. Fixed flag is required to disable auto-negotiation.
Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
It's necessary to set 1 on TXGBE_PX_INTA register to get interrupts
normally, when legacy interrupt mode is used.
Fixes: 2fc745e6b6 ("net/txgbe: add interrupt operation")
Cc: stable@dpdk.org
Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
Update immediate value/pointer source operand support
for modify field RTE Flow action:
- source operand data can be presented by byte buffer
(instead of former uint64_t) or by pointer
- no host byte ordering is assumed anymore for immediate
data buffer (not uint64_t anymore)
- no immediate value offset is expected (the source
subfield is located at the same offset as in destination)
Signed-off-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
Setting 'RTE_LIBRTE_BNXT_TRUFLOW_DEBUG' macro cause build error,
removing it.
Also with meson build system compile time debug macros should be
documented in driver documentation, since there is no other way to
figure out their existence.
Fixes: ad9eed0248 ("net/bnxt: support flow template for Thor")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reported by "gcc (GCC) 12.0.0 20211003 (experimental)":
./drivers/net/softnic/rte_eth_softnic_cli.c:
In function ‘tmgr_hierarchy_default’:
./drivers/net/softnic/rte_eth_softnic_cli.c:634:73:
error: the comparison will always evaluate as ‘true’ for the
address of ‘tc_valid’ will never be NULL [-Werror=address]
634 | (¶ms->shared_shaper_id.tc_valid[0]) ? 1 : 0,
| ^
Fixing it by removing useless check.
Fixes: 1af2dc5111 ("net/softnic: add command for default tmgr hierarchy")
Fixes: 5eb676d74f ("net/softnic: add config flexibility to TM")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Jasvinder Singh <jasvinder.singh@intel.com>
Add support for item PORT_REPRESENTOR which should
be used instead of ambiguous item PORT_ID.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Action PORT_ID implementation assumes ingress only. Its semantics
suggests that support for equal action PORT_REPRESENTOR be added.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Semantics of the existing support for action PORT_ID suggests
that support for equal action REPRESENTED_PORT be implemented.
Helper functions keep port_id suffix since action
MLX5_FLOW_ACTION_PORT_ID is still used internally.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Add support for actions PORT_REPRESENTOR and REPRESENTED_PORT
based on the existing support for action PORT_ID.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Hyong Youb Kim <hyonkim@cisco.com>
Add support for actions PORT_REPRESENTOR and REPRESENTED_PORT
based on the existing support for action PORT_ID.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Add support for items PORT_REPRESENTOR and REPRESENTED_PORT
based on the existing support for item PORT_ID.
The use of item PORT_ID depends on the specified direction attribute.
Items PORT_REPRESENTOR and REPRESENTED_PORT, in turn, define traffic
direction themselves. The former matches traffic from the driver's
vNIC. The latter matches packets from either a v-port (network) or
a VF's vNIC (if the driver's port is a VF representor).
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Move rte_eth_dev, rte_eth_dev_data, rte_eth_rxtx_callback and related
data into private header (ethdev_driver.h).
Few minor changes to keep DPDK building after that.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Tested-by: Feifei Wang <feifei.wang2@arm.com>
Currently majority of fast-path ethdev ops take pointers to internal
queue data structures as an input parameter.
While eth_rx_queue_count() takes a pointer to rte_eth_dev and queue
index.
For future work to hide rte_eth_devices[] and friends it would be
plausible to unify parameters list of all fast-path ethdev ops.
This patch changes eth_rx_queue_count() to accept pointer to internal
queue data as input parameter.
While this change is transparent to user, it still counts as an ABI change,
as eth_rx_queue_count_t is used by ethdev public inline function
rte_eth_rx_queue_count().
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Hyong Youb Kim <hyonkim@cisco.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Tested-by: Feifei Wang <feifei.wang2@arm.com>
By design, in a GROUP flow, outer match criteria go to "ENC" fields
of the action rule match specification. The current HW/FW hasn't
got support for these fields (except the VXLAN VNI) yet.
As a workaround, start parsing the pattern from the tunnel item.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Such a counter will only report the number of hits, which is actually
a sum of two contributions (the JUMP rule's own counter + indirect
increments issued by counters of the associated GROUP rules.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
By design, JUMP flows should be represented solely by the outer rules. But
the HW/FW hasn't got support for setting Rx mark from RECIRC_ID on outer
rule lookup yet. Neither does it support outer rule counters. As a
workaround, an action rule of lower priority is used to do the job.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
The current HW/FW doesn't allow to match on MAC addresses in outer rules.
One day this will change for sure, but right now a workaround is needed.
Match on VLAN presence in outer rules is also unsupported. Ignore it.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Support generic callbacks which callers will invoke to get
PMD-specific actions and items used to produce JUMP and
GROUP flows and to detect tunnel information.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
GROUP is an in-house term for so-called "tunnel_match" flows.
On parsing, they are detected by virtue of PMD-internal item
MARK. It associates a given flow with its tunnel context.
Such a flow is represented by a MAE action rule which is
chained with the corresponding JUMP rule's outer rule
by virtue of matching on its recirculation ID.
GROUP flows do narrower match than JUMP flows do and
decapsulate matching packets (full offload).
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
JUMP is an in-house term for so-called "tunnel_set" flows. On parsing,
they are identified by virtue of actions MARK (PMD-internal) and JUMP.
The action MARK associates a given flow with its tunnel context.
Such a flow is represented by a MAE outer rule (OR) which has its
recirculation ID set. This ID is also associated with the tunnel
context. The OR is supposed to set this ID in 8 high bits of
Rx mark in matching packets. It also counts the packets.
Packets that hit the OR but miss in action rule (AR) table,
should go to MAE admin PF (that is, to DPDK) by default.
Support for the use of action COUNT in JUMP
flows will be introduced by later patches.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Later patches add support for tunnel offload on Riverhead (EF100).
A board can host at most 254 tunnels. Partially offloaded (missed)
tunnel packets are identified by virtue of 8 high bits in Rx mark.
Add basic definitions of the upcoming tunnel offload support and
take care of the dedicated bits in Rx mark across the driver.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
vnic_dev_capable_filter_mode() currently fails when
CMD_CAPABILITY(CMD_ADD_FILTER) returns ERR_EPERM. In turn, this
failure causes the driver initialization to fail.
But, firmware may legitimately return ERR_EPERM. For example, VF vNIC
returns ERR_EPERM when it does not support filtering at all. So, treat
ERR_EPERM as "no filtering available" instead of an unexpected error.
Fixes: 322b355f21 ("net/enic/base: bring NIC interface functions up to date")
Cc: stable@dpdk.org
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
Reviewed-by: John Daley <johndale@cisco.com>
The intr_handle->intr_vec is allocated by rte_zmalloc(), but freed by
free(), this patch fixes it.
Fixes: 02a7b55657 ("net/hns3: support Rx interrupt")
Cc: stable@dpdk.org
Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Detect the flag in Rx prefix and pass it to users.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
MAE counter engine gets generation counts by virtue of the mark,
so the code to extract the field is already in place, but flow
action MARK doesn't benefit from it. Support this use case, too.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Initial support for the method. Later patches will extend it to
make FLAG and MARK delivery available on EF100 native datapath.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Enhance support for RSS action in the non-TruFlow path.
This will allow the user or application to update the RSS settings
using RTE_FLOW API.
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Fix Rx queue state on device start.
The state of Rx queues could be incorrect in some cases
because instead of updating the state for all the Rx queues,
we are updating it for queues in a VNIC.
Fixes: 0105ea1296 ("net/bnxt: support runtime queue setup")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Lance Richardson <lance.richardson@broadcom.com>
Aggregation rings are needed when PMD needs to support jumbo frames, LRO.
Currently we are creating the aggregation rings whether jumbo frames or
LRO has been enabled or disabled. This causes unnecessary allocation of
mbufs needing larger mbuf pool which is not used at all.
This patch modifies the code to create aggregation rings only when
needed.
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Support of the keep-CRC offloading by checking
the relevant FW capability (scatter_fcs) for NIC support.
Supported offload:
DEV_RX_OFFLOAD_KEEP_CRC
Signed-off-by: Tal Shnaiderman <talshn@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Tested-by: Idan Hackmon <idanhac@nvidia.com>
Support of the VLAN stripping offloading by checking
the relevant FW capability (vlan_cap) for NIC support.
Supported offload:
DEV_RX_OFFLOAD_VLAN_STRIP
Signed-off-by: Tal Shnaiderman <talshn@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Tested-by: Idan Hackmon <idanhac@nvidia.com>
Support of the TSO offloading by checking
the relevant FW capability for NIC support.
Supported offloads:
DEV_TX_OFFLOAD_TCP_TSO
DEV_TX_OFFLOAD_VXLAN_TNL_TSO
DEV_TX_OFFLOAD_GRE_TNL_TSO
DEV_TX_OFFLOAD_GENEVE_TNL_TSO
Signed-off-by: Tal Shnaiderman <talshn@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Tested-by: Idan Hackmon <idanhac@nvidia.com>
Query tunneling supported on the NIC.
Save the offloads values in a config parameter.
This is needed for the following TSO support:
DEV_TX_OFFLOAD_VXLAN_TNL_TSO
DEV_TX_OFFLOAD_GRE_TNL_TSO
DEV_TX_OFFLOAD_GENEVE_TNL_TSO
Signed-off-by: Tal Shnaiderman <talshn@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Tested-by: Idan Hackmon <idanhac@nvidia.com>
Currently, the PMD decides if the tunneling offload
can enable VXLAN/GRE/GENEVE tunneled TSO support by checking
config->tunnel_en (single bit) and config->tso.
This is incorrect, the right way is to check the following
flags returned by the mlx5dv_query_device function:
MLX5DV_RAW_PACKET_CAP_TUNNELED_OFFLOAD_VXLAN - if supported the offload
DEV_TX_OFFLOAD_VXLAN_TNL_TSO can be enabled.
MLX5DV_RAW_PACKET_CAP_TUNNELED_OFFLOAD_GRE - if supported the offload
DEV_TX_OFFLOAD_GRE_TNL_TSO can be enabled.
MLX5DV_RAW_PACKET_CAP_TUNNELED_OFFLOAD_GENEVE - if supported the offload
DEV_TX_OFFLOAD_GENEVE_TNL_TSO can be enabled.
The fix enables the offloads according to the correct
flags returned by the kernel.
Fixes: dbccb4cddc ("net/mlx5: convert to new Tx offloads API")
Cc: stable@dpdk.org
Signed-off-by: Tal Shnaiderman <talshn@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Tested-by: Idan Hackmon <idanhac@nvidia.com>
Query software parsing supported on the NIC.
Save the offloads values in a config parameter.
This is needed for the outer IPv4 checksum and
IP and UDP tunneled packet TSO support.
Signed-off-by: Tal Shnaiderman <talshn@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Tested-by: Idan Hackmon <idanhac@nvidia.com>
Currently, the PMD decides if the software parsing
offload can enable outer IPv4 checksum and tunneled
TSO support by checking config->hw_csum and config->tso
respectively.
This is incorrect, the right way is to check the following
flags returned by the mlx5dv_query_device function:
MLX5DV_SW_PARSING - check general swp support.
MLX5DV_SW_PARSING_CSUM - check swp checksum support.
MLX5DV_SW_PARSING_LSO - check swp LSO/TSO support.
The fix enables the offloads according to the correct
flags returned by the kernel.
Fixes: e46821e9fc ("net/mlx5: separate generic tunnel TSO from the standard one")
Cc: stable@dpdk.org
Signed-off-by: Tal Shnaiderman <talshn@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Tested-by: Idan Hackmon <idanhac@nvidia.com>
When the iavf_adapter instance is not initialized completely in the
primary process, the secondary process accesses its "rte_eth_dev"
member, it causes secondary process crash.
This patch replaces eth_dev with eth_dev_data in iavf_adapter.
Fixes: f978c1c9b3 ("net/iavf: add RSS hash parsing in AVX path")
Fixes: 9c9aa00403 ("net/iavf: add offload path for Rx AVX512 flex descriptor")
Fixes: 63660ea3ee ("net/iavf: add RSS hash parsing in SSE path")
Cc: stable@dpdk.org
Signed-off-by: Dapeng Yu <dapengx.yu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
This patch adds some defines related to DDP Track ID.
Signed-off-by: Artur Tyminski <arturx.tyminski@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
Update FVL FW API version to 1.15
Signed-off-by: Maciej Paczkowski <maciej.paczkowski@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
Add raw format for i40e_32byte_rx_desc, right now this only be used
by kernel driver, the commit is just to sync with kernel driver.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
Add macros and structures for MAC frequency calculation in case the link
is not present.
Remove duplicate definition in i40e_ethdev.c
Signed-off-by: Piotr Kwapulinski <piotr.kwapulinski@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
The variable checksum from i40e_calc_nvm_checksum is used before return
value is checked. Fix this logic.
Fixes: 8db9e2a1b2 ("i40e: base driver")
Fixes: 3ed6c3246f ("i40e/base: handle AQ timeout when releasing NVM")
Cc: stable@dpdk.org
Signed-off-by: Christopher Pau <christopher.pau@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
The status of i40e_read_nvm_word is not checked, so variables set
from this function could be used uninitialized. In this case, preserve
the existing flow that does not block initialization by initializing
these values from the start.
Fixes: 8d6c51fcd2 ("i40e/base: get OEM version")
Fixes: 2db7057424 ("net/i40e/base: limit PF/VF specific code to that driver only")
Cc: stable@dpdk.org
Signed-off-by: Christopher Pau <christopher.pau@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
Fix mismatched function name in comments.
Fixes: 8db9e2a1b2 ("i40e: base driver")
Fixes: 842ea19963 ("i40e/base: save link module type")
Fixes: fd72a2284a ("i40e/base: support LED blinking with new PHY")
Fixes: 788fc17b2d ("i40e/base: support proxy config for X722")
Cc: stable@dpdk.org
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
Add flags for outer VLAN and include set port parameters.
Add flags, which describe port and switch state for both double VLAN
functionality and outer VLAN processing.
Signed-off-by: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
For Active Optical Cable (AOC) the correct media type is "Fibre",
not "Direct Attach Copper".
Fixes: d749d4d899 ("i40e/base: add AOC PHY types")
Fixes: aa153cc89f ("net/i40e/base: add new PHY types for 25G AOC and ACC")
Cc: stable@dpdk.org
Signed-off-by: Dawid Lukwinski <dawid.lukwinski@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
The X722 card has 'Link Type' information elsewhere than the X710.
Previously, for all cards, the 'Link Type' information was retrieved by
opcode 0x0607 and this value was wrong for all X722 cards.
Now this information for X722 only is taken by opcode 0x0600
(function: i40e_aq_get_phy_capabilities) instead of an opcode
0x0607 (function: i40e_aq_get_link_info).
All other parameters read by opcode 0x0607 unchanged.
Fixes: e6691b428e ("i40e/base: fix PHY NVM interaction")
Fixes: 75c3de654e ("net/i40e/base: fix long link down notification time")
Cc: stable@dpdk.org
Signed-off-by: Jaroslaw Gawin <jaroslawx.gawin@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
PF has to delete all the filters during reset.
If it is fully loaded with filters then it is possible that it will take
more than 200 ms to finish the reset resulting in timeout during pf_reset
and PF reset failed, -15 error indication.
Increasing the timeout value for PF reset from 200 to 1000 to give PF
more time to finish reset if it is loaded with filters.
Fixes: 1e32378f07 ("i40e/base: increase PF reset max loop limit")
Cc: stable@dpdk.org
Signed-off-by: Sylwester Dziedziuch <sylwesterx.dziedziuch@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
Unlike other supported adapters, 2.5G and 5G use different PHY type
identifiers for reading/writing PHY settings and for reading link status.
This commit introduces separate PHY identifiers for these two operation
types.
Fixes: 988ed63c74 ("net/i40e/base: add support for Carlsville device")
Cc: stable@dpdk.org
Signed-off-by: Dawid Lukwinski <dawid.lukwinski@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
Add definitions for Shadow RAM pointers: 6th FPA (Free Provisioning Area)
module, 5th FPA module in X722 and Preservation Rules module.
These definitions are not using by DPDK now, the purpose of this commit
is to sync base code with kernel driver.
Signed-off-by: Stanislaw Grzeszczak <stanislaw.a.grzeszczak@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
Some customers want to downgrade to an earlier FW security revision, this
already implemented by FW so that customers can have more control over
the security revisions they can use. FW also implemented a mechanism via
NVMupdate to allow the users to accept or not a baseline Min SRev version
that will limit the secure version rollback only down to that level.
This commit increments X722 API version and adds new minimal rollback
revision that related to the extended implementation of Security Revision
Opt-In for 4 more X722 modules.
These definitions are not using by DPDK now, the purpose of this commit
is sync with latest share code.
Signed-off-by: Stanislaw Grzeszczak <stanislaw.a.grzeszczak@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
ASQ(Admin Send Queue) send command functions are returning only i40e
status codes yet some calling functions also need Admin Queue status that
is stored in hw->aq.asq_last_status. Since hw object is stored on a heap
it introduces a possibility for a race condition in access to hw if
calling function is not fast enough to read hw->aq.asq_last_status before
next send ASQ command is executed.
Added new versions of send ASQ command functions that return Admin Queue
status on the stack to avoid race conditions in access to
hw->aq.asq_last_status.
Added new _v2 version of i40e_aq_remove_macvlan and i40e_aq_add_macvlan
that is using new _v2 versions of ASQ send command functions and returns
the Admin Queue status on the stack.
Signed-off-by: Sylwester Dziedziuch <sylwesterx.dziedziuch@intel.com>
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
On the vector implementation, during the tear-down, the mbufs not
drained in the RxQ and TxQ are freed based on an algorithm which
supposed that the number of descriptors is a power of 2 (max_desc).
Based on this hypothesis, this algorithm uses a bitmask in order to
detect an index overflow during the iteration, and to restart the loop
from 0.
However, there is no such power of 2 requirement in the ixgbe for the
number of descriptors in the RxQ / TxQ. The only requirement is to have
a number correctly aligned.
If a user requested to configure a number of descriptors which is not a
power of 2, as a consequence, during the tear-down, it was possible to
be in an infinite loop, and to never reach the exit loop condition.
By removing the bitmask and changing the loop method, we can avoid this
issue, and allow the user to configure a RxQ / TxQ which is not a power
of 2.
Fixes: c95584dc2b ("ixgbe: new vectorized functions for Rx/Tx")
Cc: stable@dpdk.org
Signed-off-by: Julien Meunier <julien.meunier@nokia.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
Some packets are discarded by the NIC because they are larger than
the MTU, these packets should be counted as "RX error" instead of
"RX packet", for example:
pkt1 = Ether()/IP()/Raw('x' * 1400)
pkt2 = Ether()/IP()/Raw('x' * 1500)
---------------- Forward statistics for port 0 -----------------
RX-packets: 2 RX-dropped: 0 RX-total: 2
TX-packets: 1 TX-dropped: 0 TX-total: 1
----------------------------------------------------------------
Here the packet pkt2 has been discarded, but still was counted
by "RX-packets"
The register 'GL_RXERR1' can count above discarded packets.
This patch adds reading and calculation of the 'GL_RXERR1' counter
when reporting DPDK statistics.
Fixes: f4a91c38b4 ("i40e: add extended stats")
Cc: stable@dpdk.org
Signed-off-by: Alvin Zhang <alvinx.zhang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
If GTPU Extension header has no pdu_type setting, the parsed value of
gtp_psc_spec->hdr.type will be 0, which is same as IAVF_GTPU_EH_DWLINK.
Thus, for this case, we should check gtp_psc_mask->hdr.type instead,
to set QFI field bit of GTPU_EH first.
Fixes: cd212c4669 ("net/iavf: fix QFI fields of GTPU UL/DL for flow director")
Cc: stable@dpdk.org
Signed-off-by: Junfeng Guo <junfeng.guo@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Indirect actions should be used to do shared counters.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>
Acked-by: Ori Kam <orika@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
The representor support has been implemented to some extent, and the fact
that ethdev mport is equivalent to entity mport is by design.
Fixes: 1fb65e4dae ("net/sfc: support flow action port ID in transfer rules")
Cc: stable@dpdk.org
Signed-off-by: Viacheslav Galaktionov <viacheslav.galaktionov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Let the driver provide the user with information about available
representors by implementing the representor_info_get operation.
Due to the lack of any structure to representor IDs, every ID range
describes exactly one representor.
Signed-off-by: Viacheslav Galaktionov <viacheslav.galaktionov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Representor IDs must be unique for each representor. VFs, which are
currently used, are not unique as they may repeat in combination with
different PCI controllers and PFs. On the other hand, switch port IDs
are unique, so they are a better fit for this role.
Signed-off-by: Viacheslav Galaktionov <viacheslav.galaktionov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Allow the user to specify representor entities using the structured
parameter values.
Signed-off-by: Viacheslav Galaktionov <viacheslav.galaktionov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Make representor names unique on multi-host configurations.
Signed-off-by: Viacheslav Galaktionov <viacheslav.galaktionov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
This information will be useful when representor info API is implemented.
Signed-off-by: Viacheslav Galaktionov <viacheslav.galaktionov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Newer hardware may have arbitrarily complex controller configurations,
and for this reason the mapping has been made dynamic: it is represented
with a dynamic array that is indexed by controller numbers and each
element contains an EFX interface number. Since the number of controllers
is expected to be small, this approach should not hurt the performance.
Signed-off-by: Viacheslav Galaktionov <viacheslav.galaktionov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
If for some reason the hardware switch ID initialization function fails,
MAE lock is still held after the function finishes. This patch fixes that.
Fixes: 1e7fbdf0ba ("net/sfc: support concept of switch domains/ports")
Cc: stable@dpdk.org
Signed-off-by: Viacheslav Galaktionov <viacheslav.galaktionov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Gather statistics of enqueued and dequeued packets in Rx and Tx burst
callbacks to report in stats_get callback.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Implement traffic forwarding for representor and representor proxy
from virtual functions to representor Rx queues.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Set ingress mport dynamic field in mbuf in EF100.
For a given PF, Rx queues of representor devices
function on top of the only Rx queue operated by
the PF representor proxy facility. This field is
a means to demultiplex traffic hitting the queue.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Layout of the EF100 Rx prefix is defined in terms of a 32 bytes long
value type (xword). Replace oword with xword to avoid truncation.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Forward traffic that is transmitted from a port representor to the
corresponding virtual function using the dedicated TxQ.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Support probing the device multiple times so that additional port
representors can be created with hotplug EAL API. To hotplug a
representor, the PF must be hotplugged with different representor
device argument.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Implement the callback by reporting link down if the representor
is not started, otherwise report link up with undefined link speed.
Link speed is undefined since representors can pass traffic to each
other even if the PF link is down.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Implement queue start and stop operation both in port
representors and representor proxy.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Implement Tx queue start and stop in port representor proxy.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Add extra libefx flags to Rx queue information initialization
function interface to be able to specify the ingress m-port
flag for a representor RxQ. Rx prefix of packets on that queue
will contain ingress m-port field required for packet forwarding
in representor proxy.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Implement queue creation and destruction both in port representors
and representor proxy.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
The API is required to create and destroy representor proxy
port assigned to representor.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Provide minimal implementation for port representors that only can be
configured and can provide device information.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Adapter state will be reused by representors, that will have
a separate adapter. Rename adapter state to ethdev state
so that the meaning of it is clearer.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
A Tx/Rx queue pair is required to forward traffic between
port representors and virtual functions.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Representor proxy is a mediator between virtual functions and port
representors. It forwards traffic between virtual functions and port
representors performing base PF ethdev + VF's representor traffic
(de-)multiplexing. The implementation will be provided by later patches.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Add internal mbuf dynamic flag and field to request EF100 native
Tx datapath to use Tx prefix descriptor to override egress m-port.
Overriding egress m-port is necessary on representor Tx burst
so that the packet will reach corresponding VF.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
By default, the firmware is in EVB mode, but insertion of the first MAE
rule resets it to switchdev mode automatically and removes all automatic
MAE rules added by EVB support. On initialisation, insert MAE rules that
forward traffic between PHY and PF.
Add an API for creation and insertion of driver-internal MAE
rules(flows).
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Add the argument that allows user to choose either switchdev or legacy
mode. Legacy mode enables switching by using Ethernet virtual bridging
(EVB) API. In switchdev mode, VF traffic goes via port representor
(if any) on PF, and software virtual switch (for example, Open vSwitch)
steers the traffic.
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Andy Moreton <amoreton@xilinx.com>
Reviewed-by: Ivan Malov <ivan.malov@oktetlabs.ru>
The patch is required for all PMDs which do not provide representors
info on the representor itself.
The function, rte_eth_representor_id_get(), is used in
eth_representor_cmp() which is required in ethdev class iterator to
search ethdev port ID by name (representor case). Before the patch
the function is called on the representor itself and tries to get
representors info to match.
Search of port ID by name is used after hotplug to find out port ID
of the just plugged device.
Getting a list of representors from a representor does not make sense.
Instead, a backer device should be used.
To this end, extend the rte_eth_dev_data structure to include the port ID
of the backing device for representors.
Signed-off-by: Viacheslav Galaktionov <viacheslav.galaktionov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
Reviewed-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
If Tx datapath collects per queue statistics, use these stats
to provide opackets and obytes in basic ethdev stats.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
If Rx datapath collects per queue statistics, use these stats
to provide ipackets and ibytes in basic ethdev stats.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Add support for grouping SW stats together. When stats are
grouped the corresponding stats values for each queue
are obtained during calling one read callback. This is useful
to group per-queue stats 'packets' and 'bytes' to keep stats
consistent, i.e. a number of bytes corresponds to a number of
packets. These stats will be added in the following patches.
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Add toggle to disable total SW stat. This is useful
for per-queue 'packets' and 'bytes' to not conflict
with corresponding basic stats. These stats will be
added in the following patches.
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Global description structure of SW stats is used currently.
Following patches introduce SW stats that may be unavailable
for some adapters, so add per-adapter descriptions to safely
work with multiple adapters.
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Optimize getting number of SW stats by caching the
value during device configure since it's the only
place it may change.
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Description is a global variable shared by all adapters and must
not be modified.
Fixes: fdd7719eb3 ("net/sfc: add xstats for Rx/Tx doorbells")
Cc: stable@dpdk.org
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Fix cleanup order of SW stats structures to be reversed
to initialization order.
Fixes: fdd7719eb3 ("net/sfc: add xstats for Rx/Tx doorbells")
Cc: stable@dpdk.org
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Delete 'x' in:
1. per-adapter 'struct sfc_sw_xstats' with corresponding
field 'sw_xstats'.
2. 'struct sfc_sw_xstat_descr' and callback prototype
'sfc_get_sw_xstat_val_t' with its implementations.
The 'x' stands for 'extended' in RTE but from sfc point of
view these are just SW stats.
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
This is a better word used in RTE docs.
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Rename global array of SW stats descriptions declared as
'struct sfc_sw_xstat_descr sfc_sw_xstats[]' to
'sfc_sw_stats_descr[]' to avoid ambiguity since there is
structure declared as 'struct sfc_sw_xstats'.
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reported by "gcc (GCC) 12.0.0 20211003 (experimental)":
./drivers/net/ena/ena_rss.c: In function ‘ena_rss_reta_query’:
./drivers/net/ena/ena_rss.c:140:66:
error: the comparison will always evaluate as ‘false’ for the
pointer operand in ‘reta_conf + 136’ must not be NULL
[-Werror=address]
140 | (reta_size > RTE_RETA_GROUP_SIZE && ((reta_conf + 1) == NULL)))
| ^~
Fixing it by removing useless check.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>
For different capabilities, we declare different macro functions to
determine whether the capabilities are supported.
This patch declare a unified macro function to judge capabilities.
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
rte_eth_rx_descriptor_status() should be used as a replacement.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Currently the hash_key_size information has not been set. So, apps can
not get the key size from dev_info(), this make some problem.
e.g, in testpmd, the hash_key_size will be checked before configure
or get the hash key:
testpmd> show port 4 rss-hash
dev_info did not provide a valid hash key size
testpmd> show port 4 rss-hash key
dev_info did not provide a valid hash key size
testpmd> port config 4 rss-hash-key ipv4 (hash key)
dev_info did not provide a valid hash key size
In this patch, the meaning of rss_key_len has been modified. It only
indicated the length of the configured hash key before. Therefore,
its value depends on the user's configuration. This seems unreasonable.
And now, it indicates the minimum hash key length required by the
bonded device. Its value will be the shortest hash key among all slave
drivers.
Fixes: 734ce47f71 ("bonding: support RSS dynamic configuration")
Cc: stable@dpdk.org
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
If the vector burst mode is selected, the dedicated queue mode will not
take effect on some PMDs because these PMDs may have some limitations
in vector burst mode. For example, the limit on burst size. Currently,
both hns3 and intel I40E require four alignments when receiving packets
in vector mode. As a result, they can't accept packets if burst size
below four. However, in dedicated queue mode, the burst size of periodic
packets processing is one.
This patch fixes the above problem by modifying the burst size to 32.
This approach also makes the packet processing of the dedicated queue
mode more reasonable. Currently, if multiple LACP protocol packets are
received in the hardware queue in a cycle, only one LACP packet will be
processed in this cycle, and the left packets will be processed in the
following cycle. After the modification, all the LACP packets will be
processed at one time, which seems more reasonable and closer to the
behavior of the bonding driver when the dedicated queue is not turned on.
Fixes: 112891cd27 ("net/bonding: add dedicated HW queues for LACP control")
Cc: stable@dpdk.org
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
Since the AF_XDP PMD does not work for secondary processes as reported
in Bugzilla 805, check for the process type at the beginning of probe
and return ENOTSUP if the process type is secondary.
It is planned that secondary processes will be supported by the PMD in
full in a future release by using rte_mp_msg to pass the state to the
secondary process that it requires in order to work.
Bugzilla ID: 805
Fixes: f1debd77ef ("net/af_xdp: introduce AF_XDP PMD")
Cc: stable@dpdk.org
Reported-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Add RTE_ prefix to internal API defined in public header.
Use the prefix instead of double underscore.
Use uppercase for macros in the case of name conflict.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Fix the mempool flags namespace by adding an RTE_ prefix to the name.
The old flags remain usable, to be deprecated in the future.
Flag MEMPOOL_F_NON_IO added in the release is just renamed to have RTE_
prefix.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
When the first port in a given protection domain (PD) starts,
install a mempool event callback for this PD and register all existing
memory regions (MR) for it. When the last port in a PD closes,
remove the callback and unregister all mempools for this PD.
This behavior can be switched off with a new devarg: mr_mempool_reg_en.
On TX slow path, i.e. when an MR key for the address of the buffer
to send is not in the local cache, first try to retrieve it from
the database of registered mempools. Supported are direct and indirect
mbufs, as well as externally-attached ones from MLX5 MPRQ feature.
Lookup in the database of non-mempool memory is used as the last resort.
RX mempools are registered regardless of the devarg value.
On RX data path only the local cache and the mempool database is used.
If implicit mempool registration is disabled, these mempools
are unregistered at port stop, releasing the MRs.
Signed-off-by: Dmitry Kozlyuk <dkozlyuk@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
No barriers are required when stats are incremented or read.
Fixes: 96fd2bd69b ("net/sfc: support flow action count in transfer rules")
Cc: stable@dpdk.org
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
When adding multicast and unicast MAC addresses, three descriptors and
one descriptor are required for querying or adding MAC VLAN table,
respectively. This patch uses the number of descriptors as input
parameter to complete this task to make the function more secure.
Fixes: 7d7f9f80bb ("net/hns3: support MAC address related operations")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
This problem occurs in the following scenarios:
1) reset is encountered when the adapter is running.
2) set a new default MAC address
After the above two steps, the old default MAC address should be not
take effect. But the current behavior is contrary to that. This is due
to the change of the "default_addr_setted" in hw->mac from 'true' to
'false' after the reset. As a result, the old MAC address is not removed
when the new default MAC address is set. This variable controls whether
to delete the old default MAC address when setting the default MAC
address. It is only used when the mac_addr_set API is called for the
first time. In fact, when a unicast MAC address is deleted, if the
address isn't in the MAC address table, the driver doesn't return
failure. So this patch remove the redundant and troublesome variables to
resolve this problem.
Fixes: 7d7f9f80bb ("net/hns3: support MAC address related operations")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
Normally when closing the device the queue memzone should be
freed. But the memzone will be not freed, when device setup
ops like:
rte_eth_bond_slave_remove
-->__eth_bond_slave_remove_lock_free
---->slave_remove
------>rte_eth_dev_internal_reset
-------->rte_eth_dev_rx_queue_config
---------->eth_dev_rx_queue_config
------------>ixgbe_dev_rx_queue_release
rte_eth_dev_close
-->ixgbe_dev_close
---->ixgbe_dev_free_queues
------>ixgbe_dev_rx_queue_release
(not been called due to nb_rx_queues and nb_tx_queues are 0)
And when queue number is changed to small size, the BIG memzone
queue index will be lost. This will lead to a memory leak. So we
should release the memzone when releasing queues.
Fixes: 460d167958 ("drivers/net: delete HW rings while freeing queues")
Cc: stable@dpdk.org
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
Normally when closing the device the queue memzone should be
freed. But the memzone will be not freed, when device setup
ops like:
rte_eth_bond_slave_remove
-->__eth_bond_slave_remove_lock_free
---->slave_remove
------>rte_eth_dev_internal_reset
-------->rte_eth_dev_rx_queue_config
---------->eth_dev_rx_queue_config
------------>i40e_dev_rx_queue_release
rte_eth_dev_close
-->i40e_dev_close
---->i40e_dev_free_queues
------>i40e_dev_rx_queue_release
(not been called due to nb_rx_queues and nb_tx_queues are 0)
And when queue number is changed to small size, the BIG memzone
queue index will be lost. This will lead to a memory leak. So we
should release the memzone when releasing queues.
Fixes: 460d167958 ("drivers/net: delete HW rings while freeing queues")
Cc: stable@dpdk.org
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
Normally when closing the device the queue memzone should be
freed. But the memzone will be not freed, when device setup
ops like:
rte_eth_bond_slave_remove
-->__eth_bond_slave_remove_lock_free
---->slave_remove
------>rte_eth_dev_internal_reset
-------->rte_eth_dev_rx_queue_config
---------->eth_dev_rx_queue_config
------------>ice_rx_queue_release
rte_eth_dev_close
-->ice_dev_close
---->ice_free_queues
------>ice_rx_queue_release
(not been called due to nb_rx_queues and nb_tx_queues are 0)
And when queue number is changed to small size, the BIG memzone
queue index will be lost. This will lead to a memory leak. So we
should release the memzone when releasing queues.
Fixes: 460d167958 ("drivers/net: delete HW rings while freeing queues")
Cc: stable@dpdk.org
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
Normally when closing the device the queue memzone should be
freed. But the memzone will be not freed, when device setup
ops like:
rte_eth_bond_slave_remove
-->__eth_bond_slave_remove_lock_free
---->slave_remove
------>rte_eth_dev_internal_reset
-------->rte_eth_dev_rx_queue_config
---------->eth_dev_rx_queue_config
------------>em_rx_queue_release
rte_eth_dev_close
-->eth_em_close
---->em_dev_free_queues
------>em_rx_queue_release
(not been called due to nb_rx_queues and nb_tx_queues are 0)
And when queue number is changed to small size, the BIG memzone
queue index will be lost. This will lead to a memory leak. So we
should release the memzone when releasing queues.
Fixes: 460d167958 ("drivers/net: delete HW rings while freeing queues")
Cc: stable@dpdk.org
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
A more fine-grain flow API action RTE_FLOW_ACTION_TYPE_SAMPLE should
be used instead of it.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
The link state change interrupt handler of the NFP PMD will delay the
actual LSC work for a short period to ensure the link is stable. If the
link of the port changes state and the port is closed immediately after
the link event then a segmentation fault will occur. This happens
because the delayed LSC work eventually triggers and this logic will try
to access private port data that had been released when the port was
closed.
Fixes: 6c53f87b34 ("nfp: add link status interrupt")
Cc: stable@dpdk.org
Signed-off-by: Heinrich Kuhn <heinrich.kuhn@corigine.com>
Signed-off-by: Simon Horman <simon.horman@corigine.com>
Currently, most ethdev callback API use queue ID as parameter, but Rx
and Tx queue release callback use queue object which is used by Rx and
Tx burst data plane callback.
To align with other eth device queue configuration callbacks:
- queue release callbacks are changed to use queue ID
- all drivers are adapted
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Some drivers don't need Rx and Tx queue release callback, make them
optional. Clean up empty queue release callbacks for some drivers.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Reflect globally enabled Rx and Tx offloads in queue conf.
Also fix issue with lmt data prepare for multi seg.
Fixes: a24af6361e ("net/cnxk: add Tx queue setup and release")
Fixes: a86144cd9d ("net/cnxk: add Rx queue setup and release")
Fixes: 305ca2c4c3 ("net/cnxk: support multi-segment vector Tx")
Cc: stable@dpdk.org
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
This patch adds support to configure channel mask which will
be used by rte flow when adding flow rules with inline IPsec
action.
Signed-off-by: Satheesh Paul <psatheesh@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Adds capabitlities for AES_CBC and HMAC_SHA1 for 9k
security offload.
Signed-off-by: Srujana Challa <schalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Sets IP6_UDP_OPT in NIX RX config to allow optional
UDP checksum for IPv6 in case of security offload.
Also disable drop_re when inline inbound is enabled.
Signed-off-by: Srujana Challa <schalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Adds support to update ethertype for mixed IPsec tunnel
versions. And also sets et_overwr for inbound IPsec.
Signed-off-by: Srujana Challa <schalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Adds anti replay support for cn9k platform using
SW anti replay check.
Signed-off-by: Srujana Challa <schalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Add support to create and submit CPT instructions on Tx
on CN10K.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Add support to receive CPT processed packets on Rx via
second pass on CN10K.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Add support to create and submit CPT instructions on Tx
on CN9K SoC.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Add support to receive CPT processed packets on Rx for
CN9K.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Add support for inline inbound and outbound IPSec for SA create,
destroy and other NIX / CPT LF configurations.
This patch also changes dpdk-devbind.py to list new inline
device as misc device.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Add support for inline inbound and outbound IPSec for SA create,
destroy and other NIX / CPT LF configurations.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
reported by "gcc (GCC) 12.0.0 20211003 (experimental)":
../drivers/net/cxgbe/cxgbe_ethdev.c:
In function ‘cxgbe_dev_rx_queue_setup’:
../drivers/net/cxgbe/cxgbe_ethdev.c:682:24:
error: the comparison will always evaluate as ‘true’ for the
address of ‘fl’ will never be NULL [-Werror=address]
682 | if ((&rxq->fl) != NULL)
| ^~
Fixing it by removing useless check.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Adjust parameters order to eth_xstats_get_by_id_t prototype.
Make ids the second parameter similar to eth_xstats_get_by_id_t.
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
The af_packet pmd driver binds to a raw socket and allows sending and
receiving of packets through the kernel.
Since commit [1], the kernel strips the vlan tags early in
__netif_receive_skb_core(), so we receive untagged packets while running
with the af_packet pmd.
Luckily for us, the skb vlan-related fields are still populated from the
stripped vlan tags, so we end up having all the information that we need
in the mbuf.
Having the pmd driver support DEV_RX_OFFLOAD_VLAN_STRIP allows the
application to control the desired vlan stripping behavior, until we
have a way to describe offloads that can't be disabled by pmd drivers.
This patch will cause a change in the default way that the af_packet pmd
treats received vlan-tagged frames. While previously, the application
was required to check the PKT_RX_VLAN_STRIPPED flag, after this patch,
the pmd will re-insert the vlan tag transparently to the user, unless
the DEV_RX_OFFLOAD_VLAN_STRIP is enabled in rxmode.offloads.
I've attempted a preliminary benchmark to understand if the change could
cause a sizable performance hit.
Setup:
Two virtual machines running on top of an ESXi hypervisor
Tx: DPDK app (running on top of vmxnet3 PMD)
Rx: af_packet (running on top of a kernel vmxnet3 interface)
Packet size :68 (packet contains a vlan tag)
Rates:
Tx - 1.419 Mpps
Rx (without vlan insertion) - 1227636 pps
Rx (with vlan insertion) - 1220081 pps
At a first glance, we don't seem to have a large degradation in terms of
packet rate.
[1]
https://github.com/torvalds/linux/commit/bcc6d47903612c3861201cc3a866fb60
Signed-off-by: Tudor Cornea <tudor.cornea@gmail.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Add support to fetch port and queue stats via xstats API. Also remove
queue stats from basic stats because they're now available via xstats
API for the VF.
Signed-off-by: Nikhil Vasoya <nikhil.vasoya@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Previously, memif socket hash is always allocated on NUMA socket 0.
If the application is entirely running on another NUMA socket and EAL
--socket-limit prevents memory allocation on NUMA socket 0, memif
creation fails with "HASH: memory allocation failed" error.
This patch allows allocating memif socket hash on any NUMA socket.
Signed-off-by: Junxiao Shi <git@mail1.yoursunny.com>
Reviewed-by: Jakub Grajciar <jgrajcia@cisco.com>
Definition of `rte_ether_addr` structure used a workaround allowing DPDK
and Windows SDK headers to be used in the same file, because Windows SDK
defines `s_addr` as a macro. Rename `s_addr` to `src_addr` and `d_addr`
to `dst_addr` to avoid the conflict and remove the workaround.
Deprecation notice:
https://mails.dpdk.org/archives/dev/2021-July/215270.html
Signed-off-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
This patch enables building the ixgbe driver for Windows.
It also enables its dependencies on security and cryptodev.
I tested on AWS using ixgbe VF device, using dpdk-testpmd.
Signed-off-by: William Tu <u9012063@gmail.com>
Acked-by: Pallavi Kadam <pallavi.kadam@intel.com>
Acked-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
This patch adds comment to explain how dpaa_port_fmc_ccnode_parse
function is working to get the HW queue from FMC policy file
Signed-off-by: Rohit Raj <rohit.raj@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
This patch updates the RSS support to support following additional
distributions:
- VLAN
- ESP
- AH
- PPPOE
Signed-off-by: Vanshika Shukla <vanshika.shukla@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
TX redirection support by flow action RTE_FLOW_ACTION_TYPE_PHY_PORT
and RTE_FLOW_ACTION_TYPE_PORT_ID
This action is executed by HW to forward packets between ports.
If the ingress packets match the rule, the packets are switched
without software involved and perf is improved as well.
Signed-off-by: Jun Yang <jun.yang@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Updating MC firmware support APIs to be latest. It supports
improved DPDMUX (SRIOV equivalent) for traffic split between
dpnis and additional PTP APIs.
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Fixed with ./devtools/update-abi.sh $(cat ABI_VERSION)
Fixes: e73a7ab224 ("net/softnic: promote manage API")
Fixes: 8f532a34c4 ("fib: promote API to stable")
Fixes: 4aeb92396b ("rib: promote API to stable")
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Ray Kinsella <mdr@ashroe.eu>
Code changes done for build issue as reported in Bug 817
error: dereferencing type-punned pointer will break strict-aliasing rules.
added union to avoid pointer dereferencing
The build issue has been reported with both gcc 4.8.5 (RHEL 7) and
gcc 5.4.0 (Ubuntu 16.04).
Bugzilla ID: 817
Fixes: 39925373a3 ("net/ice/base: add parser execution main loop")
Signed-off-by: Aman Deep Singh <aman.deep.singh@intel.com>
Tested-by: David Marchand <david.marchand@redhat.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
VFs are not allowed to change physical link params when a port
module change is detected. The firmware already returns appropriate
permission error when VF tries to change physical link params. But,
make sure to avoid sending the command to firmware from VF in the
first place and prevent flooding firmware debug logs with permission
errors.
Fixes: a83041b1e9 ("net/cxgbe: rework and simplify link handling")
Cc: stable@dpdk.org
Signed-off-by: Nikhil Vasoya <nikhil.vasoya@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
The metadata can be set in the mbuf dynamic field and then used in
flow rules steering for egress direction. The hardware requires
network order for both the insertion of a rule and sending a packet.
Indeed, there is no strict restriction for the endianness. The order
for sending a packet and its steering rule should be consistent.
In the past, there was no endianness conversion due to the
performance reason. The flow rule converted the metadata into little
endian for hardware (if needed) and the packet hit the flow rule also
with little endian.
After the metadata was converted to big endian, the missing adaption
in the data path resulted in a flow miss of the egress packets.
Converting the metadata to big endian before posting a WQE to the
hardware solves this issue.
Fixes: b57e414b48 ("net/mlx5: convert meta register to big-endian")
Cc: stable@dpdk.org
Signed-off-by: Bing Zhao <bingz@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
In the function mlx5_alloc_shared_dr(), there are various reasons
to result in a failure and error clean up process. While in the
caller of mlx5_dev_spawn(), once there is a error occurring after
the mlx5_alloc_shared_dr(), the mlx5_os_free_shared_dr() is called
to release all the resources.
To prevent a double release, the pointers of the resources should
be checked before the releasing and set to NULL after done.
In the mlx5_free_table_hash_list(), after the releasing, the pointer
was missed to set to NULL and a double release may cause a crash.
By setting the tables pointer to NULL as done for other resources,
the double release and crash could be solved.
Fixes: 54534725d2 ("net/mlx5: fix flow table hash list conversion")
Cc: stable@dpdk.org
Signed-off-by: Bing Zhao <bingz@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
This patch support new global device syntax like:
bus=pci,addr=BB:DD.F/class=eth/driver=mlx5,devargs,..
In driver parameters check, ignores "driver" key which is part of new
global device syntax instead of reporting error.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
This patch removes the simplification in Virtio descriptors
handling, where their buffer addresses are IOVAs for Virtio
PCI devices, and VA-only for Virtio-user devices, which
added a requirement on Virtio-user that it only supported
IOVA as VA.
This change introduced a regression for applications using
Virtio-user and other physical PMDs that require IOVA as PA
because they don't use an IOMMU.
This patch reverts to the old behaviour, but needed to be
reworked because of the refactoring that happened in v21.02.
Fixes: 17043a2909 ("net/virtio: force IOVA as VA mode for virtio-user")
Cc: stable@dpdk.org
Reported-by: Olivier Matz <olivier.matz@6wind.com>
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Tested-by: Olivier Matz <olivier.matz@6wind.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
When attaching to an existing mono queue tap, the virtio-user was not
reporting that the virtio device was not properly initialised which
prevented from starting the port later.
$ ip tuntap add test mode tap
$ dpdk-testpmd --vdev \
net_virtio_user0,iface=test,path=/dev/vhost-net,queues=2 -- -i
...
virtio_user_dev_init_mac(): (/dev/vhost-net) No valid MAC in devargs or
device, use random
vhost_kernel_open_tap(): TUNSETIFF failed: Invalid argument
vhost_kernel_enable_queue_pair(): fail to open tap for vhost kernel
virtio_user_start_device(): (/dev/vhost-net) Failed to start device
...
Configuring Port 0 (socket 0)
vhost_kernel_open_tap(): TUNSETIFF failed: Invalid argument
vhost_kernel_enable_queue_pair(): fail to open tap for vhost kernel
virtio_set_multiple_queues(): Multiqueue configured but send command
failed, this is too late now...
Fail to start port 0: Invalid argument
Please stop the ports first
Done
The virtio-user with vhost-kernel backend was going through a lot
of complications to initialise tap fds only when using them.
For each qp enabled for the first time, a tapfd was created via
TUNSETIFF with unneeded additional steps (see below) and then mapped to
the right qp in the vhost-net backend.
Unneeded steps (as long as it has been done once for the port):
- tap features were queried while this is a constant on a running
system,
- the device name in DPDK was updated,
- the mac address of the tap was set,
On subsequent qps state change, the vhost-net backend fd mapping was
updated and the associated queue/tapfd were disabled/enabled via
TUNSETQUEUE.
Now, this patch simplifies the whole logic by keeping all tapfds opened
and in enabled state (from the tap point of view) at all time.
Unused ioctl defines are removed.
Tap features are validated earlier to fail initialisation asap.
Tap name discovery and mac address configuration are moved when
configuring qp 0.
To support attaching to mono queue tap, the virtio-user driver now tries
to attach in multi queue first, then fallbacks to mono queue.
Finally (but this is more for consistency), VIRTIO_NET_F_MQ feature is
exposed only if the underlying tap supports multi queue.
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Fix the tunnel port counting logic.
Currently we are incrementing the port count without checking
the if bnxt_hwrm_tunnel_dst_port_alloc would return success or failure.
Modify the logic to increment it only if the firmware returns success.
Fixes: 10d074b202 ("net/bnxt: support tunneling")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Lance Richardson <lance.richardson@broadcom.com>
The error recovery async event messages are often mistaken
for errors. Improved the wording to clarify the meaning of
these events.
Also, take the first step towards more inclusive language.
The references to master will be changed to primary.
For example: "bnxt_is_master_func" will be renamed to
"bnxt_is_primary_func()".
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>
The device cleanup logic was freeing most of the ring related memory,
but was not freeing up the memzone associated with the rings.
This patch fixes the issue.
Fixes: 2eb53b134a ("net/bnxt: add initial Rx code")
Fixes: 6eb3cc2294 ("net/bnxt: add initial Tx code")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Default queue state of Tx queues on startup is not correct.
Fix this by setting the state when the port is started.
Fixes: 6eb3cc2294 ("net/bnxt: add initial Tx code")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Lance Richardson <lance.richardson@broadcom.com>
1. Fix to use correct fields in the request structure of
HWRM_FUNC_DRV_RGTR.
2. Remove the "flags" argument to bnxt_hwrm_func_driver_unregister()
as it is not needed.
Fixes: beb3087f50 ("net/bnxt: add driver register/unregister")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Add ice support for new ethdev APIs to enable/disable and read/write/adjust
IEEE1588 PTP timestamps. Currently, only scalar path supports 1588 PTP,
vector path doesn't.
The example command for running ptpclient is as below:
./build/examples/dpdk-ptpclient -c 1 -n 3 -- -T 0 -p 0x1
Signed-off-by: Simei Su <simei.su@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
The request of getting VF VSI map request may fail when DCF is busy,
this patch adds retry mechanism to make it able to succeed.
Fixes: b09d34ac85 ("net/ice: fix flow redirector")
Cc: stable@dpdk.org
Signed-off-by: Dapeng Yu <dapengx.yu@intel.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
In the iavf_config_rx_queues_irqs function, the memory pointed to by the
intr_handle->intr_vec and qv_map addresses is not released in the
subsequent hook branch, resulting in resource leakage.
Fixes: f593944fc9 ("net/iavf: enable IRQ mapping configuration for large VF")
Cc: stable@dpdk.org
Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Use the dynamic mbuf to register timestamp field and flag.
The ice has the feature to dump Rx timestamp value into dynamic
mbuf field by flex descriptor. This feature is turned on by dev
config "enable-rx-timestamp". Currently, it's only supported
under scalar path.
Signed-off-by: Simei Su <simei.su@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
The default VF driver for Intel 700 Series Ethernet Controller already
switch to iavf in DPDK 21.05. And i40evf is no need to maintain now,
so remove i40evf related code.
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Due to i40evf will be removed, so there's no need to keep the devargs
option "driver=i40evf" in iavf.
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
If call ice_flow_rem_entry() directly without checking entry_id, may
cause an ACL flow entry to be freed more than once.
This patch tries to find entry_id first, then call ice_flow_rem_entry()
to avoid the defect.
Fixes: 40d466fa9f ("net/ice: support ACL filter in DCF")
Cc: stable@dpdk.org
Signed-off-by: Dapeng Yu <dapengx.yu@intel.com>
Reviewed-by: Simei Su <simei.su@intel.com>
There is currently a scenario test, which will continuously obtain port
statistics, causing the CPU usage to soar, which does not meet the
demand. After positioning analysis, it is found that the VF and PF
command interaction is completed through the iavf_execute_vf_cmd
function.
After the message is sent, it needs to wait for the interrupt thread to
obtain the response from the PF. For the data, the rte_delay_ms
interface is used here to wait, but the CPU will not be released during
the waiting period of this interface, which will cause the statistics to
keep occupying the CPU. This is also the root cause of the soaring CPU.
The command interaction should belong to the control plane, and there
will not be too high requirements for performance. It is recommended to
wait for the interface iavf_msec_delay to complete without taking up the
CPU time.
Fixes: 22b123a36d ("net/avf: initialize PMD")
Cc: stable@dpdk.org
Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
The macros PMD_*_LOG already include the line feed character.
Redundant \n are removed.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Chenbo Xia <chenbo.xia@intel.com>
