net/mlx5: fix polling CQEs

In certain situations, the receive queue (rxq) fails to replenish its
internal ring with memory buffers (mbufs) from the pool. This can happen
when the pool has a limited

net/mlx5: fix polling CQEs

In certain situations, the receive queue (rxq) fails to replenish its
internal ring with memory buffers (mbufs) from the pool. This can happen
when the pool has a limited number of mbufs allocated, and the user
application holds incoming packets for an extended period, resulting in a
delayed release of mbufs. Consequently, the pool becomes depleted,
preventing the rxq from replenishing from it.

There was a bug in the behavior of the vectorized rxq_cq_process_v routine,
which handled completion queue entries (CQEs) in batches of four. This
routine consistently accessed four mbufs from the internal queue ring,
regardless of whether they had been replenished. As a result, it could
access mbufs that no longer belonged to the poll mode driver (PMD).

The fix involves checking if there are four replenished mbufs available
before allowing rxq_cq_process_v to handle the batch. Once replenishment
succeeds during the polling process, the routine will resume its operation.

Fixes: 1ded26239aa0 ("net/mlx5: refactor vectorized Rx")
Cc: stable@dpdk.org

Reported-by: Changqi Dingluo <dingluochangqi.ck@bytedance.com>
Signed-off-by: Gavin Hu <gahu@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>

show more ...

net/ngbe: add probe and remove for VF device

Introduce virtual function driver in ngbe PMD, add simple init and
uninit function to probe and remove the device.

Signed-off-by: Zaiyu Wang <zaiyuwang@

net/ngbe: add probe and remove for VF device

Introduce virtual function driver in ngbe PMD, add simple init and
uninit function to probe and remove the device.

Signed-off-by: Zaiyu Wang <zaiyuwang@trustnetic.com>

show more ...

examples/flow_filtering: remove duplicate assignment

Last action in template was being assigned twice.

Coverity issue: 451221
Fixes: 16158f349000 ("examples/flow_filtering: introduce use cases snip

examples/flow_filtering: remove duplicate assignment

Last action in template was being assigned twice.

Coverity issue: 451221
Fixes: 16158f349000 ("examples/flow_filtering: introduce use cases snippets")
Cc: stable@dpdk.org

Signed-off-by: Ariel Otilibili <otilibil@eurecom.fr>

show more ...

net/ena: fix missing default LLQ policy

The driver failed to set a default llq_header_policy,
causing LLQ to be disabled when the devarg value is
not specified by the application.

Fixes: d7918d19d2

net/ena: fix missing default LLQ policy

The driver failed to set a default llq_header_policy,
causing LLQ to be disabled when the devarg value is
not specified by the application.

Fixes: d7918d19d25e ("net/ena: restructure LLQ policy user setting")
Cc: stable@dpdk.org

Signed-off-by: Yosef Raisman <yraisman@amazon.com>
Signed-off-by: Shai Brandes <shaibran@amazon.com>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>

show more ...

eal/x86: fix some intrinsics header include for Windows

Error reported:
../lib/net/net_crc_sse.c:49:17: error: call to undeclared function
'_mm_clmulepi64_si128'; ISO C99 and later do not support im

eal/x86: fix some intrinsics header include for Windows

Error reported:
../lib/net/net_crc_sse.c:49:17: error: call to undeclared function
'_mm_clmulepi64_si128'; ISO C99 and later do not support implicit
function declarations [-Wimplicit-function-declaration]

The fix is to remove the unnecessary ifdef around the inclusion of
header file immintrin.h. This header also contains functions that do
not require AVX instructions, so should not be included only when AVX
is available.

Bugzilla ID: 1595
Fixes: da826b7135a4 ("eal: introduce ymm type for AVX 256-bit")
Cc: stable@dpdk.org

Reported-by: Pier Damouny <pdamouny@nvidia.com>
Signed-off-by: Andre Muezerie <andremue@linux.microsoft.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>

show more ...

vhost: clear ring addresses when getting vring base

Clear ring addresses during vring base retrieval to handle guest reboot
scenarios correctly. This is particularly important for vdpa-blk devices
w

vhost: clear ring addresses when getting vring base

Clear ring addresses during vring base retrieval to handle guest reboot
scenarios correctly. This is particularly important for vdpa-blk devices
where the following issue occurs:

When a guest OS with vdpa-blk device reboots, during UEFI stage, only
one vring is actually used and configured. However, QEMU still sends
enable messages for all configured queues. The remaining queues retain
their addresses from before reboot, which reference invalid memory
mappings in the rebooted guest.

The issue manifests in vq_is_ready():

static bool
vq_is_ready(struct virtio_net *dev, struct vhost_virtqueue *vq)
{
/* Only checks pointer validity, not address freshness */
rings_ok = vq->desc && vq->avail && vq->used;
...
}

vq_is_ready() incorrectly considers these queues as ready because it
only checks if desc/avail/used pointers are non-NULL, but cannot
detect that these addresses are stale from the previous boot.

Clear the ring addresses in vhost_user_get_vring_base() to force
the guest driver to reconfigure them before use. This ensures that
vq_is_ready() will return false for queues with stale addresses
until they are properly reconfigured by the guest driver.

Fixes: 3ea7052f4b1b ("vhost: postpone rings addresses translation")
Cc: stable@dpdk.org

Signed-off-by: Jianping Zhao <zhao305149619@gmail.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>

show more ...

ring: make dump function more verbose

The current rte_ring_dump function uses the generic rte_ring_headtail
structure to access head/tail positions. This is incorrect for the RTS
case where the head

ring: make dump function more verbose

The current rte_ring_dump function uses the generic rte_ring_headtail
structure to access head/tail positions. This is incorrect for the RTS
case where the head is stored in a different offset in the union of
structs. Switching to a separate function for each sync type allows
to dump correct head/tail values and extra metadata.

Signed-off-by: Eimear Morrissey <eimear.morrissey@huawei.com>
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@huawei.com>
Acked-by: Morten Brørup <mb@smartsharesystems.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>

show more ...

doc: reword some sample app guides

I have reviewed these sections for grammar/clarity
and made small modifications to the formatting of sections
to adhere to a template which will create uniformalit

doc: reword some sample app guides

I have reviewed these sections for grammar/clarity
and made small modifications to the formatting of sections
to adhere to a template which will create uniformality
in the sample application user guides overall.

Signed-off-by: Nandini Persad <nandinipersad361@gmail.com>
Acked-by: Chengwen Feng <fengchengwen@huawei.com>

show more ...

dts: add l2fwd test suite

Add a basic L2 forwarding test suite which tests the correct
functionality of the forwarding facility built-in in the DPDK.

The tests are performed with different queues n

dts: add l2fwd test suite

Add a basic L2 forwarding test suite which tests the correct
functionality of the forwarding facility built-in in the DPDK.

The tests are performed with different queues numbers per port.

Signed-off-by: Luca Vizzarro <luca.vizzarro@arm.com>
Signed-off-by: Thomas Wilks <thomas.wilks@arm.com>
Reviewed-by: Paul Szczepanek <paul.szczepanek@arm.com>
Reviewed-by: Dean Marx <dmarx@iol.unh.edu>
Reviewed-by: Patrick Robb <probb@iol.unh.edu>
Tested-by: Patrick Robb <probb@iol.unh.edu>

show more ...

dts: rename build target to DPDK build

Since the DPDK may already be built, some more general name
is needed that includes both the DPDK location and the build
config (if we are going to build).

Si

dts: rename build target to DPDK build

Since the DPDK may already be built, some more general name
is needed that includes both the DPDK location and the build
config (if we are going to build).

Signed-off-by: Tomáš Ďurovec <tomas.durovec@pantheon.tech>
Signed-off-by: Luca Vizzarro <luca.vizzarro@arm.com>

show more ...

raw/gdtc: introduce driver skeleton

Introduce rawdev driver support for GDTC which
can help to connect two separate hosts with each other.

Signed-off-by: Yong Zhang <zhang.yong25@zte.com.cn>

usertools/devbind: support all VFIO no-IOMMU sysfs values

This patch adds support for 'y' and 'Y' values when reading
vfio-pci unsafe_noiommu_mode flag.

Possible values were taken from linux kernel

usertools/devbind: support all VFIO no-IOMMU sysfs values

This patch adds support for 'y' and 'Y' values when reading
vfio-pci unsafe_noiommu_mode flag.

Possible values were taken from linux kernel
(sysfs__read_bool() in tools/lib/api/fs/fs.c)

Signed-off-by: Rogelio Domínguez Hernández <rogelio.dominguez@gmail.com>
Reviewed-by: Robin Jarry <rjarry@redhat.com>

show more ...

doc: modernize build instructions on Windows

Meson 0.57 was an unstable version and is now outdated.
The referenced bug in Meson 0.58 is fixed in stable releases.
Recommend the first version contain

doc: modernize build instructions on Windows

Meson 0.57 was an unstable version and is now outdated.
The referenced bug in Meson 0.58 is fixed in stable releases.
Recommend the first version containing the fix: 1.5.2.

Building DPDK applications that run on 32-bit Windows is
currently not supported. However, some Visual Studio environments
default to producing 32-bit binaries.
Recommend instructing the developer prompt to produce 64-bit binaries
when that is the case.

Signed-off-by: Andre Muezerie <andremue@linux.microsoft.com>
Acked-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>

show more ...

net/r8169: add driver skeleton

Meson build infrastructure, r8169_ethdev minimal skeleton,
header with Realtek NIC device and vendor IDs.

Signed-off-by: Howard Wang <howard_wang@realsil.com.cn>
Revi

net/r8169: add driver skeleton

Meson build infrastructure, r8169_ethdev minimal skeleton,
header with Realtek NIC device and vendor IDs.

Signed-off-by: Howard Wang <howard_wang@realsil.com.cn>
Reviewed-by: Ferruh Yigit <ferruh.yigit@amd.com>

show more ...

net/bnxt: fix LRO offload capability

Fix LRO offload capability for P7 devices.
Export the capability to the application only if compressed
Rx CQE mode is not enabled.

LRO aka TPA is not supported

net/bnxt: fix LRO offload capability

Fix LRO offload capability for P7 devices.
Export the capability to the application only if compressed
Rx CQE mode is not enabled.

LRO aka TPA is not supported when compressed CQE mode is set.

Fixes: 3b56c3ffc182 ("net/bnxt: refactor code to support P7 devices")
Cc: stable@dpdk.org

Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Vasuthevan Maheswaran <vasuthevan.maheswaran@broadcom.com>

show more ...

net/bnxt: fix reading SFF-8436 SFP EEPROMs

If a SFP which supports SFF-8436 is present then
currently the DDM information present in the eeprom
is not read. Furthermore bnxt_get_module_eeprom()
will

net/bnxt: fix reading SFF-8436 SFP EEPROMs

If a SFP which supports SFF-8436 is present then
currently the DDM information present in the eeprom
is not read. Furthermore bnxt_get_module_eeprom()
will return -EINVAL for these eeproms since the
length of these eeproms is 512 bytes but we are
only ever selecting 2 pages (256 bytes) to read.

Fixes: 6253a23491a4 ("net/bnxt: dump SFP module info")
Cc: stable@dpdk.org

Signed-off-by: Peter Morrow <peter@graphiant.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>

show more ...

net/bnxt/tf_core: fix slice count in case of HA entry move

When entries are moved during HA, a shared move function transfers
TCAM entries by using get/set message APIs, and the slice number of the

net/bnxt/tf_core: fix slice count in case of HA entry move

When entries are moved during HA, a shared move function transfers
TCAM entries by using get/set message APIs, and the slice number of the
entry is required to accomplish the movement. The slice number is
calculated as the product of row_slice and entry size. Before calling
get/set message APIs, the source entry size should be updated with the
destination entry size; otherwise, it might corrupt the slice number field,
which may result in writing an incorrect entry. A fix is made which now
copies the entry size from the source to the destination before calling
get/set message APIs, ensuring the correct slice number is modified.

Fixes: 97435d7906d7 ("net/bnxt: update Truflow core")
Cc: stable@dpdk.org

Signed-off-by: Sangtani Parag Satishbhai <parag-satishbhai.sangtani@broadcom.com>
Reviewed-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>

show more ...

net/bonding: promote experimental API stable

Remove rte_experimental macros from the stable functions
as they have been part of the stable API since 23.11.

Signed-off-by: Lukas Sismis <sismis@cesne

net/bonding: promote experimental API stable

Remove rte_experimental macros from the stable functions
as they have been part of the stable API since 23.11.

Signed-off-by: Lukas Sismis <sismis@cesnet.cz>
Acked-by: Huisong Li <lihuisong@huawei.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>

show more ...

net/zxdh: add driver skeleton

Add basic zxdh ethdev init and register PCI probe functions.
Update doc files.

Signed-off-by: Junlong Wang <wang.junlong1@zte.com.cn>
Acked-by: Ferruh Yigit <ferruh.yi

net/zxdh: add driver skeleton

Add basic zxdh ethdev init and register PCI probe functions.
Update doc files.

Signed-off-by: Junlong Wang <wang.junlong1@zte.com.cn>
Acked-by: Ferruh Yigit <ferruh.yigit@amd.com>

show more ...

examples/l3fwd: fix read beyond boundaries

ASAN report:
ERROR: AddressSanitizer: unknown-crash on address 0x7ffffef92e32 at pc 0x00000053d1e9 bp 0x7ffffef92c00 sp 0x7ffffef92bf8
READ of size 16 at 0

examples/l3fwd: fix read beyond boundaries

ASAN report:
ERROR: AddressSanitizer: unknown-crash on address 0x7ffffef92e32 at pc 0x00000053d1e9 bp 0x7ffffef92c00 sp 0x7ffffef92bf8
READ of size 16 at 0x7ffffef92e32 thread T0
#0 0x53d1e8 in _mm_loadu_si128 /usr/lib64/gcc/x86_64-suse-linux/11/include/emmintrin.h:703
#1 0x53d1e8 in send_packets_multi ../examples/l3fwd/l3fwd_sse.h:125
#2 0x53d1e8 in acl_send_packets ../examples/l3fwd/l3fwd_acl.c:1048
#3 0x53ec18 in acl_main_loop ../examples/l3fwd/l3fwd_acl.c:1127
#4 0x12151eb in rte_eal_mp_remote_launch ../lib/eal/common/eal_common_launch.c:83
#5 0x5bf2df in main ../examples/l3fwd/main.c:1647
#6 0x7f6d42a0d2bc in __libc_start_main (/lib64/libc.so.6+0x352bc)
#7 0x527499 in _start (/home/kananyev/dpdk-l3fwd-acl/x86_64-native-linuxapp-gcc-dbg-b1/examples/dpdk-l3fwd+0x527499)

Reason for that is that send_packets_multi() uses 16B loads to access
input dst_port[]and might read beyond array boundaries.
Right now, it doesn't cause any real issue - junk values are ignored, also
inside l3fwd we always allocate dst_port[] array on the stack, so
memory beyond it is always available.
Anyway, it probably need to be fixed.
The patch below simply allocates extra space for dst_port[], so
send_packets_multi() will never read beyond its boundaries.

Probably a better fix would be to change send_packets_multi()
itself to avoid access beyond 'nb_rx' entries.

Bugzilla ID: 1502
Fixes: 94c54b4158d5 ("examples/l3fwd: rework exact-match")
Cc: stable@dpdk.org

Signed-off-by: Konstantin Ananyev <konstantin.ananyev@huawei.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>

show more ...

doc: add lcore variables guide

Add lcore variables programmer's guide. This guide gives both an
overview of the API, its implementation, and alternatives to the use
of lcore variables for maintainin

doc: add lcore variables guide

Add lcore variables programmer's guide. This guide gives both an
overview of the API, its implementation, and alternatives to the use
of lcore variables for maintaining per-lcore id data.

It has pictures, too.

Signed-off-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
Reviewed-by: Luka Jankovic <luka.jankovic@ericsson.com>

show more ...

net/mlx5: fix real time counter reading from PCI BAR

There is the mlx5_txpp_read_clock() routine reading
the 64-bit real time counter from the device PCI BAR.
It introduced two issues:

- it check

net/mlx5: fix real time counter reading from PCI BAR

There is the mlx5_txpp_read_clock() routine reading
the 64-bit real time counter from the device PCI BAR.
It introduced two issues:

- it checks the PCI BAR mapping into process address
space and tries to map this on demand. This might be
problematic if something goes wrong and mapping fails.
It happens on every read_clock API call, invokes kernel
taking a long time and causing application malfunction.

- the 64-bit counter should be read in single atomic
transaction

Fixes: 9b31fc9007f9 ("net/mlx5: fix read device clock in real time mode")
Cc: stable@dpdk.org

Signed-off-by: Tim Martin <timothym@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>

show more ...

net/af_packet: support link status update

For net_af_packet PMD, eth_link_update was an empty function.
Application may be interested in link up/down status.
This patch adds implementation that upda

net/af_packet: support link status update

For net_af_packet PMD, eth_link_update was an empty function.
Application may be interested in link up/down status.
This patch adds implementation that updates dev_link->link_status based
on socket IFF_RUNNING flag.

Signed-off-by: Gur Stavi <gur.stavi@huawei.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>

show more ...

net/nfp: add more failure logs

Add more logs to debug probe process, and modify some log level.

Signed-off-by: Huaxing Zhu <huaxing.zhu@corigine.com>
Reviewed-by: Long Wu <long.wu@corigine.com>
Rev

net/nfp: add more failure logs

Add more logs to debug probe process, and modify some log level.

Signed-off-by: Huaxing Zhu <huaxing.zhu@corigine.com>
Reviewed-by: Long Wu <long.wu@corigine.com>
Reviewed-by: Peng Zhang <peng.zhang@corigine.com>
Reviewed-by: Chaoyong He <chaoyong.he@corigine.com>

show more ...

net/pcap: fix blocking Rx

Use pcap_next_ex rather than just pcap_next because pcap_next
always blocks if there is no packets to receive.

Bugzilla ID: 1526
Fixes: 4c173302c307 ("pcap: add new driver

net/pcap: fix blocking Rx

Use pcap_next_ex rather than just pcap_next because pcap_next
always blocks if there is no packets to receive.

Bugzilla ID: 1526
Fixes: 4c173302c307 ("pcap: add new driver")
Cc: stable@dpdk.org

Reported-by: Ofer Dagan <ofer.d@claroty.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Tested-by: Ofer Dagan <ofer.d@claroty.com>

show more ...