This commit enables asymmetric crypto in generation four
devices (4xxx).
Signed-off-by: Arek Kusztal <arkadiuszx.kusztal@intel.com>
Acked-by: Kai Ji <kai.ji@intel.com>
This patch fix the cipher & auth offset and length values when convert
mbuf to vector chain for QAT build op.
Fixes: a815a04cea ("crypto/qat: support symmetric build op request")
Cc: stable@dpdk.org
Signed-off-by: Kai Ji <kai.ji@intel.com>
Acked-by: Fan Zhang <roy.fan.zhang@intel.com>
The API of the OpenSSL library has changed with version 3.0. This results
in a lot of compiler warnings like
../dpdk/drivers/crypto/ccp/ccp_crypto.c:182:9:
warning: ‘SHA256_Transform’ is deprecated:
Since OpenSSL 3.0 [-Wdeprecated-declarations]
As many Linux distributions still use elder OpenSSL libraries we cannot
change the used API now. Instead define OPENSSL_API_COMPAT to indicate
that we are using the OpenSSL 1.1.0 API.
OPENSSL_API_COMPAT is introduced in *.c files and not in *.h files as some
*.c files directly include OpenSSL headers.
Fixes: d61f70b4c9 ("crypto/libcrypto: add driver for OpenSSL library")
Cc: stable@dpdk.org
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Tested-by: Daxue Gao <daxuex.gao@intel.com>
Tested-by: David Marchand <david.marchand@redhat.com>
Acked-by: Kai Ji <kai.ji@intel.com>
Add SGL support for chacha20_poly1305 algorithm through JOB API.
Supports IN-PLACE SGL, OOP SGL IN and LB OUT,
and OOP SGL IN and SGL OUT.
Feature flags not added, as the PMD does not support SGL for all
other algorithms.
Signed-off-by: Ciara Power <ciara.power@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Add SGL support for GCM algorithm through JOB API.
This change supports IN-PLACE SGL, OOP SGL IN and LB OUT,
and OOP SGL IN and SGL OUT.
Feature flags are not added, as the PMD does not yet support SGL for
all other algorithms.
If an SGL op for an unsupported algorithm is being processed,
a NULL job is submitted instead.
Signed-off-by: Ciara Power <ciara.power@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
For getting event crypto metadata from crypto_op,
the new API rte_cryptodev_get_session_event_mdata can be used
directly instead of getting userdata inside PMD.
Signed-off-by: Akhil Goyal <gakhil@marvell.com>
Acked-by: Fan Zhang <roy.fan.zhang@intel.com>
Acked-by: Abhinandan Gujjar <abhinandan.gujjar@intel.com>
Acked-by: Anoob Joseph <anoobj@marvell.com>
Using crypto devs requires the user to log in and the supplied DEK to be
encrypted with a KEK (keys encryption key).
KEK is burned once on the nic, along with credentials for users,
and for a user to log in, he is needed to supply his creds wrapped with
the KEK.
A device comes out of the Mellanox factory with a pre-defined import
method for each algorithm. The defined method could be wrapped
mode, so the device can be used as described above, or
plaintext mode, without the need to log in and wrap supplied DEKs.
Support crypto operations with the plaintext import method.
Signed-off-by: Raja Zidane <rzidane@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
This patch support copy, submit, completed and
completed status functionality of DMA driver.
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Add additional PMD APIs for DPAA2 QDMA driver for configuring
RBP, Ultra Short format, and Scatter Gather support
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
This patch support basic DMA operations which includes
device capability and channel setup.
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
The DPAA2 DMA driver is an implementation of the dmadev APIs,
that provide means to initiate a DMA transaction from CPU.
Earlier this was part of RAW driver, but with DMA drivers
added as separate flavor of drivers, this driver is being
moved to DMA drivers.
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
With DMA devices supported as a separate flavor of devices,
the DPAA2 QDMA driver is moved in the DMA devices.
This change removes the DPAA2 QDMA driver from raw devices.
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
The AltiVec header file is defining "vector", except in C++ build.
The keyword "vector" may conflict easily.
As a rule, it is better to use the alternative keyword "__vector",
so we will be able to #undef vector after including AltiVec header.
Later it may become possible to #undef vector in rte_altivec.h
with a compatibility breakage.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: David Christensen <drc@linux.vnet.ibm.com>
Enable GPU_REGISTERED flag in gpu/cuda driver in the memory list.
If a GPU memory address CPU mapped is freed before being
unmapped, CUDA driver unmaps it before freeing the memory.
Signed-off-by: Elena Agostini <eagostini@nvidia.com>
If an event queue flush does not complete after a fixed number of tries,
remaining queues are flushed before retrying the one with incomplete
flush.
Signed-off-by: Shijith Thotton <sthotton@marvell.com>
Added API to set queue attributes at runtime and API to get weight and
affinity.
Signed-off-by: Shijith Thotton <sthotton@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Since mailbox is now accessed from multiple threads, use lock to
synchronize access.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Signed-off-by: Shijith Thotton <sthotton@marvell.com>
Move event post-processing to a separate function.
Do complete event post-processing in tear-down functions to prevent
incorrect memory free.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Add additional checks while performing RTE_EVENT_OP_RELEASE to
ensure that there are no pending SWTAGs and FLUSHEs in flight.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Updated to allow overriding the default CQ depth of 32. Since there are
only 2048 DLB history list entries, increasing the CQ depth decreases
the number of available LDB ports to 2048/max_cq_depth. Resource query
will take this into account and return the correct maximum number of
LDB ports.
Signed-off-by: Timothy McDaniel <timothy.mcdaniel@intel.com>
Function roc_sso_hws_head_wait() expects a base as input pointer, and it
will itself get tag_op from the base. By passing tag_op instead of base
pointer to this function will add SSOW_LF_GWS_TAG register offset twice,
which will lead to accessing wrong register.
Fixes: 1f5b3d55c0 ("event/cnxk: store and reuse workslot status")
Cc: stable@dpdk.org
Signed-off-by: Volodymyr Fialko <vfialko@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
In packets with ESP header, the inner IP will be encrypted, and
its fields cannot be used for RSS hashing. So, ESP packets
can be hashed only by the outer IP layer.
So, when using RSS on ESP packets, hashing may not be efficient,
because the fields used by the hash functions are only the outer IPs,
causing all traffic belonging to all tunnels between a given
pair of GWs to land on one core.
Adding the SPI hash field can extend the spreading of IPsec packets.
Signed-off-by: Raja Zidane <rzidane@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
When a meter with RSS action being used, there might be several
sub-flows using different sub-policies in the flow splitting stage.
If there's no green action, there's an error that will always use the
same sub-policy for all sub-flows, some resources will be
overwritten and cannot be released, leading assert during port close.
This patch fixes this issue by checking both green and yellow queue
index during getting a blank sub-policy, to avoid the incorrect
resource overwrite.
Fixes: b38a12272b ("net/mlx5: split meter color policy handling")
Cc: stable@dpdk.org
Signed-off-by: Shun Hao <shunh@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Invoking bnxt_link_update_op() with wait_for_completion set would
result in the driver waiting for 10s in case the port link is down to
complete port initialization (dev_start_op()).
Change it by not waiting for the completion when invoking it in
dev_start_op()
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
In bnxt_free_all_filters(), all the filters attached to a vnic are removed.
But each of these filters hold a backreference ptr to the vnic and they
need to be reset to NULL now. Otherwise, during a normal testpmd quit, as
part of dev_close_op(), first bnxt_free_all_filters() is invoked in
dev_stop, followed by bnxt_free_filter_mem() from bnxt_uninit_resources(),
which finds a filter with a vnic back reference ptr and now
bnxt_hwrm_clean_up_l2_filter() also tries to remove the filter from the
vnic's filter list which was already done as part of
bnxt_free_all_filters().
Fixes: f0f6b5e6cf ("net/bnxt: fix reusing L2 filter")
Cc: stable@dpdk.org
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
If Firmware is still in reset process and returns the error
HWRM_ERR_CODE_HOT_RESET_PROGRESS, retry VER_GET command.
We have to do it in bnxt_handle_if_change_status().
Fixes: 0b53359123 ("net/bnxt: inform firmware about IF state changes")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Driver forces link down during port stop. But device is not obliged
link down in certain scenarios, even when forced. In that case,
subsequent link queries returns link as up.
Fixed to return link status as down when port is stopped.
Driver is already doing that for VF/NPAR/MH functions.
Fixes: c09f57b49c ("net/bnxt: add start/stop/link update operations")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Device is not obliged link down in certain scenarios, even
when forced. When FW does not allow any user other than the BMC
to shutdown the port, bnxt_get_hwrm_link_config() call always
returns link up. Force phy update always in that case,
else user configuration for speed/autoneg would not get applied
correctly.
Fixes: 7bc8e9a227 ("net/bnxt: support async link notification")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
The "active_fec_signal_mode" in HWRM_PORT_PHY_QCFG response
does not return correct value till the link is up. Driver cannot
rely on active_fec_signal_mode while setting autoneg speed.
While setting autoneg speed, driver is currently checking only
"auto_link_speed_mask". Fixed to check "auto_pam4_link_speed_mask"
as well. Also, while setting auto mode and setting speed mask,
driver will have to set both NRZ and PAM4 mask.
Fixes: c23f9ded03 ("net/bnxt: support 200G PAM4 link")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
The "active_fec_signal_mode" in HWRM_PORT_PHY_QCFG response is uint8_t.
So no need of endianness conversion while parsing response.
Also, signal_mode is the first 4bits of "active_fec_signal_mode".
Fixes: c23f9ded03 ("net/bnxt: support 200G PAM4 link")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
The programming of the RSS table was not taking into account if
any of the queues in the set were stopped prior to the flow
creation, hence leading to a vnic RSS config cmd failure thrown by
the FW.
Fix by programming only the active queues in the RSS action queue
set.
Fixes: 239695f754 ("net/bnxt: enhance RSS action support")
Cc: stable@dpdk.org
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Currently driver does not have a check for duplicate queue ids.
User must either specify all Rx queues created or no queues in the
flow create command. Repeating the queue index is invalid.
Also, moved the check for invalid queue to the beginning of the function.
Fixes: 239695f754 ("net/bnxt: enhance RSS action support")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
When an Rx queue is stopped and restarted, as part of that workflow,
for cards that have ring groups, we free and reallocate the ring group.
This new ring group is not communicated to the VNIC though via
HWRM_VNIC_CFG cmd.
Fix to issue HWRM_VNIC_CFG cmd on all adapters now in this scenario.
Fixes: ed0ae3502f ("net/bnxt: update ring group after ring stop start")
Cc: stable@dpdk.org
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Specifying a subset of Rx queues created by the application in
the "flow create" command is invalid.
User must either specify all Rx queues created or no queues.
Also removed a wrong comment as RSS action will not be supported
if user or application specifies MARK or COUNT action.
Fixes: 239695f754 ("net/bnxt: enhance RSS action support")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
'Count' action was never really implemented in the legacy/AFM model.
But there was some place holder code, remove it so that the user
will see a failure when a flow with 'count' action is being
created.
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
The HW only supports tunnel header parsing globally for supported tunnel
types. When a function uses one default VNIC to receive both the tunnel
and non-tunnel packets, applying the same stateless offload operation to
both tunnel and non-tunnel packets can cause problems in certain scenarios.
To workaround these problems, the firmware advertises no tunnel header
parsing capabilities to the driver using the HWRM_FUNC_QCAPS.
The driver must check this flag setting and accordingly not advertise
tunnel packet stateless offload capabilities to the stack.
If the device supports VXLAN, GRE, IPIP and GENEVE tunnel parsing,
then reports RX_OFFLOAD_OUTER_IPV4_CKSUM, RX_OFFLOAD_OUTER_UDP_CKSUM
and TX_OFFLOAD_OUTER_IPV4_CKSUM in the Rx/Tx offload capabilities of
the device.
Also, advertise tunnel TSO capabilities based on FW support.
Fixes: 0a6d2a7200 ("net/bnxt: get device infos")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
We are currently not handling RX/RSS modes correctly.
After launching testpmd with multiple RXQs, if the user tries to set
the number of RXQs to 1, driver is not updating the "hash_type"
and "hash_mode" values of the VNICs. As a result, driver issues
bnxt_vnic_rss_configure() unnecessarily and the FW command fails.
Fixed bnxt_mq_rx_configure() to update VNIC RSS fields unconditionally.
Fixes: 4191bc8f79 ("net/bnxt: handle multi queue mode properly")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
1. Added two functions bnxt_get_tx_port_offloads() and
bnxt_get_rx_port_offloads() to report the device
tx/rx offload capabilities to the application.
2. This avoids few duplicate code in the driver and make
VF-rep capability the same as VF.
3. This will help in selectively reporting offload capabilities
based on FW support.
Fixes: 0a6d2a7200 ("net/bnxt: get device infos")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Brought in the latest hsi_struct_def_dpdk.h.
HWRM API is now updated to version "1.10.2.83".
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>
Rx descriptor contains a valid bit which indicates readiness of the rest
of descriptor words. Hence, the word contains valid bit must be read
prior to other words.
In NEON vector path, two contiguous 8B descriptor are loaded to a single
NEON register. Given vector load ensures no 16B atomicity, read of the
word that includes valid bit could be reordered after read of other words.
In this case, data could be invalid.
Reloaded lower 64b after read barrier. This ensures what fetched is
correct.
Also fixed comments that not pertains to Arm platform architecture.
Fixes: deae85145c ("net/bnxt: handle multiple packets per loop in vector Rx")
Cc: stable@dpdk.org
Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
NEON vector path is built only when Arm platform is 64-bit.
The ifdefs in NEON path are of no use, hence remove.
Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
When no packet is received, there is no need to update completion raw cons.
Moved update down to remove unnecessary store in this case.
Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Add support for per port RQ in inline device thereby using
Aura/Pool attributes from that port specific first RQ.
When inline device is used with channel masking, it will
fallback to single RQ for all ethdev ports.
Also remove clamping up of CQ size for LBK ethdev when
inline inbound is enabled as now backpressure is supported
even on LBK ethdevs.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Fix hotplug detach sequence to handle case where first PCI
device that is hosting NPA LF is being destroyed while in use.
Fixes: 5a4341c849 ("net/cnxk: add platform specific probe and remove")
Cc: stable@dpdk.org
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Fix multi-seg extraction in vwqe path to avoid updating mbuf[]
array until it is used via cq0 path.
Fixes: 7fbbc981d5 ("event/cnxk: support vectorized Rx event fast path")
Cc: stable@dpdk.org
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Perform early MTU setup for event mode path in order
to update the Rx/Tx offload flags before Rx adapter setup
starts.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Add support for flow control in outbound inline path using
FC updates from CPT.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Enabled rte_security stats operation based on the configuration
of SA options set while creating session.
Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com>
Signed-off-by: Akhil Goyal <gakhil@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Added supported capabilities for various IPsec SA options.
Signed-off-by: Akhil Goyal <gakhil@marvell.com>
Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
When the packet is processed with inline IPsec offload,
the ol_flags were updated only with RTE_MBUF_F_RX_SEC_OFFLOAD.
But the hardware can also update the L3/L4 csum offload flags.
Hence, ol_flags are updated with RTE_MBUF_F_RX_IP_CKSUM_GOOD,
RTE_MBUF_F_RX_L4_CKSUM_GOOD, etc based on the microcode completion
codes.
Signed-off-by: Akhil Goyal <gakhil@marvell.com>
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Optimize Rx fast path for security packets by preprocessing
most of the operations such as sa pointer compute,
inner WQE pointer fetch and microcode completion translation
before the pkt is characterized as inbound inline pkt.
Preprocessed info will be discarded if packet is not
found to be security pkt. Also fix fetching of CQ word5
for vector mode. Get ucode completion code from CPT parse
header and RLEN from IP4v/IPv6 decrypted packet as it is
in same 64B cacheline as CPT parse header in most of
the cases. By this method, we avoid accessing an extra
cacheline
Fixes: c062f5726f ("net/cnxk: support IP reassembly")
Cc: stable@dpdk.org
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Added support for decrementing TTL(IPv4)/hoplimit(IPv6)
while doing inline IPsec processing if the security session
SA options is enabled with dec_ttl.
Signed-off-by: Akhil Goyal <gakhil@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
The rx offload flag need to be reset if IP reassembly flag
is not set while calling reassembly_conf_set.
Signed-off-by: Akhil Goyal <gakhil@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Changed environment variable name for specifying
debug IV for unit testing of inline IPsec offload
with known test vectors.
Signed-off-by: Akhil Goyal <gakhil@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Inline PF_FUNC is updated in ethdev_tel_handle_info(),
when inline device is attached to any dpdk process
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
For transport mode, roundup needs to be based on L4 data
and shouldn't include L3 length.
By including L3 length, rlen that is calculated and put in
send hdr would cross the final length of the packet in some
scenarios where padding is necessary.
Also when outer and inner checksum offload flags are enabled,
get the l2_len and l3_len from il3ptr and il4ptr.
Fixes: 55bfac717c ("net/cnxk: support Tx security offload on cn10k")
Cc: stable@dpdk.org
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Disable default inner L3/L4 checksum generation for outbound inline
path and enable based on SA options or RTE_MBUF flags as per
the spec. Though the checksum generation is not impacting much
performance, it is overwriting zero checksum for UDP packets
which is not always good.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Add barrier after meta batch free in scalar routine when
LMT lines are exactly full to make sure that next LMT line user
in Tx only starts writing the lines only when previous stoerl's
are complete.
Fixes: 4382a7ccf7 ("net/cnxk: support Rx security offload on cn10k")
Cc: stable@dpdk.org
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Update link info of LBK ethdev i.e AF's VF's as always up
and 100G. This is because there is no phy for the LBK interfaces
and driver won't get a link update notification for the same.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Support internal loopback mode on AF VF's using ROC by setting
Tx channel same as Rx channel.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Use aggregate level Round Robin Priority from mbox response instead of
fixing it to single macro. This is useful when kernel AF driver
changes the constant.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Inbound SA SPI if not in min-max range specified in devargs,
was marked as a warning. But this is not converted to debug
print because if the entry is found to be duplicate in the mask,
it will give another error print. Hence, warning print is not needed
and is now converted to debug print.
Signed-off-by: Akhil Goyal <gakhil@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Fix issues in mode where soft expiry is disabled in ROC.
When soft expiry support is not enabled in inline device,
memory is not allocated for the ring base array and should
not be accessed.
Fixes: bea5d990a9 ("net/cnxk: support outbound soft expiry notification")
Cc: stable@dpdk.org
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
SoC run platform file is not present in CN9k so probing
is done for CN10k devices
Signed-off-by: Rakesh Kudurumalla <rkudurumalla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Fix SQ flush sequence to issue NIX RX SW Sync after SMQ flush.
This sync ensures that all the packets that were in-flight are
flushed out of memory.
This patch also fixes NULL return issues reported by
static analysis tool in Traffic Manager and sync's mailbox
to that of the kernel version.
Fixes: 05d727e8b1 ("common/cnxk: support NIX traffic management")
Fixes: 0b7e667ee3 ("common/cnxk: enable packet marking")
Cc: stable@dpdk.org
Signed-off-by: Satha Rao <skoteshwar@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Add new API to configure the SA table entries with new CPT PKIND
when timestamp is enabled.
Signed-off-by: Vidya Sagar Velumuri <vvelumuri@marvell.com>
Acked-by: Ray Kinsella <mdr@ashroe.eu>
Acked-by: Jerin Jacob <jerinj@marvell.com>
With timestamp enabled, time stamp will be added to second pass packets
from CPT. NPC needs different configuration to parse second pass packets
with and without timestamp.
New PKIND is defined for CPT when time stamp is enabled on NIX.
CPT should use this PKIND for second pass packets when TS is enabled for
corresponding ethdev port.
Signed-off-by: Vidya Sagar Velumuri <vvelumuri@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
The SDP interfaces also need to be configured for NIX receive channel
backpressure for packet receive.
Signed-off-by: Radha Mohan Chintakuntla <radhac@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Currently only base channel number is configured as default
channel for all the SDP send queues. Due to this, packets
sent on different SQ's are landing on the same output queue
on the host. Channel number in the send queue should be
configured according to the number of queues assigned to the
SDP PF or VF device.
Signed-off-by: Subrahmanyam Nilla <snilla@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Support matching IPv6 fragment extension header
with RTE_FLOW_ITEM_TYPE_IPV6_FRAG_EXT flow pattern item.
Signed-off-by: Satheesh Paul <psatheesh@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Adding cnxk device driver support to configure custom SA index.
Custom SA index can be configured as part of the session create
as SPI, and later original SPI can be updated using session update.
Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
Acked-by: Ray Kinsella <mdr@ashroe.eu>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Adding devargs support to parse custom SA action.
Devargs can be specified in the following way.
-a 0002:02:00.0,custom_sa_act=1
Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Adding ROC Flow changes to parse custom SA action for cnxk device.
When custom sa action is enabled, VTAG actions are not allowed.
And custom SA index will be calculated based on SA_HI and SA_LO
values. This allows the potential for a MCAM entry to match
many SAs, rather than only match a single SA.
Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Set the error message with rte_flow_error_set() API
when flow parsing fails.
Fixes: 8c009b4505 ("net/cnxk: support flow API")
Cc: stable@dpdk.org
Signed-off-by: Satheesh Paul <psatheesh@marvell.com>
Reviewed-by: Kiran Kumar K <kirankumark@marvell.com>
Used errata APIs roc_errata_nix_has_no_drop_re() and
roc_errata_nix_has_cq_min_size_4k() instead of direct
ROC model check.
Signed-off-by: Rahul Bhansali <rbhansali@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Created roc_errata.h to list the errata handled in userspace drivers.
Added no_drop_re, cq_min_size_4k, no_fc_stype_ststp, no_drop_aging,
no_vwqe_flush_op etc erratas.
Signed-off-by: Rahul Bhansali <rbhansali@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
hot_plug application does not perform any port setup
configuration via rte_eth_dev_configure() API. All the probed
Ethernet ports do not contain any Rx and Tx queues.
While detaching a device via rte_eal_hotplug_remove(), CNXK
driver expects Rx and Tx queues structures populated during
reset of PFC. So application gets crashed as data->rx_queues
and data->tx_queues are NULL.
Fixes: 9544713564 ("net/cnxk: support priority flow control")
Cc: stable@dpdk.org
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Inline device could be null in cases when it was not bound or missing.
Added check to prevent null pointer access.
Fixes: fe5846bcc0 ("net/cnxk: add devargs for min-max SPI")
Cc: stable@dpdk.org
Signed-off-by: Volodymyr Fialko <vfialko@marvell.com>
ROC code is assuming presence of vlan extension headers in
case of QinQ, because of this, there is incompatibility
between the driver and ROC. Fixed this in ROC by treating
QINQ as multiple VLAN pattern items for DPDK (as opposed to
treating QINQ as separate pattern item).
Fixes: b8ac8b089c ("common/cnxk: support matching VLAN existence")
Cc: stable@dpdk.org
Signed-off-by: Satheesh Paul <psatheesh@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Corrected some names to follow ROC naming convention in ROC NPC code.
Signed-off-by: Satheesh Paul <psatheesh@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Added support for RTE_FLOW_ITEM_TYPE_MARK. This item type
can be used to create ingress flow rules to match packets
from CPT's second pass packets.
Signed-off-by: Satheesh Paul <psatheesh@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Added support to create flow rules to match packets
from CPT's second pass packets. With this change, ingress
rules will be created with bits 10 and 11 of channel field
in the MCAM ignored by default. For rules specific to
second pass packets, the CPT channel bits will be set
in the MCAM.
Signed-off-by: Satheesh Paul <psatheesh@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Whether the RSS is enabled depends on RTE_ETH_MQ_RX_RSS_FLAG and packet
tuple are enabled. So the RSS switch is unnecessary.
Fixes: 5e782bc257 ("net/hns3: fix configuring RSS hash when rules are flushed")
Fixes: fd81968387 ("net/hns3: fix configuring device with RSS enabled")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
The 'rss_tuple_fields' in struct struct hns3_rss_conf::rss_tuple_sets is
redundant. Because the enabled RSS tuple in PMD is already managed by
the 'types' in struct hns3_rss_conf::conf. This patch removes this
redundant variable.
Fixes: c37ca66f2b ("net/hns3: support RSS")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
The RSS tuple isn't restored when RSS key length is invalid or setting
algo key failed. This patch fixes it.
Fixes: c37ca66f2b ("net/hns3: support RSS")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
Currently, hns3 PMD disable RSS by resetting redirection table when user
set rss_hf to 0 so as to all packets go to queue 0. The implementation
may cause following problems:
1) the same type packet may go to different queue on the case of
disabling all tuples and partial tuples. The problem is determined by
hardware design.
2) affect the configuration of redirection table and user experience.
For hns3 hardware, the packets with RSS disabled are always go to the
queue corresponding to first entry of the redirection table. Generally,
disable RSS should be implemented by disabling all tuples, This patch
fix the implementation.
Fixes: c37ca66f2b ("net/hns3: support RSS")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
Currently, the hns3 PMD may free more mbufs than free_cnt parameter,
this is an incorrect implementation. This patch fixes it.
Fixes: 0b77e8f3d3 ("net/hns3: optimize Tx performance")
Cc: stable@dpdk.org
Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
Some fields in the end of 'struct hns3_rx_queue' and
'struct hns3_tx_queue' are not accessed in the I/O path.
But these fields may be accessed in other threads, which may lead to the
problem of cache pseudo-sharing of IO threads. This patch add a
cacheline alignment to avoid it.
Fixes: 9261fd3caf ("net/hns3: improve IO path data cache usage")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
The MAC and queues statistics are 32-bit registers in hardware. If
hardware statistics are not obtained for a long time, these statistics
will be overflow.
So PF and VF driver have to periodically obtain and save these
statistics. Since the periodical task and the stats API are in different
threads, we introduce a statistics lock to protect the statistics.
Fixes: 8839c5e202 ("net/hns3: support device stats")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
Clearing imissed registers in PF hardware depends on the
'drop_stats_mode' in struct hns3_hw. The variable is initialized after
the "hns3_get_configuration". But, in current code, the clearing
operation runs before the function.
So this patch fixes this order. In addition, this patch extracts a
public function to initialize and uninitialize statistics to improve the
maintainability of these codes.
Fixes: 3e9f3042d7 ("net/hns3: add imissed packet stats")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
rte_pmd_tun/tap_probe() allocates pmd->intr_handle in eth_dev_tap_create()
and it should not be freed until rte_pmd_tap_remove() is called.
Inspection of tap_rx_intr_vec_set() shows that the call to
tap_tx_intr_vec_uninstall() was calling rte_intr_instance_free() but
tap_tx_intr_vec_install() can then be immediately called, and this then
uses pmd->intr_handle without it being reallocated.
Move rte_intr_instance_free() call from tap_tx_intr_vec_uninstall()
to rte_pmd_tap_remove().
Fixes: d61138d4f0 ("drivers: remove direct access to interrupt handle")
Cc: stable@dpdk.org
Signed-off-by: Quentin Armitage <quentin@armitage.org.uk>
Reviewed-by: David Marchand <david.marchand@redhat.com>
All slaves will be stopped and removed when closing a bonded port.
But the while loop can not end if both rte_eth_dev_stop and
rte_eth_bond_slave_remove fails, runs infinitely.
This is because the skipped slave port counted in both function failures
but it should be counted only one.
Fixing by not continue to process in the loop after first failure.
Fixes: fb0379bc5d ("net/bonding: check stop call status")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
When stopping a bonded port, all slaves should be stopped. But only
active slaves are stopped.
So fix by stopping all slave ports and later do "deactivate_slave()" for
active slaves.
Fixes: 0911d4ec01 ("net/bonding: fix crash when stopping mode 4 port")
Cc: stable@dpdk.org
Signed-off-by: Huisong Li <lihuisong@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
Implement support for the power management API by implementing a
'get_monitor_addr' function that will return an address of an RX ring's
status bit.
Signed-off-by: Kevin Liu <kevinx.liu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Support configuration of unicast and multicast promisc on dcf.
Signed-off-by: Alvin Zhang <alvinx.zhang@intel.com>
Signed-off-by: Kevin Liu <kevinx.liu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Add API "dev_supported_ptypes_get" to DCF, that DCF pmd can get
ptypes through the new API.
Signed-off-by: Jie Wang <jie1x.wang@intel.com>
Signed-off-by: Kevin Liu <kevinx.liu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
add API "mtu_set" to dcf, and it can configure the port mtu through
cmdline.
Signed-off-by: Jie Wang <jie1x.wang@intel.com>
Signed-off-by: Kevin Liu <kevinx.liu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Add support for ops rte_eth_tx_done_cleanup in dcf
Signed-off-by: Robin Zhang <robinx.zhang@intel.com>
Signed-off-by: Kevin Liu <kevinx.liu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
RSS HASH should be updated and queried by application,
Add related ops ('.rss_hash_update', '.rss_hash_conf_get') for DCF.
Because DCF doesn't support configure RSS HASH, only HASH key can be
updated within ops '.rss_hash_update'.
Signed-off-by: Steve Yang <stevex.yang@intel.com>
Signed-off-by: Kevin Liu <kevinx.liu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
RSS RETA should be updated and queried by application,
Add related ops ('.reta_update', '.reta_query') for DCF.
Signed-off-by: Steve Yang <stevex.yang@intel.com>
Signed-off-by: Kevin Liu <kevinx.liu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
In this patch, We use CPU ticks instead of HW register
to determine whether low 32 bits timestamp has turned
over. It can avoid requesting register value frequently
and improve receiving performance.
Signed-off-by: Wenjun Wu <wenjun1.wu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
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>
Add new ops and structures to support VF to support Rx timestamp
on flex descriptor.
"VIRTCHNL_OP_1588_PTP_GET_CAPS" ops is sent by the VF to request PTP
capabilities and responded by the PF with capabilities enabled for
that VF.
"VIRTCHNL_OP_1588_PTP_GET_TIME" ops is sent by the VF to request
the current time of the PHC. The PF will respond by reading the
device time and reporting it back to the VF.
Signed-off-by: Simei Su <simei.su@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Remove the limitation that max queue pair number must be 2^n.
With this patch, even on a 8 ports device, the max queue pair
number increased from 128 to 254.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Wenjun Wu <wenjun1.wu@intel.com>
The 'max_rx_pktlen' value was previously read from hardware, which was
set by the running firmware. This caused confusion due to different
meanings of 'MAX_MTU'. This patch updates the 'max_rx_pktlen' to the
maximum value that the NFP NIC can support. The 'max_mtu' value that is
read from hardware, is assigned to the 'dev_info->max_mtu' variable.
If more layer 2 metadata must be used, the firmware can be updated to
report a smaller 'max_mtu' value.
The constant defined for NFP_FRAME_SIZE_MAX is derived for the maximum
supported buffer size of 10240, minus 136 bytes that is reserved by the
hardware and another 56 bytes reserved for expansion in firmware. This
results in a usable maximum packet length of 10048 bytes.
Signed-off-by: Walter Heymans <walter.heymans@corigine.com>
Signed-off-by: Niklas Söderlund <niklas.soderlund@corigine.com>
Reviewed-by: Louis Peens <louis.peens@corigine.com>
Reviewed-by: Chaoyong He <chaoyong.he@corigine.com>
Reviewed-by: Richard Donkin <richard.donkin@corigine.com>
In order to speed-up the device suspend and resume, make the statistics
counters persistent in reconfiguration until the device gets removed.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
This patch supports device cleanup callback API which is called when
the device is disconnected from the VM. Cached resources like VM MR and
VQ memory are released.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
During device suspend and resume, resources are not changed normally.
When huge resources were allocated to VM, like huge memory size or lots
of queues, time spent on release and recreate became significant.
To speed up, this patch reuses resources like VM MR and VirtQ memory if
not changed.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
To speed up device resume, create reuseable resources during device
probe state, release when device is removed. Reused resources includes
TIS,
TD, VAR Doorbell mmap, error handling event channel and interrupt
handler, UAR, Rx event channel, NULL MR, steer domain and table.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
When Qemu suspends a VM, HW notifier is un-mmapped while vCPU thread may
still be active and write notifier through kick socket.
PMD kick handler thread tries to install HW notifier through client
socket. In such case, it will timeout and slow down device close.
This patch skips HW notifier install if VQ or device in middle of
shutdown.
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
In Ctrl+C handling, sometimes kick handling thread gets endless EGAIN
error and fall into dead lock.
Kick happens frequently in real system due to busy traffic or retry
mechanism. This patch simplifies kick firmware anyway and skip setting
hardware notifier due to potential device error, notifier could be set
in next successful kick request.
Fixes: 62c813706e ("vdpa/mlx5: map doorbell")
Cc: stable@dpdk.org
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Move message handler description and callbacks into a single array and
remove unneeded VHOST_USER_MAX and VHOST_SLAVE_MAX enums.
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
By default, TSO feature should be disabled because it requires
application's support to be functional as mentioned in the
documentation.
However, if "tso" devarg was not specified, the feature did
not get disabled.
This patch fixes this issue, so that TSO is disabled, even if
"tso=0" is not passed as devarg.
Fixes: e289400669 ("net/vhost: support TSO disabling")
Cc: stable@dpdk.org
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Those optimisations were only enabled with make builds, fix the meson
part.
Fixes: 77d66da838 ("net/virtio: add vectorized packed ring Rx")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
This patch fixes heap-use-after-free reported by ASan.
It is possible for the rte_vhost_dequeue_burst() to access the vq
is freed when numa_realloc() gets called in the device running state.
The control plane will set the vq->access_lock to protected the vq
from the data plane. Unfortunately the lock will fail at the moment
the vq is freed, allowing the rte_vhost_dequeue_burst() to access
the fields of the vq, which will trigger a heap-use-after-free error.
In the case of multiple queues, the vhost pmd can access other queues
that are not ready when the first queue is ready, which makes no sense
and also allows numa_realloc() and rte_vhost_dequeue_burst() access to
vq to happen at the same time. By controlling vq->allow_queuing we can make
the pmd access only the queues that are ready.
Fixes: 1ce3c7fe14 ("net/vhost: emulate device start/stop behavior")
Signed-off-by: Yuan Wang <yuanx.wang@intel.com>
Tested-by: Wei Ling <weix.ling@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
In patch [1], NAPI has been supported in kernel tun driver to accelerate
packet processing received from vhost-net. This will greatly improve the
throughput of the tap device in the vhost-net backend.
[1]: https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next.git/commit/?id=fb3f903769e8
Signed-off-by: Harold Huang <baymaxhuang@gmail.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Match the closest supported Rx payload buffer size with the mempool
data size and program it for the Rx queue. This removes unnecessary
need for handling additional padding, packing, and alignment, when
posting Rx buffers to hardware.
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
When trying to coalesce mbufs with chain on Tx side, it is possible
to get stuck during queue wrap around. When coalescing this mbuf
chain fails, the Tx path returns EBUSY and when the same packet
is retried again, it couldn't get coalesced again, and the loop
repeats. Fix by pushing the packet through the normal Tx path.
Also use FW_ETH_TX_PKTS_WR to handle mbufs with chain for FW
to optimize.
Fixes: 6c2809628c ("net/cxgbe: improve latency for slow traffic")
Cc: stable@dpdk.org
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Fill the correct DPDK ethdev port_id, instead of local adapter
physical port_id in mbufs allocated for Rx.
Fixes: 78fc1a716a ("cxgbe: improve Rx performance")
Cc: stable@dpdk.org
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Rx/Tx queue info dump and pvid info dump is both supported
in PF and VF. This patch fixed it.
Fixes: 1a03c659cb ("net/hns3: dump device basic info")
Cc: stable@dpdk.org
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
Shouldn't dump TM info when TM is not supported by the NIC.
This patch fixed it.
Fixes: e4cfe6bb91 ("net/hns3: dump TM configuration info")
Cc: stable@dpdk.org
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
This patch refactors queue info dump.
Fixes: 6038c8a3f6 ("net/hns3: dump queue info")
Cc: stable@dpdk.org
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
This patch refactors adapter state dump.
Fixes: 1a03c659cb ("net/hns3: dump device basic info")
Cc: stable@dpdk.org
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
When creating a bonding device, if the slave device's
RSS key length = standard_rss_key length + extended_hash_key length,
then bonding device will be same as slave,
in function bond_ethdev_configure(), the default_rss_key length is 40,
it is not matched, so it should calculate a new key for bonding device
if the default key could not be used.
Fixes: 6b1a001ec5 ("net/bonding: fix RSS key length")
Cc: stable@dpdk.org
Signed-off-by: Ke Zhang <ke1x.zhang@intel.com>
Acked-by: Min Hu (Connor) <humin29@huawei.com>
