crypto/cnxk: update maximum sec crypto capabilities

Update the macro to include newly added ciphers.
Updated the functions populating caps to throw error
when max is exceeded.

Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>

crypto/cnxk: clear session data before populating

Clear session data before populating fields to not have garbage data.

Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>

crypto/cnxk: support lookaside IPsec AES-CBC-HMAC-SHA256

Adding AES-CBC-HMAC-SHA256 support to lookaside IPsec PMD.

Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>

crypto/cnxk: enable allocated queues only

Only enable/disable queue pairs that are allocated during cryptodev
start/stop.

Fixes: 52008104e9a6 ("crypto/cnxk: update instruction queue in start/stop")
Cc: stable@dpdk.org
Signed-off-by: Shijith Thotton <sthotton@marvell.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>

common/cnxk: update completion code

Update completion code to match v1.19 microcode release.

Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>

common/cnxk: add missing reserved fields

Added missing bitfields for ctx flush and add err
print for ctx flush failure.

Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>

common/cnxk: fix reset of fields

Copy DF/DSCP fields would get set based on ipsec_xform in the code
preceding this. Setting it again would cause the options to be reset.

Fixes: 78d03027f2cc ("common/cnxk: add IPsec common code")
Cc: stable@dpdk.org
Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>

common/cnxk: add bit fields for params

Added new structure with bit fields for params.

Signed-off-by: Archana Muniganti <marchana@marvell.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>

common/cnxk: support AES-XCBC key derivation

Add support for AES-XCBC key derivation.

Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>

common/cnxk: define minor opcodes for MISC

MISC CPT instruction behaves differently based on minor opcode.
Define the missing minor opcodes for MISC major opcode.

Signed-off-by: Aakash Sasidharan <asasidharan@marvell.com>
Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>

event/cnxk: add timer adapter periodic mode support

Add support for event timer adapter periodic mode capability.

Signed-off-by: Shijith Thotton <sthotton@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

common/cnxk: use XAQ create API for inline device

Use the XAQ AURA create and free API while initializing the
inline device.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

common/cnxk: add workaround for vWQE flush

Due to an errata writing to vWQE flush register might hang NIX.
Add workaround for vWQE flush hang by waiting for the max
coalescing timeout to flush out any pending vWQEs.

Fixes: ee48f711f3b0 ("common/cnxk: support NIX inline inbound and outbound setup")
Cc: stable@dpdk.org
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

common/cnxk: add telemetry endpoints to SSO

Add common telemetry endpoints for SSO.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

event/cnxk: fix QoS devargs parsing

Fix qos devargs parsing using incorrect datatype for the
structure elements.

Fixes: 38c2e3240ba8 ("event/cnxk: add option to control SSO HWGRP QoS")
Cc: stable@dpdk.org
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>

eventdev/eth_rx: add event port get API

This patch introduces new api for retrieving event port id
of eth rx adapter.

Signed-off-by: Naga Harish K S V <s.v.naga.harish.k@intel.com>
Acked-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>

examples/l3fwd: fix Rx burst size for event mode

While dequeuing the packets from the event device, burst size
is provided in the API. This was not getting properly
configured in the application. This patch correctly configures
the burst size.

Fixes: aaf58cb85b62 ("examples/l3fwd: add event port and queue setup")
Cc: stable@dpdk.org
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Sunil Kumar Kori <skori@marvell.com>

event/cnxk: add external clock support for timer

Add external clock support for cnxk timer adapter.

External clock mapping is as follows:
RTE_EVENT_TIMER_ADAPTER_EXT_CLK0 = TIM_CLK_SRC_10NS,
RTE_EVENT_TIMER_ADAPTER_EXT_CLK1 = TIM_CLK_SRC_GPIO,
RTE_EVENT_TIMER_ADAPTER_EXT_CLK2 = TIM_CLK_SRC_PTP,
RTE_EVENT_TIMER_ADAPTER_EXT_CLK3 = TIM_CLK_SRC_SYNCE,

TIM supports clock input from external GPIO, PTP, SYNCE clocks.
Input resolution is adjusted based on CNTVCT frequency for better
estimation.

Since TIM is unaware of input clock frequency, application is
expected to pass the frequency.
Example:
-a 0002:0e:00.0,tim_eclk_freq=122880000-0-0

The order of frequencies above is GPIO-PTP-SYNCE.

Signed-off-by: Shijith Thotton <sthotton@marvell.com>
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>

event/cnxk: update minimum interval calculation

Minimum supported interval should now be retrieved from
mailbox based on the clock source and clock frequency.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>

eventdev/eth_rx: fix missing internal port checks

When event delivery is through internal port, stats are maintained
by HW and we should avoid reading SW data structures for stats.
Fix missing internal port checks.

Fixes: 995b150c1ae1 ("eventdev/eth_rx: add queue stats API")
Cc: stable@dpdk.org
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jay Jayatheerthan <jay.jayatheerthan@intel.com>

doc: fix dlb2 guide

Number of direct credits, atomic inflight and history list are
updated to DLB2.0 supported sizes. As DLB2.0 does not provide
dev arg to override the default per-queue atomic inflight
allocation, it is removed from the documentation.

Fixes: f3cad285bb88 ("event/dlb2: add infos get and configure")
Cc: stable@dpdk.org
Signed-off-by: Rashmi Shetty <rashmi.shetty@intel.com>
Reviewed-by: Timothy McDaniel <timothy.mcdaniel@intel.com>

eal/linux: log hugepage create errors with filename

While debugging running DPDK service in a container, it is
useful to see which file creation failed.  Don't hide this
failure with DEBUG.

Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>

build: remove custom dependency checks in drivers

Some drivers currently have their own checks and give some non
consistent reasons when an internal dependency is unavailable.

drivers/meson.build also checks for internal dependencies via 'deps'.
Let's rely on it for consistency, and smaller code.

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Long Li <longli@microsoft.com>

build: make cfgfile optional

Allow disabling of the cfgfile library in builds.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Morten Brørup <mb@smartsharesystems.com>
Acked-by: David Marchand <david.marchand@redhat.com>

build: make "packet framework" optional

Add port, table and pipeline libraries - collectively often known as
the "packet framework" -  to the list of optional libraries, and
ensure tests can build with them disabled.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Morten Brørup <mb@smartsharesystems.com>
Acked-by: David Marchand <david.marchand@redhat.com>

build: make flow classification optional

Add the flow_classify library to the list of optional libraries, and
ensure tests can build with it disabled.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Morten Brørup <mb@smartsharesystems.com>
Acked-by: David Marchand <david.marchand@redhat.com>

build: make node optional

Allow the 'node' library to be disabled in builds.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Morten Brørup <mb@smartsharesystems.com>
Acked-by: David Marchand <david.marchand@redhat.com>

test: link against all enabled libs

Rather than maintaining a list of the libraries the unit tests need, and
having to conditionally include/omit optional libs from the list, we can
just link against all available libraries, simplifying the code
considerably.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Morten Brørup <mb@smartsharesystems.com>
Acked-by: David Marchand <david.marchand@redhat.com>

build: allow recursive disabling of libraries

Align the code in lib/meson.build with that in drivers/meson.build to
enable recursive disabling of libraries, i.e. if library b depends on
library a, disable library b if a is disabled (either explicitly or
implicitly). This allows libraries to be optional even if other DPDK
libs depend on them, something that was not previously possible.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Morten Brørup <mb@smartsharesystems.com>
Acked-by: David Marchand <david.marchand@redhat.com>

gpudev: add alignment for memory allocation

Similarly to rte_malloc, rte_gpu_mem_alloc accepts as
input the memory alignment size.

GPU driver should return GPU memory address aligned
with the input value.

Signed-off-by: Elena Agostini <eagostini@nvidia.com>

config: add arch define for Arm

As per design document, RTE_ARCH is the name of the architecture.
However, the definition was missing on Arm with meson build.
It impacts applications that refers to this string.

Added for Arm builds.

Fixes: b1d48c41189a ("build: support ARM with meson")
Cc: stable@dpdk.org
Reported-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>

examples/performance-thread: remove

Remove sample application which is not clear if it is still relevant.

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>

gpu/cuda: add NVIDIA GPU A100 identifier for DPU

Adding a new NVIDIA GPU identifier to let
driver recognize the A100 on a DPU card.

Signed-off-by: Elena Agostini <eagostini@nvidia.com>

gpu/cuda: fix memory list cleanup

Memory list cleanup (called by cuda_mem_free)
was not properly set the new head of the list
when deleting an entry.

Fixes: 1306a73b1958 ("gpu/cuda: introduce CUDA driver")
Cc: stable@dpdk.org
Signed-off-by: Elena Agostini <eagostini@nvidia.com>

test/dma: increase iterations of capacity test case

To ensure we catch any bugs in calculation due to wrap-around of the id
values, increase the number of iterations of the burst_capacity test.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Kevin Laatz <kevin.laatz@intel.com>

dma/idxd: fix wrap-around in burst capacity calculation

The burst capacity calculation code assumes that the write and read
(i.e. ids_returned) values both wrap at the ring-size, but the read
value instead wraps as UINT16_MAX. Therefore, instead of just adding
ring-size to the write value in case the read is greater, we need to
just always mask the result to ensure a correct, in-range, value.

Fixes: 9459de4edc99 ("dma/idxd: add burst capacity")
Cc: stable@dpdk.org
Reported-by: Sunil Pai G <sunil.pai.g@intel.com>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Tested-by: Sunil Pai G <sunil.pai.g@intel.com>
Acked-by: Kevin Laatz <kevin.laatz@intel.com>

dma/idxd: fix paths to driver sysfs directory

Recent kernel changes[1][2] mean that we cannot guarantee that the paths
in sysfs used for creating/binding a DSA or workqueue instance will be
as given in the utility script, since they are now "compatibility-mode
only". Update script to support both new paths and compatibility ones.

[1] https://lore.kernel.org/all/162637445139.744545.6008938867943724701.stgit@djiang5-desk3.ch.intel.com/
[2] https://lore.kernel.org/all/162637468705.744545.4399080971745974435.stgit@djiang5-desk3.ch.intel.com/

Fixes: 01863b9d2354 ("raw/ioat: include example configuration script")
Cc: stable@dpdk.org
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Kevin Laatz <kevin.laatz@intel.com>

dma/idxd: fix burst capacity calculation

When the maximum burst size supported by HW is less than the available
ring space, incorrect capacity was returned when there was already some
jobs queued up for submission. This was because the capacity calculation
failed to subtract the number of already-enqueued jobs from the max
burst size. After subtraction is done, ensure that any negative values
(which should never occur if the user respects the reported limits), are
clamped to zero.

Fixes: 9459de4edc99 ("dma/idxd: add burst capacity")
Cc: stable@dpdk.org
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Kevin Laatz <kevin.laatz@intel.com>
Tested-by: Jiayu Hu <jiayu.hu@intel.com>

bus/ifpga: remove useless check while browsing devices

reported by code analysis tool C++test (version 10.4):

  /build/dpdk-20.11/drivers/bus/ifpga/ifpga_bus.c
  67    Condition "afu_dev" is always evaluated to true
  81    Condition "afu_dev" is always evaluated to true

The "for" loop already checks that afu_dev is not NULL.

Fixes: 05fa3d4a6539 ("bus/ifpga: add Intel FPGA bus library")
Cc: stable@dpdk.org
Signed-off-by: Maxime Gouin <maxime.gouin@6wind.com>
Reviewed-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Kevin Traynor <ktraynor@redhat.com>
Acked-by: Rosen Xu <rosen.xu@intel.com>

eal: add OS defines for C conditional checks

Define a set of macros in the build configuration to allow C runtime
code to check the current OS environment. This saves the user having to
use ifdefs for e.g. disabling particular tests on Windows.
See included documentation changes for usage examples.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

devtools: remove ugly workaround from get maintainer

Linux kernel 'get_maintainer.pl' script supports running out of Linux
tree since commit
31bb82c9caa9 ("get_maintainer: allow usage outside of kernel tree")

As commit is a few years old now, integrating it to DPDK and removing
ugly workaround for it.

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>

doc: remove dependency on findutils on FreeBSD

Standard "find" on BSD does not support the "-printf" so gfind from
findutils package was used to enable full doc builds. We can remove this
extra dependency by using "sed" and "tr" to adjust the output from
regular find instead.

Fixes: 8260f4f98cfe ("mk: use script to generate examples.dox")
Fixes: 499fe9dfcfc7 ("doc: add dependency on examples for API doxygen")
Fixes: 897e55c8d27f ("doc: fix Doxygen examples build on FreeBSD")
Cc: stable@dpdk.org
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>

buildtools: fix AVX512 check for Python 3.5

Python 3.5 subprocess.run() does not have a capture_output
parameter (it is present only in 3.7 and up).
Capture output by using subprocess.PIPE for stdout instead.

Fixes: bb9cd91095b3 ("buildtools: make AVX512 check portable")
Cc: stable@dpdk.org
Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Acked-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>

drivers: remove octeontx2 drivers

As per the deprecation notice,  In the view of enabling unified driver
for octeontx2(cn9k)/octeontx3(cn10k), removing drivers/octeontx2
drivers and replace with drivers/cnxk/ which
supports both octeontx2(cn9k) and octeontx3(cn10k) SoCs.

This patch does the following

- Replace drivers/common/octeontx2/ with drivers/common/cnxk/
- Replace drivers/mempool/octeontx2/ with drivers/mempool/cnxk/
- Replace drivers/net/octeontx2/ with drivers/net/cnxk/
- Replace drivers/event/octeontx2/ with drivers/event/cnxk/
- Replace drivers/crypto/octeontx2/ with drivers/crypto/cnxk/
- Rename config/arm/arm64_octeontx2_linux_gcc as
  config/arm/arm64_cn9k_linux_gcc
- Update the documentation and MAINTAINERS to reflect the same.
- Change the reference to OCTEONTX2 as OCTEON 9. Old release notes and
the kernel related documentation is not accounted for this change.

Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>
Acked-by: Ruifeng Wang <ruifeng.wang@arm.com>

regex/cn9k: use cnxk infrastructure

update driver to use the REE cnxk code
replace octeontx2/otx2 with cn9k

Signed-off-by: Liron Himi <lironh@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

common/cnxk: support REE

extend cnxk infrastructure to support REE

Signed-off-by: Liron Himi <lironh@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

common/cnxk: add REE mbox definitions

add REE mbox definitions

Signed-off-by: Liron Himi <lironh@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

common/cnxk: add REE HW definitions

adding REE (Regular Expression Engine) HW definitions

Signed-off-by: Liron Himi <lironh@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

doc: update LTS release cadence

Regular LTS releases have previously aligned to DPDK main branch
releases so that fixes being backported have already gone through
DPDK main branch release validation.

Now that DPDK main branch has moved to 3 releases per year, the LTS
releases should continue to align with it and follow a similar release
cadence.

Update stable docs to reflect this.

Signed-off-by: Kevin Traynor <ktraynor@redhat.com>

maintainers: update for stable branches

Christian and Xueming are both already maintaining LTS releases.

Update the MAINTAINERS file to reflect this.

Cc: stable@dpdk.org
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>

maintainers: update for testpmd

Add Yuying Zhang as a co-maintainer.

Signed-off-by: Yuying Zhang <yuying.zhang@intel.com>
Acked-by: Xiaoyun Li <xiaoyun.li@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>

doc: replace deprecated distutils version parsing

When using Python 3.10, this warning appears:
  DeprecationWarning: The distutils package is deprecated
  and slated for removal in Python 3.12.
  Use setuptools or check PEP 632 for potential alternatives

The PEP 632 recommends replacing "distutils.version" with "packaging".

Bugzilla ID: 914
Cc: stable@dpdk.org
Reported-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Tested-by: Jerin Jacob <jerinj@marvell.com>

fix spelling in comments and strings

The tool comes from https://github.com/jsoref

Signed-off-by: Josh Soref <jsoref@gmail.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>

version: 22.03-rc0

Start a new release cycle with empty release notes.
Bump version and ABI minor.
Enable ABI checks.

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>

version: 21.11.0

Signed-off-by: David Marchand <david.marchand@redhat.com>

maintainers: update for compressdev

Replacing myself with Fan as compressdev API maintainer.

Signed-off-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Fan Zhang <roy.fan.zhang@intel.com>
Acked-by: Ashish Gupta <ashish.gupta@marvell.com>

devtools: forbid additions of __reserved

__reserved is a reserved keyword in Windows system headers.

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Tested-by: Conor Walsh <conor.walsh@intel.com>

ethdev: announce migration to generic flow modify action

The generic RTE_FLOW_ACTION_TYPE_MODIFY_FIELD action was
introduced by [1]. This action provides an unified way
to perform various arithmetic and transfer operations over
packet network header fields and packet metadata.

[1] 73b68f4c54a0 ("ethdev: introduce generic modify flow action")

On other side there are a bunch of multiple legacy actions,
that can be superseded by the generic MODIFY_FIELD action:

RTE_FLOW_ACTION_TYPE_OF_SET_MPLS_TTL
RTE_FLOW_ACTION_TYPE_OF_DEC_MPLS_TTL
RTE_FLOW_ACTION_TYPE_OF_SET_NW_TTL
RTE_FLOW_ACTION_TYPE_OF_DEC_NW_TTL      sfc
RTE_FLOW_ACTION_TYPE_OF_COPY_TTL_OUT
RTE_FLOW_ACTION_TYPE_OF_COPY_TTL_IN
RTE_FLOW_ACTION_TYPE_SET_IPV4_SRC       bnxt, cxgbe, mlx5
RTE_FLOW_ACTION_TYPE_SET_IPV4_DST       bnxt, cxgbe, mlx5
RTE_FLOW_ACTION_TYPE_SET_IPV6_SRC       cxgbe, mlx5
RTE_FLOW_ACTION_TYPE_SET_IPV6_DST       cxgbe, mlx5
RTE_FLOW_ACTION_TYPE_SET_TP_SRC         cxgbe, mlx5
RTE_FLOW_ACTION_TYPE_SET_TP_DST         cxgbe, mlx5
RTE_FLOW_ACTION_TYPE_DEC_TTL            mlx5, sfc
RTE_FLOW_ACTION_TYPE_SET_TTL            mlx5
RTE_FLOW_ACTION_TYPE_SET_MAC_SRC        cxgbe, mlx5
RTE_FLOW_ACTION_TYPE_SET_MAC_DST        cxgbe, mlx5
RTE_FLOW_ACTION_TYPE_INC_TCP_SEQ        mlx5
RTE_FLOW_ACTION_TYPE_DEC_TCP_SEQ        mlx5
RTE_FLOW_ACTION_TYPE_INC_TCP_ACK        mlx5
RTE_FLOW_ACTION_TYPE_DEC_TCP_ACK        mlx5
RTE_FLOW_ACTION_TYPE_SET_IPV4_DSCP      mlx5
RTE_FLOW_ACTION_TYPE_SET_IPV6_DSCP      mlx5
RTE_FLOW_ACTION_TYPE_OF_SET_VLAN_VID    bnxt, cnxk, cxgbe, enic,
                                        mlx5, octeontx2, sfc
RTE_FLOW_ACTION_TYPE_OF_SET_VLAN_PCP    bnxt, cnxk, cxgbe, enic,
                                        mlx5, octeontx2, sfc
RTE_FLOW_ACTION_TYPE_SET_TAG            mlx5
RTE_FLOW_ACTION_TYPE_SET_META           mlx5

This note deprecates the following RTE Flow actions,
as not supported by any of PMDs:

RTE_FLOW_ACTION_TYPE_OF_SET_MPLS_TTL
RTE_FLOW_ACTION_TYPE_OF_DEC_MPLS_TTL
RTE_FLOW_ACTION_TYPE_OF_SET_NW_TTL
RTE_FLOW_ACTION_TYPE_OF_COPY_TTL_OUT
RTE_FLOW_ACTION_TYPE_OF_COPY_TTL_IN

The following actions are supposed to be deprecated in 22.07
and replaced by generic field modify action:

RTE_FLOW_ACTION_TYPE_OF_DEC_NW_TTL
RTE_FLOW_ACTION_TYPE_SET_IPV4_SRC
RTE_FLOW_ACTION_TYPE_SET_IPV4_DST
RTE_FLOW_ACTION_TYPE_SET_IPV6_SRC
RTE_FLOW_ACTION_TYPE_SET_IPV6_DST
RTE_FLOW_ACTION_TYPE_SET_TP_SRC
RTE_FLOW_ACTION_TYPE_SET_TP_DST
RTE_FLOW_ACTION_TYPE_DEC_TTL
RTE_FLOW_ACTION_TYPE_SET_TTL
RTE_FLOW_ACTION_TYPE_SET_MAC_SRC
RTE_FLOW_ACTION_TYPE_SET_MAC_DST
RTE_FLOW_ACTION_TYPE_INC_TCP_SEQ
RTE_FLOW_ACTION_TYPE_DEC_TCP_SEQ
RTE_FLOW_ACTION_TYPE_INC_TCP_ACK
RTE_FLOW_ACTION_TYPE_DEC_TCP_ACK
RTE_FLOW_ACTION_TYPE_SET_IPV4_DSCP
RTE_FLOW_ACTION_TYPE_SET_IPV6_DSCP
RTE_FLOW_ACTION_TYPE_SET_TAG
RTE_FLOW_ACTION_TYPE_SET_META

The VLAN set actions are interrelated to VLAN header insertion/removal
and supported by multiple PMDs and widely used by applications and
not supposed to be deprecated due to potential large impact on
drivers and applications.

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Olivier Matz <olivier.matz@6wind.com>

devtools/cocci: add script for ethdev namespace change

The cocci script is to help add prefix 'RTE_ETH' namespace to enum
& macro of ethdev library. It helps in automating these changes for
applications. The script won't make changes in the code comment part.
Usage: spatch <script path> <app path>

Signed-off-by: Aman Singh <aman.deep.singh@intel.com>
Tested-by: Ferruh Yigit <ferruh.yigit@intel.com>

doc: add details for new test structure

The testing guide is now updated to include details about
using sub-testsuites.
Some example code is given to demonstrate how they can be used.

A note is also added to highlight the need for using vdev EAL args
when running cryptodev tests.

Signed-off-by: Ciara Power <ciara.power@intel.com>
Acked-by: Fan Zhang <roy.fan.zhang@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>

doc: add a guide for developing unit tests

The DPDK testing infrastructure includes a comprehensive set of
libraries, utilities, and CI integrations for developers to test
their code changes.  This isn't well documented, however.

Document the basics for adding a test suite to the infrastructure
and enabling that test suite for continuous integration platforms
so that newer developers can understand how to develop test suites
and test cases.

Signed-off-by: Aaron Conole <aconole@redhat.com>
Acked-by: Ciara Power <ciara.power@intel.com>
Acked-by: Fan Zhang <roy.fan.zhang@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>

doc: add eventdev feature matrices

Patch adds feature matrices for event dev, Rx/Tx, Crypto
and Timer adapters.

Along with, .ini files are added for all supported PMDs.

Signed-off-by: Sunil Kumar Kori <skori@marvell.com>

doc: add RIB and FIB programmer guides

Currently, programmer's guide for the RIB and FIB libraries are missing.
This commit adds them.

Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>
Reviewed-by: Conor Walsh <conor.walsh@intel.com>

doc: fix a typo in EAL guide

Change from "how many segments each segment can have" to
"how many segments each segment list can have".

Fixes: b31739328354 ("doc: update guides for memory subsystem")
Cc: stable@dpdk.org
Signed-off-by: Kefu Chai <tchaikov@gmail.com>

doc: replace machine option in build guide

Update the docs to reflect the two new variables, cpu_instruction_set
for non-arm builds and platform for arm builds.

Fixes: bf66003b51ec ("build: use platform for generic and native builds")
Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech>

doc: add traffic metering API walk-through

Added a diagram to document meter library components
and added text for steps performed by the application to
configure the traffic meter and policing library.

Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>

doc: fix typos in examples code

Fix documentation typos that are generating spurious CI warnings.

Fixes: 9a212dc06c7a ("doc: use code snippets in sample app guides")
Signed-off-by: Ray Kinsella <mdr@ashroe.eu>
Acked-by: Aaron Conole <aconole@redhat.com>
Reviewed-by: Conor Walsh <conor.walsh@intel.com>

doc: add tested platforms with Mellanox NICs

Add tested platforms with Mellanox NICs to the 21.11 release notes.

Signed-off-by: Raslan Darawsheh <rasland@nvidia.com>

doc: add tested platforms with NXP SoCs

Add tested NXP platform to v21.11 release note for NXP enetfec PMD.

Signed-off-by: Apeksha Gupta <apeksha.gupta@nxp.com>

doc: add power management scale mode reaction time note

When using PMD Power Management, scale mode reacts slower than
monitor mode and pause mode. Add note in user guide to this
effect.

Signed-off-by: David Hunt <david.hunt@intel.com>

net/mlx5: fix devargs validation for multi-class probing

The mlx5_args function reads the devargs and checks if they are valid
for this driver and if not it returns an error.

This was normal behavior as long as all the devargs come to this driver,
but since it is possible to run several drivers together, the function
may return an error for another driver's devarg even though it is
completely valid.
In addition the function does not allow the user to know which of the
devargs is incorrect, but returns an error without printing the
unknown devarg.

This patch eliminates the error return in the case of an unknown devarg,
and prints a warning for each such devarg specifically.

Fixes: 7b4f1e6bd367 ("common/mlx5: introduce common library")
Cc: stable@dpdk.org
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>

common/mlx5: fix memory region lookup on slow path

Memory region (MR) was being looked up incorrectly
for the data address of an externally-attached mbuf.
A search was attempted for the mempool of the mbuf,
while mbuf data address does not belong to this mempool
in case of externally-attached mbuf.
Only attempt the search:
1) for not externally-attached mbufs;
2) for mbufs coming from MPRQ mempool;
3) for externally-attached mbufs from mempools
   with pinned external buffers.

Fixes: 08ac03580ef2 ("common/mlx5: fix mempool registration")
Signed-off-by: Dmitry Kozlyuk <dkozlyuk@nvidia.com>
Reviewed-by: Matan Azrad <matan@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>

app/gpudev: remove memory leaks

Remove all memory leaks in case of errors in
test-gpudev application.

Fixes: e818c4e2bf50 ("gpudev: add memory API")
Fixes: c7ebd65c1372 ("gpudev: add communication list")
Signed-off-by: Elena Agostini <eagostini@nvidia.com>

gpudev: remove unnecessary memory barrier

Remove unnecessary rte_gpu_wmb from rte_gpu_comm_populate_list_pkts.
It causes a performance degradation in case of NVIDIA GPU V100.

This change doesn't affect any functionality as the status resides
in CPU registered memory.

Fixes: c7ebd65c1372 ("gpudev: add communication list")
Signed-off-by: Elena Agostini <eagostini@nvidia.com>

compress/mlx5: fix double close of device context

The context of the device opens once in the common probe and closes with
its removal.

If the probe of one of the drivers fails, it releases its resources and
then the common closes the context.
But mistakenly in the compress probe, if there isn't enough capabilities
to support compress operations, it closes the device and then common
probe closes it again.

Remove the redundant closing from compress probe.

Fixes: 2efd26544554 ("compress/mlx5: support partial transformation")
Cc: stable@dpdk.org
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>

examples/ipsec-secgw: fix L4 length for TSO packets

Using RTE_MBUF_F_TX_TCP_SEG requires L4 length to be set.

Fixes: a7f32947a316 ("examples/ipsec-secgw: support TCP TSO")
Signed-off-by: Radu Nicolau <radu.nicolau@intel.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>

maintainers: update for testpmd

Add Aman Singh as co-maintainer of testpmd.

Make the title description more accurate for this tool.

Signed-off-by: Aman Singh <aman.deep.singh@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>

doc: fix typo in coding style

This patch fixes a typo in DPDK Coding Style, in Return Values section,
i.e. replaces "indicated may" with "indicated by".

Fixes: 36032e46be40 ("doc: add coding style")
Cc: stable@dpdk.org
Signed-off-by: Dariusz Sosnowski <dsosnowski@nvidia.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>

doc: capitalise PMD

The doc's contain references to pmd but the proper use is to use PMD.

Cc: stable@dpdk.org
Signed-off-by: Sean Morrissey <sean.morrissey@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Reviewed-by: Conor Walsh <conor.walsh@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

fix PMD wording

Removing the use of driver following PMD as its unnecessary.

Cc: stable@dpdk.org
Signed-off-by: Sean Morrissey <sean.morrissey@intel.com>
Signed-off-by: Conor Fogarty <conor.fogarty@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Reviewed-by: Conor Walsh <conor.walsh@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

remove repeated 'the' in the code

Remove the use of double "the" as it does not make sense.

Cc: stable@dpdk.org
Signed-off-by: Sean Morrissey <sean.morrissey@intel.com>
Signed-off-by: Conor Fogarty <conor.fogarty@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Reviewed-by: Conor Walsh <conor.walsh@intel.com>
Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

devtools: suppress checkpatch warning about strlcpy

Current version of Linux kernel checkpatch now complains about
usages of strlcpy because the Linux kernel prefers strscpy.

WARNING:STRLCPY: Prefer strscpy over strlcpy

But DPDK project allows strlcpy, therefore suppress the warning.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>

devtools: check prefix for libraries patches

The convention in DPDK is to directly use library names as prefix,
without a lib/.

Signed-off-by: David Marchand <david.marchand@redhat.com>

devtools: add acronyms in dictionary for commit checks

FD      -> File Descriptor
GUID    -> Globally Unique Identifier
Netlink -> Linux Netlink socket family
PCIe    -> Peripheral Component Interconnect Express

Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>

devtools: add Marvell acronyms for commit checks

Update word list with Marvell specific acronyms.

CPT  -> Cryptographic Accelerator Unit
CQ   -> Completion Queue
LBK  -> Loopback Interface Unit
LMT  -> Large Atomic Store Unit
MCAM -> Match Content Addressable Memory
NIX  -> Network Interface Controller Unit
NPA  -> Network Pool Allocator
NPC  -> Network Parser and CAM Unit
ROC  -> Rest Of Chip
RQ   -> Receive Queue
RVU  -> Resource Virtualization Unit
SQ   -> Send Queue
SSO  -> Schedule Synchronize Order Unit
TIM  -> Timer Unit

Suggested-by: Ferruh Yigit <ferruh.yigit@intel.com>
Signed-off-by: Ashwin Sekhar T K <asekhar@marvell.com>
Reviewed-by: Jerin Jacob <jerinj@marvell.com>

devtools: clarify that lines up to 100 characters are ok

Since we allow line lengths of up to 100, and the CI checkpatches job
only check for that amount, the rest of our tooling and docs should
reflect this reality. Therefore we can:

* adjust the editorconfig to use that value, to save editors (e.g. vim)
  from automatically wrapping lines at 80 characters when typing.
  [Since python checkers all seem to expect 79 character lines max, add
  for python only a 79-char max line length.]

* change the default line length setting in checkpatches script to 100
  so as it matches CI and pre-merge checks.

* update the docs to clarify that while 80 chars is recommended, up to
  100 characters is acceptable.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Chenbo Xia <chenbo.xia@intel.com>
Acked-by: Chengwen Feng <fengchengwen@huawei.com>
Acked-by: Conor Walsh <conor.walsh@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>

version: 21.11-rc4

Signed-off-by: David Marchand <david.marchand@redhat.com>

net/mlx5: fix shared Rx queue segment configuration match

While joining the shared Rx queue to the existing queue group,
the queue configurations is checked to be the same as it was
specified in the first group queue creation - all shared
queues should be created with identical configurations.

During the Rx queue creation the buffer split segment
configuration can be altered - the zero segment sizes are
substituted with the actual ones, inherited from the pools,
number of segments can be extended to cover the maximal
packet length, etc. It means the actual queue segment
configuration can not be used directly to match the
configuration provided in the queue setup call.

To resolve an issue we should store original parameters
in the shared queue structure and perform the check against
one of these stored ones.

Fixes: 09c2555303be ("net/mlx5: support shared Rx queue")
Signed-off-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>

net/mlx5: fix GENEVE and VXLAN-GPE flow item matching

GENEVE and VXLAN-GPE item matching is done similarly to GRE matching.
Users can skip the specification of the protocol type and expect that
this type is deducted from the inner header type automatically.
But the inner header type may not be specified in order to match all the
protocol types. In this case, PMD should not specify the protocol type.
Check if we have the inner header type before setting the protocol type.

Fixes: 690391dd0e8b ("net/mlx5: fix GENEVE protocol type translation")
Fixes: 861fa3796f75 ("net/mlx5: fix VXLAN-GPE next protocol translation")
Cc: stable@dpdk.org
Signed-off-by: Alexander Kozyrev <akozyrev@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>

net/mlx5: fix GRE flow item matching

GRE protocol type is implicitly set in the matching translation in case
an application doesn't specify any type explicitly in a flow rule.
It is extracted from the inner header type, but this type may be absent.
In this case, GRE item matching is broken. Check if we have the inner
header type before setting it to allow matching on all GRE packets.

Fixes: be26e81bfc1c ("net/mlx5: fix GRE protocol type translation")
Cc: stable@dpdk.org
Signed-off-by: Alexander Kozyrev <akozyrev@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>

net/mlx5: fix Rx queue reference count for indirect RSS

mlx5_ind_table_obj_modify() was not changing the reference counters
of neither the new set of RxQs, nor the old set of RxQs.
On the other hand, creation of the RSS incremented the RxQ refcnt.
If an RxQ was present in both the initial and the modified set,
its reference counter was incremented one extra time
compared to the queues that were only present in the new set.
This prevented releasing said RxQ resources on port stop:

    flow indirect_action 0 create action_id 1 \
        action rss queues 0 1 end / end
    flow indirect_action 0 update 1 \
        action rss queues 2 3 end / end
    quit
    ...
    mlx5_net: mlx5.c:1622: mlx5_dev_close():
        port 0 some Rx queue objects still remain
    mlx5_net: mlx5.c:1626: mlx5_dev_close():
        port 0 some Rx queues still remain

Increment reference counters for the new set of RxQs
and decrement them for the old set of RxQs when needed.
Remove explicit referencing of RxQ from mlx5_ind_table_obj_attach()
because it reuses mlx5_ind_table_obj_modify() code doing this.

Fixes: ec4e11d41d12 ("net/mlx5: preserve indirect actions on restart")
Cc: stable@dpdk.org
Signed-off-by: Dmitry Kozlyuk <dkozlyuk@nvidia.com>
Reviewed-by: Matan Azrad <matan@nvidia.com>

net/mlx5: fix indirect RSS creation when port is stopped

mlx5_ind_table_obj_setup() was incrementing RxQ reference counters
even when the port was stopped, which prevented RxQ release
and triggered an internal assertion.
Only increment reference counter when the port is started.

Fixes: ec4e11d41d12 ("net/mlx5: preserve indirect actions on restart")
Cc: stable@dpdk.org
Signed-off-by: Dmitry Kozlyuk <dkozlyuk@nvidia.com>
Reviewed-by: Matan Azrad <matan@nvidia.com>

app/testpmd: fix hexadecimal parser with odd length

Current hex string parser assumes input has even characters number.
The parser fails input string with odd length.

The patch parses hex strings with even and odd length.
Parse result of an input with odd length will match result of
even length input, that has `0` as MSB, following by the original
sequence.
For example:
"0x1" results in *dst={0x01, 0x00}, *size=1
"0xabc" results in *dst={0x0a, 0xbc, 0x00}, *size=2

Fixes: 169a9fed1f4c ("app/testpmd: fix hex string parser support for flow API")
Cc: stable@dpdk.org
Signed-off-by: Gregory Etelson <getelson@nvidia.com>
Reviewed-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>

doc: fix memif driver acronyms

The commit d250589d5702 ("net/memif: replace master/slave arguments")
replaced master/slave terms to server/client terms.
Fix the documentation to reflect the same.

Fixes: d250589d5702 ("net/memif: replace master/slave arguments")
Cc: stable@dpdk.org
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/memif: allow stopping and closing device

Bugzilla ID: 888
Fixes: febc855b358e ("ethdev: forbid closing started device")
Cc: stable@dpdk.org
Signed-off-by: Junxiao Shi <git@mail1.yoursunny.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

doc: remove flow mark Rx offload deprecation notice

The problem is solved using Rx metadata delivery negotiation API [1].

[1] commit f6d8a6d3fad7 ("ethdev: negotiate delivery of packet metadata from HW to PMD")

Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>

doc: remove custom kernel patch link in ena guide

ENAv2 device requires write combining support which isn't supported by
the upstream vfio-pci. amzn-driver repository provided non-upstream
patch to enable this feature and it was linked directly by the ENA PMD
guide.

To avoid custom kernel patch linking, the user is now guided to the AWS
ENA PMD documentation, which describes vfio-pci and ENAv2 issue more
deeply with possible workarounds on how to resolve it.

Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>

net/mlx5: fix crash on close after failed start

If mlx5_rxq_start() failed and rxq_ctrl was not initialized,
mlx5_rxq_obj_verify() would segfault in an attempt to dereference it.
Add a check that rxq_ctrl is not NULL before accessing its members.

Fixes: 09c2555303be ("net/mlx5: support shared Rx queue")
Signed-off-by: Dmitry Kozlyuk <dkozlyuk@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>

common/mlx5: fix shared memory region ranges allocation

Memory regions (MRs) were allocated in one chunk of memory with a
mempool registration object. However, MRs can be reused among
different mempool registrations.

When the registration that allocated the MRs originally was
destroyed, the dangling pointers to the MRs could be left in other
registrations sharing these MRs.

Splitting the memory allocation of registration structure and MRs in
this commit solves this pointer reference issue.

Fixes: 690b2a88c2f7 ("common/mlx5: add mempool registration facilities")
Signed-off-by: Bing Zhao <bingz@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
Reviewed-by: Dmitry Kozlyuk <dkozlyuk@nvidia.com>

net/mlx5: fix reference count on detached indirect action

This patch fixes segfault which was triggered when port, with indirect
actions created, was closed. Segfault was occurring only when
RTE_LIBRTE_MLX5_DEBUG was defined. It was caused by redundant decrement
of RX queues refcount:

- refcount was decremented when port was stopped and indirect actions
were detached from RX queues (port stop),
- refcount was decremented when indirect actions objects were destroyed
(port close or destroying of indirect action).

This patch fixes behavior. Dereferencing Rx queues is done if and only
if indirect action is explicitly destroyed by the user or detached on
port stop. Dereferencing Rx queues on action destroy operation depends
on an argument to the wrapper of indirect action destroy operation,
introduced in this patch.

Fixes: ec4e11d41d12 ("net/mlx5: preserve indirect actions on restart")
Cc: stable@dpdk.org
Signed-off-by: Dariusz Sosnowski <dsosnowski@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>