Lifetime tracking is not supported by hardware and is not implemented in
software either. Return failure when lifetime is configured.
Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>
Add support for SA lifetime configuration. Expiry can
be either in units of octets or packets.
Also, updated cryptodev dequeue path to update crypto op result to
indicate soft expiry.
Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>
Adding IV in SA in lookaside IPsec debug mode. It helps
to verify lookaside PMD using known outbound vectors in
lookaside autotest.
Signed-off-by: Anoob Joseph <anoobj@marvell.com>
Signed-off-by: Tejasree Kondoj <ktejasree@marvell.com>
Acked-by: Akhil Goyal <gakhil@marvell.com>
The driver creates an indirect mkey per entry in the queue to manage the
crypto operation using the BSF fields.
The indirect mkeys were never released neither while error occurs in the
creation phase nor when the queue is released.
Clean the indirect mkeys in the above cases.
Fixes: c2a42d19d967 ("crypto/mlx5: add WQE set initialization")
Cc: stable@dpdk.org
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
The crypto QP consumer (ci) and producer (pi) indexes are increased
with each successful enqueue/dequeue operations.
However the QP pi index is calculated with a wraparound the number
of elements while the QP ci does not.
This is causing incorrect engine calculation for encqueued WQ values
(wq->pi - wq->ci) and eventually the device stops accepting new enqueue
operations.
Fixed by removing the wraparound on QP pi and using a temp calculation
where wraparound values are needed.
Fixes: 8e196c08ab53 ("crypto/mlx5: support enqueue/dequeue operations")
Cc: stable@dpdk.org
Signed-off-by: Tal Shnaiderman <talshn@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Replace pending queue with one that allows concurrent single producer and
single consumer. This relaxes the restriction of only allowing a single
lcore to operate on a given queue pair.
Signed-off-by: David George <david.george@sophos.com>
Signed-off-by: Anoob Joseph <anoobj@marvell.com>
When VM size is larger than 4G (u32) and memory region is larger than 4G,
the 32-bit GCD function overflowed and returned wrong value
that resulted in memory registration failure.
This patch calls 64-bit GCD function to avoid overflow.
Fixes: cc07a42da250 ("vdpa/mlx5: prepare memory regions")
Cc: stable@dpdk.org
Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Reviewed-by: Matan Azrad <matan@nvidia.com>
These APIs were introduced in 19.02, therefore removing
experimental tag to promote them to stable state.
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Remove experimental tag from internal API dpaa2_seqn.
This API was introduced in DPDK 20.11 and is now moved to
internal tag.
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
This API was introduced in 19.08, therefore removing
experimental tag to promote them to stable state.
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Remove experimental tag from internal API dpaa_seqn.
This API was introduced in DPDK 20.11 and is now moved to
internal tag.
Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
On older CPUs, currently numa_node returns value only for socket 0.
Instead, application should be able to make correct decision and
also to keep consistent with the Linux code,
replace the return value to -1.
Fixes: ac7c98d04f2c ("bus/pci: ignore missing NUMA node on Windows")
Cc: stable@dpdk.org
Reported-by: Vipin Varghese <vipin.varghese@intel.com>
Signed-off-by: Pallavi Kadam <pallavi.kadam@intel.com>
Acked-by: Tal Shnaiderman <talshn@nvidia.com>
Some docs and comments in Meson files are still mentioning
the old build system based on "make", removed in 20.11.
After one year, such references are better to be removed.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: David Marchand <david.marchand@redhat.com>
In RegEx device probing, there is register read trying after context
device creation.
When the reading fails, the context device was not freed what caused a
memory leak.
Free it.
Fixes: f324162e8e77 ("regex/mlx5: support combined rule file")
Cc: stable@dpdk.org
Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Acked-by: Ori Kam <orika@nvidia.com>
This patch adds more dummy packet types for QinQ packet,
it enables tcp/udp layer of ipv4/ipv6 for QinQ payload,
so we can use L4 dst/src port as input set for switch
filter.
Signed-off-by: Steve Yang <stevex.yang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
During the port probe process, there are two abnormal branches that did
not release the previously requested memory, resulting in leakage. The
patch adds an iavf_uninit_vf function, which corresponds to the
iavf_init_vf function.
Fixes: ff2d0c345c3b ("net/iavf: support generic flow API")
Cc: stable@dpdk.org
Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
The RTE_ALIGN macro is aligned upwards. If the buf_size variable is not
aligned with 1 << I40E_RXQ_CTX_DBUFF_SHIFT, the rx_buf_len is larger than
the actual mbuf memory after the operation. When receiving the packet, if
the packet is larger than the configured buf_size, it will cause a memory
stepping event.
The patch uses the RTE_ALIGN_FLOOR down alignment macro to correct the
problem.
Fixes: 69dd4c3d0898 ("net/avf: enable queue and device")
Cc: stable@dpdk.org
Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
In the i40e_init_arq function, when the i40e_config_arq_regs function
returns from processing failure, the previously applied arq_bufs resource
is not released, which leads to leakage.
The patch is processed in the same way as the i40e_init_asq function,
maintaining a unified coding style.
Fixes: 49ea51605be4 ("net/i40e/base: gracefully clean the resources")
Cc: stable@dpdk.org
Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
A local test found that repeated port start and stop operations during
the continuous SSE vector bufflist receiving process will cause the mbuf
resource to run out. The final positioning is when the port is stopped,
the mbuf of the pkt_first_seg pointer is not released. Resources leak.
The patch scheme is to judge whether the pointer is empty when the port
is stopped, and release the corresponding mbuf if it is not empty.
Fixes: 69dd4c3d0898 ("net/avf: enable queue and device")
Cc: stable@dpdk.org
Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
If DCF representor port is closed after DCF port is closed, there will
be segmentation fault because representor accesses the released resource
of DCF port.
This patch checks if the resource is present before accessing.
Fixes: 5674465a32c8 ("net/ice: add DCF VLAN handling")
Cc: stable@dpdk.org
Signed-off-by: Dapeng Yu <dapengx.yu@intel.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
In original implementation, if DCF is created on PF1, the PF ID is
still 0, but not 1. Without the right PF ID, the ACL will not work.
This patch makes VF to get its parent's physical function ID.
Fixes: 0b02c9519432 ("net/ice: handle PF initialization by DCF")
Cc: stable@dpdk.org
Signed-off-by: Dapeng Yu <dapengx.yu@intel.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
In the eth_i40e_dev_init function, the tunnel and ethertype hash table
resource release interface should be rte_hash_free instead of rte_free,
and the previously registered interrupt handling function also needs to
be removed from the interrupt list. The patch is amended to use the
correct interface to release the hash table resource and release the
interrupt handling function at the same time.
Fixes: 425c3325f0b0 ("net/i40e: store tunnel filter")
Fixes: 5c53c82c8174 ("net/i40e: store flow director filter")
Cc: stable@dpdk.org
Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
A local test found that repeated port start and stop operations during
the continuous SSE vector bufflist receiving process will cause the mbuf
resource to run out. The final positioning is when the port is stopped,
the mbuf of the pkt_first_seg pointer is not released. Resources leak.
The patch scheme is to judge whether the pointer is empty when the port
is stopped, and release the corresponding mbuf if it is not empty.
Fixes: 4861cde46116 ("i40e: new poll mode driver")
Cc: stable@dpdk.org
Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
The VIRTCHNL_OP_QUERY_FDIR_FILTER opcode is not used, so remove it.
Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Add TCP/UDP/SCTP header checksum field selectors, they can be used in
creating FDIR or RSS rules related to TCP/UDP/SCTP header checksum.
Signed-off-by: Alvin Zhang <alvinx.zhang@intel.com>
Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
The QFI is 6-bit "QoS Flow Identifier" within the GTPU Extension Header.
Add virtchnl fields QFI of GTPU UL/DL for supporting the AVF FDIR.
Signed-off-by: Junfeng Guo <junfeng.guo@intel.com>
Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Update MTU value based on PTP enable status and reserve eight
bytes in TX path to accommodate VLAN tags.
If PTP is enabled maximum allowed MTU is 9200 otherwise it's 9208.
Fixes: b5dc3140448e ("net/octeontx2: support base PTP")
Cc: stable@dpdk.org
Signed-off-by: Hanumanth Reddy Pothula <hpothula@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Adding a new callback for reading the link status. PF can read its
link status and can forward the same to VF once it comes up.
Signed-off-by: Harman Kalra <hkalra@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Currently link event is only sent to the PF by AF as soon as it comes
up, or in case of any physical change in link. PF will broadcast
these link events to all its VFs as soon as it receives it.
But no event is sent when a new VF comes up, hence it will not have
the link status.
Adding support for sending link status to the VF once it comes up
successfully.
Signed-off-by: Harman Kalra <hkalra@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Add ROC API to configure dual VLAN tag addition and removal.
Signed-off-by: Satheesh Paul <psatheesh@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Since the addition of support for runtime queue setup,
receive queues that are started by default no longer
have the correct state. Fix this by setting the state
when a port is started.
Fixes: 0105ea1296c9 ("net/bnxt: support runtime queue setup")
Cc: stable@dpdk.org
Signed-off-by: Lance Richardson <lance.richardson@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
The RSS expansion is based on DFS algorithm to traverse over the possible
expansion paths.
The current implementation breaks out, if it reaches the terminator of
the "next nodes" array, instead of going backwards to try the next path.
For example:
testpmd> flow create 0 ingress pattern eth / ipv6 / udp / vxlan / end
actions rss level 2 types tcp end / end
The paths found are:
ETH IPV6 UDP VXLAN END
ETH IPV6 UDP VXLAN ETH IPV4 TCP END
ETH IPV6 UDP VXLAN ETH IPV6 TCP END
The traversal stopped after getting to the terminator of the next nodes
of the ETH node. It missed the rest of the nodes in the next nodes array
of the VXLAN node.
The fix is to go backwards when reaching the terminator of the current
level and find if there is a "next node" to start traversing a new path.
Using the above example, the flows will be:
ETH IPV6 UDP VXLAN END
ETH IPV6 UDP VXLAN ETH IPV4 TCP END
ETH IPV6 UDP VXLAN ETH IPV6 TCP END
ETH IPV6 UDP VXLAN IPV4 TCP END
ETH IPV6 UDP VXLAN IPV6 TCP END
The traversal will find additional paths, because it traverses through
all the next nodes array of the VXLAN node.
Fixes: 4ed05fcd441b ("ethdev: add flow API to expand RSS flows")
Cc: stable@dpdk.org
Signed-off-by: Lior Margalit <lmargalit@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
The RSS expansion algorithm is using a graph to find the possible
expansion paths. A graph node with the 'explicit' flag will be skipped,
if it is not found in the flow pattern.
The current implementation misses the case where the node with the
explicit flag is in the middle of the expanded path.
For example:
testpmd> flow create 0 ingress pattern eth / ipv6 / udp / vxlan / end
actions rss level 2 types tcp end / end
The VLAN node has the explicit flag, so it is currently included in the
expanded flow:
ETH IPV6 UDP VXLAN END
ETH IPV6 UDP VXLAN ETH VLAN IPV4 TCP END
ETH IPV6 UDP VXLAN ETH VLAN IPV6 TCP END
The fix is to skip the nodes with the explicit flag while iterating over
the possible expansion paths. Using the above example, the flows will be:
ETH IPV6 UDP VXLAN END
ETH IPV6 UDP VXLAN ETH IPV4 TCP END
ETH IPV6 UDP VXLAN ETH IPV6 TCP END
Fixes: 3f02c7ff6815 ("net/mlx5: fix RSS expansion for inner tunnel VLAN")
Cc: stable@dpdk.org
Signed-off-by: Lior Margalit <lmargalit@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
Use max-pkt-len only if jumbo frames offload is requested
since otherwise this field isn't valid.
Fixes: 8b90e4358112 ("net/virtio: set offload flag for jumbo frames")
Fixes: 4e8169eb0d2d ("net/virtio: fix Rx scatter offload")
Cc: stable@dpdk.org
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
When virtio_init_queue returns error, the memory of vq is freed.
But the value of hw->vqs[queue_idx] does not restore.
If virtio_init_queue returns error, the memory of vq is freed again
in virtio_free_queues.
Fixes: 69c80d4ef89b ("net/virtio: allocate queue at init stage")
Cc: stable@dpdk.org
Signed-off-by: Gaoxiang Liu <liugaoxiang@huawei.com>
Reviewed-by: Chenbo Xia <chenbo.xia@intel.com>
The callfds[] array stores eventfds sequentially for Rx and Tx vq.
Fixes: 3d4fb6fd2505 ("net/virtio-user: support Rx interrupt")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
There is no reason to re-register a interrupt handler for LSC if this
feature was not requested in the first place.
A simple use case is when asking for Rx interrupts without LSC interrupt.
Fixes: 26b683b4f7d0 ("net/virtio: setup Rx queue interrupts")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Report max/min/align descriptors limits in device info get callback.
Before calling the callback, rte_eth_dev_info_get() provides
default values of nb_min as zero and nb_max as UINT16_MAX that are
not correct for the driver, so one can't rely on them.
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Descriptors number may be set less than queue size for split queue
vectorized Rx path. Pointers to mbufs for received packets are
obtained from SW ring, that is initially filled with them in the end
of queue setup in virtio_dev_rx_queue_setup_finish(). The begin of the
SW ring filled up to the size of descriptors number. At queue size
offset from the begin of the SW ring pointers to some fake mbuf are also
set for wrapping purpose. So the ring may contains the hole of invalid
pointers from descriptors number offset to queue size offset, and split
vectorized Rx routines could write to the invalid addresses since they
use the ring up to the queue size. Fix this by setting descriptors
number to queue size on Rx queue setup.
Fixes: fc3d66212fed ("virtio: add vector Rx")
Cc: stable@dpdk.org
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Rx queue setup callback allows to use the whole ring when
descriptor number argument equals zero. There's no point to
handle zero in any way since RTE Rx queue setup function
rte_eth_rx_queue_setup() doesn't pass zero using fallback
values.
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Rx queue setup finish function may report wrong number of
allocated mbufs in case of in-order feature. Fix the
function to not ignore allocation error and count only
successfully allocated number of buffers.
Fixes: e5f456a98d3c ("net/virtio: support in-order Rx and Tx")
Cc: stable@dpdk.org
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
This API was introduced in 18.08, therefore removing
experimental tag to promote it to stable state.
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Acked-by: Ray Kinsella <mdr@ashroe.eu>
Error occurs when configuring meson with --buildtype=minsize
with GCC 11.1.0:
drivers/vdpa/mlx5/mlx5_vdpa_mem.c: In function ‘mlx5_vdpa_mem_register’:
drivers/vdpa/mlx5/mlx5_vdpa_mem.c:183:24: error:
initialization of ‘uint64_t’ {aka ‘long unsigned int’} from ‘void *’
makes integer from pointer without a cast [-Werror=int-conversion]
| uint64_t gcd = NULL;
| ^~~~
drivers/vdpa/mlx5/mlx5_vdpa_mem.c:244:75: error:
‘mode’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
| klm_size = mode == MLX5_MKC_ACCESS_MODE_KLM ?
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| KLM_SIZE_MAX_ALIGN(empty_region_sz) : gcd;
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Matan Azrad <matan@nvidia.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Error occurs when configuring meson with --buildtype=minsize
with GCC 11.1.0:
drivers/regex/mlx5/mlx5_regex_fastpath.c:398:17: error:
‘len’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
| complete_umr_wqe(qp, sq, &qp->jobs[mkey_job_id], sq->pi,
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| klm_num, len);
| ~~~~~~~~~~~~~
drivers/regex/mlx5/mlx5_regex_fastpath.c:315:31: note: ‘len’ was declared here
| uint32_t klm_num = 0, len;
| ^~~
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Error occurs when configuring meson with --buildtype=minsize
with GCC 11.1.0:
In function ‘__internal_ram_wr_relaxed’,
inlined from ‘internal_ram_wr’ at ecore_int_api.h:166:2,
inlined from ‘qede_update_rx_prod.constprop’ at qede_rxtx.c:736:2:
drivers/net/qede/base/bcm_osal.h:136:9: error:
‘rx_prods’ is used uninitialized [-Werror=uninitialized]
| rte_write32_relaxed((_val), (_reg_addr))
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
ecore_int_api.h:151:17: note: in expansion of macro ‘DIRECT_REG_WR_RELAXED’
| DIRECT_REG_WR_RELAXED(p_hwfn, &((u32 OSAL_IOMEM *)addr)[i],
| ^~~~~~~~~~~~~~~~~~~~~
drivers/net/qede/qede_rxtx.c: In function ‘qede_update_rx_prod.constprop’:
drivers/net/qede/qede_rxtx.c:724:33: note: ‘rx_prods’ declared here
| struct eth_rx_prod_data rx_prods = { 0 };
| ^~~~~~~~
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Devendra Singh Rawat <dsinghrawat@marvell.com>
Acked-by: Rasesh Mody <rmody@marvell.com>
The PCI and vdev bus drivers cannot be disabled for DPDK builds and
special logic is put in place to not skip them when they are specified
in the disable list. This logic is broken though, as the inclusion of
the driver-specific meson.build file is only included in the "else" leg
of the condition check. This means that when they are specified as
disabled the PCI and vdev buses are not disabled, but neither are their
source files compiled.
Fix this by moving the "subdir()" call into the next "if build" block,
ensuring that if not disabled the sources are always included. To take
account of the fact that the subdir call could itself disable the
driver, we add a break call into the following loop to ensure we quickly
fall through to the following block which stops processing appropriately
if the driver is disabled.
Fixes: 2e33309ebe03 ("config: enable/disable drivers in Arm builds")
Cc: stable@dpdk.org
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Tested-by: Radu Nicolau <radu.nicolau@intel.com>
Acked-by: Radu Nicolau <radu.nicolau@intel.com>
Apply the same fix that for iavf to DCF
commit ead06572bd8f ("net/iavf: fix performance with writeback policy")
Fixes: 4b0d391f0eab ("net/ice: add queue config in DCF")
Cc: stable@dpdk.org
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Tested-by: Lijuan Tu <lijuan.tu@intel.com>
