The TAP keep-alive queue was created in order to keep the TAP device
in Linux even in case all of its Rx/Tx queues are released (in Linux
terminology: even in case all of the TAP device file descriptors are
closed), however, the keep-alive queue itself is attached to the TAP
device like all other Rx/Tx queues and therefore the kernel will
enqueue to it some Rx packets based on the kernel RSS distribution
rules. Those packets are unknown to the application and will remain
lost in the keep-alive queue.
All queues are attached by default to the TAP device after they are
created though TUNSETIFF ioctl call.
The fix is to detach the keep-alive queue after its creation through
TUNSETQUEUE ioctl call.
Fixes: 3101191c63 ("net/tap: fix device removal when no queue exist")
Cc: stable@dpdk.org
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>
In 100G mode, we poll firmware slow path completion for every 1 second,
which is not enough and may result in completion timeout if
driver misses that window.
Patch "eal: set affinity for control threads" exposed this issue since
alarm callback runs in control thread context.
Fix this issue by update polling period to 100ms.
Fixes: d651ee4919 ("eal: set affinity for control threads")
Fixes: 2af14ca79c ("net/qede: support 100G")
Cc: stable@dpdk.org
Signed-off-by: Shahed Shaikh <shahed.shaikh@cavium.com>
On some distros, the inbox rdma-core tree can contain the Software
Parser enum while the remaining structs still missing.
Fixes: 5f8ba81c42 ("net/mlx5: support generic tunnel offloading")
Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
This is to fix the offloads introduced by commits
5f8ba81 net/mlx5: support generic tunnel offloading
5355f44 ethdev: introduce generic IP/UDP tunnel checksum and TSO
Fixes: 8589e944d0 ("net/mlx5: fix setting offsets for SW parser")
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
DPDK apps can be executed as non-root users but current NFP lock
file for avoiding concurrent accesses to CPP interface is precluding
this option or requires to modify system file permissions.
When the NFP device is bound to VFIO, this driver does not allow this
concurrent access, so the lock file is not required at all.
OVS-DPDK as executed in RedHat distributions is the main NFP user
needing this fix.
Fixes: c7e9729da6 ("net/nfp: support CPP")
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
For virtio-user server mode, one use case comes across segmentation fault.
step 1: Launch vhost side as client firstly.
step 2: launch virtio-user side as server.
The cause is: after registering virtio_interrupt_handler into
eal-intr-thread, two threads (main thread and eal-intr-thread) have
sync issues, so add rxvq pointer checking in function virtio_notify_peers
to decide if the code can continue.
Fixes: bd8f50a45d ("net/virtio-user: support server mode")
Cc: stable@dpdk.org
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
This was missed in a previous patch for using new offload API.
It turns out the line can be removed because that internal PMD
crc_len data is not being used.
Fixes: a922c6a789 ("net/nfp: support new HW offloads API")
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Some files were left with full license and wrong copyright format.
They are switched to this format:
SPDX-License-Identifier: BSD-3-Clause
Copyright 2017 Mellanox Technologies, Ltd
Fixes: 5feecc57d9 ("align SPDX Mellanox copyrights")
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
Enables TUN PMD sharing by attaching the port from the shared data.
Fixes: ee27edbe0c ("drivers/net: share vdev data to secondary process")
Signed-off-by: Vipin Varghese <vipin.varghese@intel.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>
The TX function is shared between TAP and TUN PMD. Checking TUN-TAP
type field will ensure the TAP PMD will always have protocol field
as 0.
Fixes: 204d026a39 ("net/tap: support tun")
Signed-off-by: Vipin Varghese <vipin.varghese@intel.com>
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Clear link state information during device start for fetching new state
before function completion.
Fixes: 605164c8e7 ("net/liquidio: add API to validate VF MTU")
Cc: stable@dpdk.org
Signed-off-by: Chuhong Yao <ych@panath.cn>
Acked-by: Shijith Thotton <shijith.thotton@caviumnetworks.com>
For frames where the hardware is not able to calculate checksum
we are indicating such frames to be bad. And that is incorrect.
Indicate PKT_RX_IP_CKSUM_UNKNOWN or PKT_RX_L4_CKSUM_UNKNOWN
for such frames.
Fixes: 7ec39d8c52 ("net/bnxt: update status of Rx IP/L4 CKSUM")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Query of port stats is permitted for VF interfaces as well.
Fix the code to allow VFs to query port stats.
Fixes: bfb9c2260b ("net/bnxt: support xstats get/reset")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
If the resource requirement of a VF is more than what is allotted
to it by the FW, a VF can request to reallocate some of the resources.
This is possible only when the NIC is running the Resource Manager aware
CHiMP FW.
If necessary, calculate Tx and Rx ring count using the new RM API.
Otherwise use the Tx and Rx ring count as-is. Update the cp ring count
based on the Tx and Rx ring count.
Fixes: b7778e8a1c ("net/bnxt: refactor to properly allocate resources for PF/VF")
Signed-off-by: Jay Ding <jay.ding@broadcom.com>
Signed-off-by: Scott Branden <scott.branden@broadcom.com>
Reviewed-by: Ray Jui <ray.jui@broadcom.com>
Reviewed-by: Randy Schacher <stuart.schacher@broadcom.com>
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
In order to save completion resource, use the first completion ring from
PF or VF as the default completion ring for async event & HWRM forward
response handling. Add bnxt_hwrm_set_async_event_cr() to set
async_event_cr for either PF or VF.
Fixes: 7bc8e9a227 ("net/bnxt: support async link notification")
Signed-off-by: Qingmin Liu <qingmin.liu@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
While configuring RSS, the FW needs to know the rss_hash_mode that needs
to be initialized.
Set hash mode to HWRM_VNIC_RSS_CFG_INPUT_HASH_MODE_FLAGS_DEFAULT
to allow the FW to use the global mode already configured in the device.
Fixes: bcd4bcc442 ("net/bnxt: configure RSS")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
We were initializing Tx and Rx burst functions only for primary
processes. Fixing it in this patch.
Fixes: 8892013668 ("net/bnxt: support xstats get by id")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
The FALLTHROUGH comment for cascading switch cases was missing in
a few places. This patch adds that.
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Since ConnectX-5, SW parser just complements HW parser. SW parser starts to
engage only if HW parser can't reach a header. For the older devices, HW
parser will not kick in if any of SWP offsets is set. Therefore, all of the
L3 offsets should be set regardless of HW offload. As IPv6 doesn't have
header checksum, the mbuf can't have PKT_TX_[OUTER_]IP_CKSUM if outer or
inner L3 is IPv6.
And if inner packet isn't IP, the inner offsets shouldn't be set.
Fixes: 5f8ba81c42 ("net/mlx5: support generic tunnel offloading")
Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Xueming Li <xuemingl@mellanox.com>
Since we move to new offload APIs, IXGBE_SIMPLE_FLAGS is not used.
Fixes: 51215925a3 ("net/ixgbe: convert to new Tx offloads API")
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
When the fail-safe device is reconfigured, it attempts to register
again for the sub-devices LSC and RMV events.
Prevent an event registration if it is already done.
Fixes: 598fb8aec6 ("net/failsafe: support device removal")
Cc: stable@dpdk.org
Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Gaetan Rivet <gaetan.rivet@6wind.com>
The fail-safe PMD registers to RMV event for each removable sub-device
port in order to cleanup the sub-device resources and switch the Tx
sub-device directly when it is plugged-out.
During removal time, the fail-safe PMD stops and closes the sub-device
but it doesn't unregister the LSC and RMV callbacks of the sub-device
port.
It can lead the callbacks to be called for a port which is no more
associated with the fail-safe sub-device, because there is not a
guarantee that a sub-device gets the same port ID for each plug-in
process. This port, for example, may belong to another sub-device of a
different fail-safe device.
Unregister the LSC and RMV callbacks for sub-devices which are not
used.
Fixes: 598fb8aec6 ("net/failsafe: support device removal")
Cc: stable@dpdk.org
Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Gaetan Rivet <gaetan.rivet@6wind.com>
Since commit 97b2217ae5 ("net/mlx4: advertise supported RSS hash
functions"), this PMD includes two similar-looking functions that convert
RSS hash fields between Verbs and DPDK formats.
This patch refactors them as a single two-way function and gets rid of
redundant helper macros.
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Below commit documents 0 as a value standing for a default set of RSS hash
types, however the mlx4 PMD doesn't interpret it correctly and still uses
its own internal special value for that (-1).
Also, its function prototype was not updated.
Fixes: ac8d22de23 ("ethdev: flatten RSS configuration in flow API")
Fixes: 1d173da83e ("net/mlx4: fix default RSS hash fields")
Cc: stable@dpdk.org
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Cleanup remain dependency on old rxmode flags since we move to
new offload APIs.
Fixes: c3ac7c5b0b ("net/i40e: convert to new Rx offloads API")
Fixes: e0cb96204b ("net/i40e: add support for representor ports")
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Beilei Xing <beilei.xing@intel.com>
rte_eth_bond_8023ad_slave_info() wasn't exported in the original work.
Useful to extract information about the current 802.3ad state.
Fixes: 46fb436836 ("bond: add mode 4")
Cc: stable@dpdk.org
Signed-off-by: Chas Williams <chas3@att.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
The vdev_netvsc driver allows an automatic probe in Hyper-V VM systems
unless it was already specified by the EAL command line.
The detection of a specified NetVSC device is wrongly done by comparing
the vdev_netvsc driver name to all the vdev devices names, including
the suffix device index. Thus, if the user specifies the vdev_netvsc
device by adding an index to the device name, the comparison fails.
Consequently, the vdev_netvsc driver may automatically probe NetVSC
devices, despite the NetVSC device that was specified by the EAL command
line.
Compare the vdev_netvsc driver name to the devices names without the
index.
Fixes: 56252de779 ("net/vdev_netvsc: add automatic probing")
Cc: stable@dpdk.org
Signed-off-by: Matan Azrad <matan@mellanox.com>
Fixes: fbe90cdd77 ("ethdev: add probing finish function")
Signed-off-by: Ed Czeck <ed.czeck@atomicrules.com>
Acked-by: John Miller <john.miller@atomicrules.com>
Reviewed-by: Thomas Monjalon <thomas@monjalon.net>
Following error reported by cppcheck:
(error) Shifting signed 32-bit value by 31 bits is
undefined behaviour. The patch fixes it.
Fixes: 8db9e2a1b2 ("i40e: base driver")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
This commit fixes a compilation error if EM_PMD is
not defined, bug IGB_PMD is. The root cause of the
issue was that log init variables are declared as
extern in a header file, while the definition of the
variables was in e1000_ethdev.c. Hence, the definitions
were not available if the e1000 PMD is disabled.
To fix this, a new file is added e1000_logs.c, which
matches the e1000_logs.h header. The log variables are
always compiled in, but the PMD logs are only registered
if a PMD is enabled in the configuration. Extra checks
are added in order to avoid duplicate registering.
Fixes: ed5bbb767c ("net/e1000: implement dynamic logging")
Cc: stable@dpdk.org
Reported-by: Vipin Varghese <vipin.varghese@intel.com>
Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: Vipin Varghese <vipin.varghese@intel.com>
With the commit af4f09f282 ("net/mlx5: prefix all functions with mlx5"),
mlx5_xstats_get() is not compliant any longer with the api.
It always returns the caller max entries count while it should return how
many entries it wrote/wanted to write.
Fixes: af4f09f282 ("net/mlx5: prefix all functions with mlx5")
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
TAP device is created following its first queue creation. Multiple
queues can be added or removed over time. In Linux terminology those
are file descriptors which are opened or closed over time. As long as
the number of opened file descriptors is positive - TAP device will
appear as a Linux device. In case all queues are released (the
equivalent of all file descriptors being closed) the TAP device will
be removed. This can lead to abnormalities in different scenarios
where the TAP device should exist even if all its queues are released.
In order to make TAP existence independent of its number of queues -
an extra file descriptor is opened on TAP creation and is closed on
TAP closure. Its only purpose is to serve as a keep-alive mechanism
for the TAP device.
Fixes: bf7b7f437b ("net/tap: create netdevice during probing")
Cc: stable@dpdk.org
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Keith Wiles <keith.wiles@intel.com>
Add force_link_up devargs to always force link as up for VFs.
This enables VFs on the same NIC to send traffic to each other
even when physical link is down.
Also add RTE_PMD_REGISTER_PARAM_STRING to export all supported
devargs.
Fixes: 011ebc236d ("net/cxgbe: add skeleton VF driver")
Signed-off-by: Shagun Agrawal <shaguna@chelsio.com>
Signed-off-by: Kumar Sanghvi <kumaras@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
The resetting of stats on a VF involves the setting of an offset
that is subtracted from future calls, rather zeroing of counters.
However doing a stats reset on the port representor was also
adjusting the values forwarded to the VF, which had the effect of
corrupting the VF's counters. The fix is for the port representor
to maintain its own stats offset, so the port representor and VF
maintain independently-resettable counters.
Fixes: e0cb96204b ("net/i40e: add support for representor ports")
Signed-off-by: Remy Horton <remy.horton@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
The old ixgbe_vlan_offload_set() is called by
rte_eth_dev_set_vlan_offload() which is meant to config VLAN
strip/filter/extend on all queues.
This old function is also called by rte_eth_dev_start()/ixgbe_dev_start()
which need support per-queue VALN strip on only parts of queues.
So add new function ixgbe_vlan_offload_config() =
old ixgbe_vlan_offload_set().
This new function is called by ixgbe_dev_start() to support VLAN strip
per-queue configuration.
New ixgbe_vlan_offload_set() = codes to align VLAN strip flags on all
queues with port level setting + new ixgbe_vlan_offload_configure().
The 2nd function is called by rte_eth_dev_set_vlan_offload to support
configure VLAN strip on all queues of whole port.
Fixes: 216f78f4d5 ("net/ixgbe: support VLAN strip per queue offloading in PF")
Fixes: 860a94d3c6 ("net/ixgbe: support VLAN strip per queue offloading in VF")
Signed-off-by: Wei Dai <wei.dai@intel.com>
Signed-off-by: Yanglong Wu <yanglong.wu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
There is no need to return failure (-EEXIST) when the requested
filter is already configured.
Fixes: d6cb17535f ("net/qede: fix VLAN filters")
Fixes: 77fac1b54f ("net/qede: fix filtering code")
Cc: stable@dpdk.org
Signed-off-by: Shahed Shaikh <shahed.shaikh@cavium.com>
Reduce the verbosity in ecore_pglueb_rbc_attn_handler() while logging
errmsg during HW init. This could happen due to ungraceful exit of
previous driver instance.
Signed-off-by: Rasesh Mody <rasesh.mody@cavium.com>
This patch removes unnecessary check in ixgbe_check_mq_mode, which
fixes the DCB configuration issue.
Fixes: 27b609cbd1 ("ethdev: move the multi-queue mode check to specific drivers")
Cc: stable@dpdk.org
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
This patch adds a pair of "()" to embrace the argument
input to the function-like macro invocation.
drivers/net/mlx5/mlx5_rxtx_vec.c:37:
drivers/net/mlx5/mlx5_rxtx_vec_neon.h:170:24: error: too many arguments
provided to function-like macro invocation
(uint16x8_t) { 0, 0, cs_flags, rte_cpu_to_be_16(len),
Fixes: 570acdb1da ("net/mlx5: add vectorized Rx/Tx burst for ARM")
Cc: stable@dpdk.org
Signed-off-by: Gavin Hu <gavin.hu@arm.com>
Signed-off-by: Sirshak Das <sirshak.das@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
The value of rte_errno must be positive in case of an error.
Fixes: d77d07391d ("net/sfc: support flow API RSS action")
Cc: stable@dpdk.org
Signed-off-by: Roman Zhukov <roman.zhukov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
