Commit Graph

29436 Commits

Author SHA1 Message Date
Lance Richardson
cc7f749488 net/bnxt: fix Rx queue startup state
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: 0105ea1296 ("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>
2021-09-15 02:07:08 +02:00
Lior Margalit
aa52e5f0f9 net/mlx5: fix RSS expansion traversal over next nodes
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: 4ed05fcd44 ("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>
2021-09-13 21:56:10 +02:00
Lior Margalit
69d268b4ff net/mlx5: fix RSS expansion for explicit graph node
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: 3f02c7ff68 ("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>
2021-09-13 21:56:09 +02:00
Ivan Ilchenko
580f3af31c net/virtio: fix device configure without jumbo Rx offload
Use max-pkt-len only if jumbo frames offload is requested
since otherwise this field isn't valid.

Fixes: 8b90e43581 ("net/virtio: set offload flag for jumbo frames")
Fixes: 4e8169eb0d ("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>
2021-09-14 13:21:57 +02:00
Chenbo Xia
945ef8a040 vhost: promote some APIs to stable
As reported by symbol bot, APIs listed in this patch have been
experimental for more than two years. This patch promotes these
18 APIs to stable.

Signed-off-by: Chenbo Xia <chenbo.xia@intel.com>
Acked-by: Ray Kinsella <mdr@ashroe.eu>
Acked-by: Kevin Traynor <ktraynor@redhat.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
2021-09-14 13:21:57 +02:00
Gaoxiang Liu
e53084d84b vhost: log socket path on adding connection
Add log print of socket path in vhost_user_add_connection.
It's useful when adding a mass of socket connections,
because the information of every connection is clearer.

Fixes: 8f972312b8 ("vhost: support vhost-user")
Cc: stable@dpdk.org

Signed-off-by: Gaoxiang Liu <liugaoxiang@huawei.com>
Reviewed-by: Chenbo Xia <chenbo.xia@intel.com>
2021-09-14 13:21:57 +02:00
Gaoxiang Liu
5d903aee8a net/virtio: fix repeated freeing of virtqueue
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: 69c80d4ef8 ("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>
2021-09-14 13:21:57 +02:00
Gaoxiang Liu
451dc0fad8 vhost: fix crash on port deletion
The rte_vhost_driver_unregister() and vhost_user_read_cb()
can be called at the same time by 2 threads.
when memory of vsocket is freed in rte_vhost_driver_unregister(),
the invalid memory of vsocket is accessed in vhost_user_read_cb().
It's a bug of both mode for vhost as server or client.

E.g., vhostuser port is created as server.
Thread1 calls rte_vhost_driver_unregister().
Before the listen fd is deleted from poll waiting fds,
"vhost-events" thread then calls vhost_user_server_new_connection(),
then a new conn fd is added in fdset when trying to reconnect.
"vhost-events" thread then calls vhost_user_read_cb() and
accesses invalid memory of socket while thread1 frees the memory of
vsocket.

E.g., vhostuser port is created as client.
Thread1 calls rte_vhost_driver_unregister().
Before vsocket of reconn is deleted from reconn list,
"vhost_reconn" thread then calls vhost_user_add_connection()
then a new conn fd is added in fdset when trying to reconnect.
"vhost-events" thread then calls vhost_user_read_cb() and
accesses invalid memory of socket while thread1 frees the memory of
vsocket.

The fix is to move the "fdset_try_del" in front of free memory of conn,
then avoid the race condition.

The core trace is:
Program terminated with signal 11, Segmentation fault.

Fixes: 52d874dc67 ("vhost: fix crash on closing in client mode")
Cc: stable@dpdk.org

Signed-off-by: Gaoxiang Liu <liugaoxiang@huawei.com>
Reviewed-by: Chenbo Xia <chenbo.xia@intel.com>
2021-09-14 13:21:57 +02:00
David Marchand
848e93d900 net/virtio-user: fix Rx interrupts with multi-queue
The callfds[] array stores eventfds sequentially for Rx and Tx vq.

Fixes: 3d4fb6fd25 ("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>
2021-09-14 13:21:57 +02:00
David Marchand
5be2325ece net/virtio: avoid unneeded link interrupt configuration
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: 26b683b4f7 ("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>
2021-09-14 13:21:57 +02:00
Ivan Ilchenko
31136836bc net/virtio: report max/min/align Rx descriptor limits
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>
2021-09-14 13:21:57 +02:00
Ivan Ilchenko
c97e73779c net/virtio: fix split queue vectorized Rx
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: fc3d66212f ("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>
2021-09-14 13:21:57 +02:00
Ivan Ilchenko
9897896ec7 net/virtio: remove handling of zero desc on Rx setup
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>
2021-09-14 13:21:57 +02:00
Ivan Ilchenko
b8d0a27842 net/virtio: fix mbuf count on Rx queue setup
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: e5f456a98d ("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>
2021-09-14 13:21:57 +02:00
Jiayu Hu
abeb865255 vhost: remove copy threshold for async path
Copy threshold has been introduced in async vhost data
path to select the appropriate copy engine to do copies
for higher efficiency.

However, it may cause packets ordering issues and also
introduces performance unpredictability.

Therefore, this patch removes copy threshold support in
async vhost data path.

Signed-off-by: Jiayu Hu <jiayu.hu@intel.com>
Signed-off-by: Cheng Jiang <cheng1.jiang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
2021-09-14 13:21:55 +02:00
Mohsin Kazmi
818ce1132a net: fix checksum offload for outer IPv4
Preparation of the headers for the hardware offload
misses the outer IPv4 checksum offload.
It results in bad checksum computed by hardware NIC.

This patch fixes the issue by setting the outer IPv4
checksum field to 0.

Fixes: 4fb7e803eb ("ethdev: add Tx preparation")
Cc: stable@dpdk.org

Signed-off-by: Mohsin Kazmi <mohsin.kazmi14@gmail.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
2021-09-15 12:51:49 +02:00
David Marchand
b37ed6def3 ethdev: promote sibling iterators to stable
This API saw no update since its introduction and will help applications
like OVS ([1] and [2]) that currently look at rte_eth_devices[] to
achieve the same.

1: https://github.com/openvswitch/ovs/blob/master/lib/netdev-dpdk.c#L1285
2: https://github.com/openvswitch/ovs/blob/master/lib/netdev-dpdk.c#L1476

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Ray Kinsella <mdr@ashroe.eu>
2021-09-15 11:48:54 +02:00
Haiyue Wang
9fecac6c3a ethdev: promote burst mode API
The DPDK Symbol Bot reports:
Please note the symbols listed below have expired. In line with the
DPDK ABI policy, they should be scheduled for removal, in the next
DPDK release.

Symbol
rte_eth_rx_burst_mode_get
rte_eth_tx_burst_mode_get

Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Ray Kinsella <mdr@ashroe.eu>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
2021-09-15 10:46:00 +02:00
Joyce Kong
97d7b92772 ethdev: fix typo in Rx queue setup API comment
Fix a typo that mb_pool was misspelt as mp_pool.

Fixes: 4ff702b5df ("ethdev: introduce Rx buffer split")
Cc: stable@dpdk.org

Signed-off-by: Joyce Kong <joyce.kong@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
2021-09-15 10:38:43 +02:00
Pavan Nikhilesh
2def522abc ethdev: promote API to set packet types
Remove experimental tag from rte_eth_dev_set_ptypes().

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Ray Kinsella <mdr@ashroe.eu>
2021-09-15 09:54:04 +02:00
Jasvinder Singh
e73a7ab224 net/softnic: promote manage API
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>
2021-09-15 09:26:23 +02:00
Xiaoyun Li
dbd34bee98 ethdev: promote API to get interrupt FD per queue
Remove the experimental tag for rte_eth_dev_rx_intr_ctl_q_get_fd API
that was introduced in 18.11 and have been around for 11 releases.

Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Ray Kinsella <mdr@ashroe.eu>
Acked-by: David Marchand <david.marchand@redhat.com>
2021-09-14 18:18:06 +02:00
Conor Walsh
4777674c44 eal: fix memory leak when saving arguments
This patch fixes a memleak which was reported in Bugzilla within the
eal_save_args function. This was caused by the function mistakenly
adding -- to the eal args instead of breaking beforehand.

Bugzilla ID: 722
Fixes: 293c53d8b2 ("eal: add telemetry callbacks")

Reported-by: Zhihong Peng <zhihongx.peng@intel.com>
Signed-off-by: Conor Walsh <conor.walsh@intel.com>
Signed-off-by: Conor Fogarty <conor.fogarty@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2021-09-16 22:24:54 +02:00
Juraj Linkeš
e7ef133d77 config/arm: make N2 crypto an optional feature
Not all Neoverse-N2 CPUs must support the crypto feature/extension
which makes it an optional feature.
Only enable the feature for SoCs which support it.

Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech>
2021-09-16 18:12:46 +02:00
Juraj Linkeš
95e0f23022 config/arm: split -march into arch and features
Older compilers may not support all arch versions and all features that
the target SoC supports, in which case it's better to figure out the
highest arch version and features that the compiler supports.
Implement a way to achieve this:
1. Find the highest arch version that the compiler supports,
keeping in mind the SoC arch version we're building.
For example, if the SoC arch version is arm8.2-a,
but the compiler only supports arm8.1-a, use arm8.1-a.
On the other hand, if the compiler supports arm8.3-a (or higher),
use armv8.2-a.
2. With the architecture version locked, iterate over SoC features and
use all that are supported.

In all cases, emit a warning if there's something unsupported by the
compiler.

Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech>
2021-09-16 18:12:46 +02:00
Ruifeng Wang
328720c594 examples/service_cores: fix lcore count check
The example has various profiles to run services on specified
number of lcores. Due to incorrect boundary condition, service
can be dispatched to a core that does not exist. This puts main
core into endless wait.

Max available number of service cores is all detected lcores
excluding main core.

Fixes: 7f6ee6aee7 ("examples/service_cores: check cores before run")
Cc: stable@dpdk.org

Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
2021-09-16 17:31:03 +02:00
Stephen Hemminger
7df485eb3d eal: remove deprecated noninclusive API
New API for these were added in 20.11 and the old API was retained
but marked deprecated. Since 21.11 is the next LTS, it is time
to remove the deprecated ones.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Acked-by: David Marchand <david.marchand@redhat.com>
2021-09-16 17:21:22 +02:00
Joyce Kong
d5559ac589 test/func_reentrancy: free memzones after test
Function reentrancy test limits maximum number of iterations
simultaneously, however it doesn't free the 'fr_test_once'
memzones after the fact, so introduce freeing 'fr_test_once'
in ring/mempool/hash/fbk/lpm_clean.

Meanwhile, add the missing free for test case on main thread.

Fixes: 104a92bd02 ("app: add reentrancy tests")
Fixes: 995eec6190 ("test: clean up memory for function reentrancy test")
Cc: stable@dpdk.org

Signed-off-by: Joyce Kong <joyce.kong@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Feifei Wang <feifei.wang2@arm.com>
Reviewed-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: David Marchand <david.marchand@redhat.com>
2021-09-16 17:17:34 +02:00
Juraj Linkeš
8ef09fdc50 build: add optional NUMA and CPU counts detection
Add an option to automatically discover the host's NUMA and CPU counts
and use those values for a non cross-build.
Give users the option to override the per-arch default values or values
from cross files by specifying them on the command line with -Dmax_lcores
and -Dmax_numa_nodes.

Signed-off-by: Juraj Linkeš <juraj.linkes@pantheon.tech>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: David Christensen <drc@linux.vnet.ibm.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2021-09-16 09:43:28 +02:00
Thomas Monjalon
9c8c806487 test/crypto: fix minsize build
Error occurs when configuring meson with --buildtype=minsize
with GCC 11.1.0:

app/test/test_cryptodev_blockcipher.c:1133:45: error:
‘blk_tcs’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
|         const struct blockcipher_test_case *blk_tcs;
|                                             ^~~~~~~

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
2021-09-15 17:12:29 +02:00
Thomas Monjalon
5dd12566f1 vdpa/mlx5: fix minsize build
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>
2021-09-15 17:12:29 +02:00
Thomas Monjalon
51d7396440 regex/mlx5: fix minsize build
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>
2021-09-15 17:12:29 +02:00
Thomas Monjalon
11c2e4b41c net/qede: fix minsize build
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>
2021-09-15 17:11:56 +02:00
David Marchand
dfb1ad1e7a config/ppc: fix build with GCC >= 10
Like for python, multiline statements in meson must either use a
backslash character (explicit continuation) or be enclosed in ()
(implicit continuation).

python PEP8 recommends the latter [1], and it looks like meson had
an issue with backslash before 0.50 [2].

1: https://www.python.org/dev/peps/pep-0008/#multiline-if-statements
2: https://github.com/mesonbuild/meson/commit/90c9b868b20b

Fixes: 394407f50c ("config/ppc: ignore GCC 11 psabi warnings")

Reported-by: Ferruh Yigit <ferruh.yigit@intel.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
Tested-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2021-09-15 17:05:44 +02:00
Bruce Richardson
0e837e584e build: fix essential drivers in disable list
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: 2e33309ebe ("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>
2021-09-15 11:00:15 +02:00
Asaf Penso
1c91ed5cdd doc: add release milestones definition
Adding more information about the release milestones.
This includes the scope of change, expectations, etc.

Signed-off-by: Asaf Penso <asafp@nvidia.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
2021-09-14 18:51:34 +02:00
Dmitry Kozlyuk
5ffa86a2b4 build: propagate Windows system dependencies to pkg-config
Windows EAL depends on some system libraries. They were linked using
add_project_link_arguments('-l<LIB>'), which prevented meson from adding
them to Libs.private of pkg-config file. As a result, applications using
pkg-config to find DPDK hit link errors, for example:

    librte_eal.a(eal_windows_eal_debug.c.obj) : error LNK2019: unresolved
    external symbol __imp_SymInitialize referenced in function
    rte_dump_stack

Reference required libraries in EAL using ext_deps meson variable.
bus/pci and net/pcap depend on lib/eal and will pull them automatically.
Drop advapi32 dependency, as MinGW locates VirtualAlloc2() dynamically.

Fixes: 2a5d547a4a ("eal/windows: implement basic memory management")
Fixes: c91717eb75 ("eal/windows: support exit and panic")
Cc: stable@dpdk.org

Reported-by: William Tu <u9012063@gmail.com>
Signed-off-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
Tested-by: William Tu <u9012063@gmail.com>
2021-09-14 15:58:34 +02:00
Thomas Monjalon
557610a8ff cryptodev: fix indent in Meson file
Fixes: af668035f7 ("cryptodev: expose driver interface as internal")

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
2021-09-14 15:58:32 +02:00
Qi Zhang
af167bcf17 net/ice: fix performance with writeback policy
Apply the same fix that for iavf to DCF
commit ead06572bd ("net/iavf: fix performance with writeback policy")

Fixes: 4b0d391f0e ("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>
2021-09-06 03:57:13 +02:00
Qiming Chen
d54b51efb6 net/ixgbe: fix mbuf leak
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: af75078fec ("first public release")
Cc: stable@dpdk.org

Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
2021-09-06 03:46:22 +02:00
Qiming Chen
cc8aaa258f net/ixgbe: fix MAC resource leak
In the eth_ixgbevf_dev_init and eth_ixgbe_dev_init functions, memory is
allocated for the MAC address, and the address is stored in the
eth_dev->data->mac_addrs member variable. If the subsequent function is
abnormal, you need to use the rte_free function to release the MAC
address memory.

Fixes: af75078fec ("first public release")
Cc: stable@dpdk.org

Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
2021-09-06 03:44:58 +02:00
Qiming Chen
3a732dce93 net/ixgbe: fix queue resource leak
In the ixgbevf_dev_start function, after initializing the rxtx queue, if
an exception occurs in the subsequent function, the rxtx queue needs to
be released. The patch solves the problem of queue resource leakage.

Fixes: 0eb609239e ("ixgbe: enable Rx queue interrupts for PF and VF")
Cc: stable@dpdk.org

Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
2021-09-06 03:43:06 +02:00
Qiming Chen
03f83d6c5c net/ixgbe: fix hash handle leak
In the ixgbe_fdir_filter_init and ixgbe_l2_tn_filter_init functions,
after the hash handle is created, the handle is not released in
subsequent abnormal branches.

Fixes: 080e3c0ee9 ("net/ixgbe: store flow director filter")
Fixes: d0c0c416ef ("net/ixgbe: store L2 tunnel filter")
Cc: stable@dpdk.org

Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
2021-09-06 03:39:46 +02:00
Dapeng Yu
1a86f4dbdf net/ice: support DCF device reset
DCF PMD needs to support rte_eth_dev_reset, the reason is when a DCF
instance is killed, all the flow rules still exists in hardware, when
DCF gets to reconnect, it already lost the flow context, and if the
application wants to create new rules, it may fail due to firmware
reports rules already exist.

The rte_eth_dev_reset API provides a more elegant way for the
application to reset DCF when reconnect happens.

Signed-off-by: Dapeng Yu <dapengx.yu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
2021-09-06 03:17:29 +02:00
Ivan Malov
142dd26a2f net/sfc: set FDIR bit for flow mark in EF100 Rx
According to flow action MARK definition, PMDs must set both
PKT_RX_FDIR and PKT_RX_FDIR_ID if the packet contains a mark.

Fixes: 1aacc3d388 ("net/sfc: support user mark and flag Rx for EF100")
Cc: stable@dpdk.org

Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
2021-09-09 19:52:36 +02:00
Min Hu (Connor)
fe9c27a337 net/hns3: fix Tx push capability
This patch fixes Tx push capability to be compatible with Kunpeng 920,
as Tx push is only supported on Kunpeng 930.

Fixes: 23e317dd1f ("net/hns3: support Tx push quick doorbell for performance")
Cc: stable@dpdk.org

Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
2021-09-09 19:35:40 +02:00
Chengchang Tang
3fb0df3138 net/hns3: fix taskqueue pair reset command
This new taskqueue pair reset command is used incorrectly, resulting in
the new command not taking effect.

This patch fixes the incorrect use.

Fixes: 6911e7c22c ("net/hns3: fix long task queue pairs reset time")
Cc: stable@dpdk.org

Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
2021-09-09 19:35:40 +02:00
Chengchang Tang
aa3497d445 net/hns3: fix queue flow action validation
The used_rx_queues only takes effect after device is started, and
its value is incorrect before the device is started. Therefore, it
is not suitable for flow action to use it to verify the queue index
before the device is started.

E.g. Enable dedicated queue in bonding device will configure a queue
flow action before start its slave devices. The above problem will
make this reasonable flow action configuration fail.

This patch use the nb_rx_queues from the configuration phase to
achieve verification.

Fixes: a951c1ed3a ("net/hns3: support different numbers of Rx and Tx queues")
Fixes: f8e7fcbfd0 ("net/hns3: support flow action of queue region")
Cc: stable@dpdk.org

Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
2021-09-09 19:35:40 +02:00
Qiming Chen
d70157c13f net/pcap: improve Rx statistics
In the receiving direction, if alloc mbuf or jumbo process failed, there
is no err_pkts count, which makes it difficult to locate the problem.
Because alloc mbuf failed, the rx_nombuf field is counted.

Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
2021-09-09 15:30:32 +02:00
Qiming Chen
da0280b3a3 net/pcap: fix resource leakage on port probe
When the port is probed, if the eth_from_pcaps function fails, the
previously opened pcap resources are not released, causing resource
leakage.

The patch solves the problem of resource leakage caused by abnormal
branch exit during the port probe process.

Fixes: 4c173302c3 ("pcap: add new driver")
Cc: stable@dpdk.org

Signed-off-by: Qiming Chen <chenqiming_huawei@163.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
2021-09-09 15:01:47 +02:00