Add new firmware API FW_PARAM_DEV_FILTER_MODE_MASK to fetch
the filtermode and filtermask values configured in hardware,
which are used to validate the match combinations in the filter
spec before offloading the filter rules to hardware. For older
firmware that doesn't support the new API, fallback to older way
of directly reading from indirect registers
Signed-off-by: Karra Satwik <kaara.satwik@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Add support to rewrite Source MAC addresses. The new Source
MAC address is written into a free entry in the SMT table
and the corresponding SMT index is used by hardware to
rewrite the Source MAC address of the packets hitting the
flow.
Signed-off-by: Karra Satwik <kaara.satwik@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Source MAC Table (SMT) is used for storing Source MAC
addresses to be written in packets transmitted on the
wire. Hence, the SMT table can be used for overwriting
Source MAC addresses in packets, hitting corresponding
filter rules inserted by the rte_flow API.
Query firmware for SMT start and size information available
to the underlying PF. Allocate and maintain the corresponding
driver's copy of the hardware SMT table, with appropriate
refcount mechanism. If SMT information is not available, then
use the entire hardware SMT table.
Signed-off-by: Karra Satwik <kaara.satwik@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Add support for overwriting destination MAC addresses.
The new MAC address is written into a free entry in the
L2T table and the corresponding L2T index is used by
hardware to overwrite the destination MAC address of the
packets hitting the flow
Signed-off-by: Karra Satwik <kaara.satwik@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Add support to match all packets received on the underlying VF.
Use new firmware API to fetch the Virtual Interface Number (VIN)
allocated to each VF by the firmware. The VIN is required to
write filter rules to match all packets on VFs, whose identifier
is beyond max 7-bit value (i.e. 127) in VIID.
If firmware doesn't support fetching the VIN information, then
fallback to manually retrieving the VIN value from the 7-bit field
in the VIID, which only supports in range of 0..127. In this case,
packets belonging to VFs, whose identifier is beyond 127 can't be
matched.
Signed-off-by: Karra Satwik <kaara.satwik@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Add support to match all packets received on the underlying PF
Note that the same 17-bit hardware tuple is shared between QinQ
and PF match. Hence, match on either QinQ or PF only can be done
at a time. Both QinQ and PF match can't be enabled at the same time.
Also, remove check to reject rules without spec because
RTE_FLOW_ITEM_TYPE_PF doesn't require a spec. Due to this check
removal, RTE_FLOW_ITEM_TYPE_PHY_PORT item needs to be updated to
handle NULL spec
Signed-off-by: Karra Satwik <kaara.satwik@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Add support to match Type of Service (TOS) field in
IPv4/IPv6 header
Signed-off-by: Karra Satwik <kaara.satwik@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Add support to match fields in 802.1ad Q-in-Q VLAN packets.
Relax check for repeated pattern items for RTE_FLOW_ITEM_TYPE_VLAN
item, since the same item is used to represent both QinQ and VLAN
packets.
When QinQ match is enabled, the ethertype field in the hardware
spec must contain the innermost VLAN header's ethertype field,
and not the Ethernet header's ethertype field. The hardware
automatically searches for ethertype 0x88A8/0x8100 in Ethernet
header, when parsing incoming packet against QinQ/VLAN pattern,
respectively.
Signed-off-by: Karra Satwik <kaara.satwik@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Make user filters a priority in EF10 datapath. When a manual
filter with a specification that is equal to an existing auto
filter is inserted, the manual filter:
- replaces auto filter if the specification is exclusive;
- is inserted along existing auto filter otherwise;
In the first case the auto filter that was replaced is saved.
This saved filter can be updated on filter reconfiguration and
is restored on the manual filter removal.
Fixes: e7cd430c86 ("net/sfc/base: import SFN7xxx family support")
Cc: stable@dpdk.org
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
The code is refactored to make it more clear and allow for
more convenient further changes.
Fixes: e7cd430c86 ("net/sfc/base: import SFN7xxx family support")
Cc: stable@dpdk.org
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
The automatic filter priority may only be used by filter implementation,
so reject the attempts to create such filters by client drivers.
Fixes: f9565517ff ("net/sfc/base: import filters support")
Cc: stable@dpdk.org
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Only implemented filter priorities are manual and auto. Remove
unused priorities since exposing unimplemented functionality
is of no benefit.
Tx filter specification default priority is changed to manual,
since required priority does not differ from it in the implementation.
Fixes: f9565517ff ("net/sfc/base: import filters support")
Cc: stable@dpdk.org
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
The priority should be explicitly set to manual for correct
usage of libefx filters.
Fixes: a9825ccf5b ("net/sfc: support flow API filters")
Cc: stable@dpdk.org
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Returned errors of ethdev callbacks are negative. Internal
sfc functions return positive errors, so convert them to
negative value.
Fixes: 9039c81257 ("ethdev: change promiscuous callbacks to return status")
Cc: stable@dpdk.org
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Any DPDK public header file which includes stdbool.h may conflict with
local definition of bool, if any, which further results in compilation
error. To avoid, used standard stdbool.h instead of defining bool
internally.
I observed this issue during a development where I included rte_uuid.h
into rte_ethdev.h. As rte_ethdev.h is included to PMD driver, it started
throwing error as given below:
CC ionic_rxtx.o
In file included from .../dpdk/build/include/rte_uuid.h:17:0,
from .../dpdk/build/include/rte_ethdev.h:161,
from .../dpdk/build/include/rte_ethdev_driver.h:18,
from .../dpdk/drivers/net/ionic/ionic_rxtx.c:34:
.../dpdk/drivers/net/ionic/ionic_osdep.h:48:17:
error: two or more data types in declaration specifiers
typedef uint8_t bool;
^
In file included from .../dpdk/drivers/net/ionic/ionic_dev.h:8:0,
from .../dpdk/drivers/net/ionic/ionic.h:13,
from .../dpdk/drivers/net/ionic/ionic_mac_api.h:8,
from .../dpdk/drivers/net/ionic/ionic_rxtx.c:45:
.../dpdk/drivers/net/ionic/ionic_osdep.h:48:1:
warning: useless type name in empty declaration
typedef uint8_t bool;
^~~~~~~
cc1: warning: unrecognized command line option
‘-Wno-address-of-packed-member’
.../dpdk/mk/internal/rte.compile-pre.mk:114:
recipe for target 'ionic_rxtx.o' failed
Fixes: 5ef518098e ("net/ionic: register and initialize adapter")
Cc: stable@dpdk.org
Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Acked-by: Alfredo Cardigliano <cardigliano@ntop.org>
Currently, when promiscuous mode is enabled, it is just allowed to
accept all the unicast and broadcast packets in hns3 PF PMD driver. It
should also be able to receive multicast packets.
Fixes: 19a3ca4c99 ("net/hns3: add start/stop and configure operations")
Cc: stable@dpdk.org
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Currently, we only support VF device is bound to vfio_pci or igb_uio and
then driven by DPDK driver when PF is driven by kernel mode hns3 ethdev
driver, VF is not supported when PF is driven by hns3 DPDK driver.
This patch adds promiscuous and allmulticast mode support for hns3 VF
PMD driver.
1) The promiscuous/allmulticast mode can be configured successfully only
based on the trusted VF device. If based on the non trusted VF
device, configuring promiscuous/allmulticast mode will fail. The hns3
VF device can be configured as trusted device by hns3 PF kernel
ethdev driver on the host by "ip link set <eth num> vf <vf id> turst
on" command.
2) After the promiscuous mode is configured successfully, hns3 VF PMD
driver can receive the ingress and outgoing traffic. In the words,
all the ingress packets, all the packets sent from the PF and other
VFs on the same physical port.
3) Note: Because of the hardware constraints, By default vlan filter is
enabled and couldn't be turned off based on VF device, so vlan filter
is still effective even in promiscuous mode. If upper applications
don't call rte_eth_dev_vlan_filter API function to set vlan based on
VF device, hns3 VF PMD driver will can't receive the packets with
vlan tag in promiscuous mode.
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
The hns3 PF/VF devices on the same port share the hardware MTU
configuration. Currently, we send mailbox to inform hns3 PF kernel
ethdev driver to finish hardware MTU configuration in hns3 VF PMD
driver, there is no need to stop the port for hns3 VF device, and
the MTU value issued by hns3 VF PMD driver must be less than or
equal to PF's MTU.
Signed-off-by: Chengchang Tang <tangchengchang@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Currently, when receiving and transmitting packets based on hns3 network
engine there are probably unexpected and redundant Tx interrupts if Rx
interrupt is enabled.
The root cause as below:
Tx and Rx queues with the same number share the interrupt vector in hns3
network engine, and in this case there are the residual hardware mapping
relationship configuration between queue and interrupt vector configured
in hns3 kernel ethdev driver.
We should clear the all hardware mapping relationship configurations in
the initialization. Because of the hardware constraints, we have to
implement clearing the relationship by binding all queues to the last
interrupt vector and reserving the last interrupt vector, this method
results in a decrease of the maximum queues when upper applications call
the rte_eth_dev_configure API function to enable Rx interrupt.
Fixes: 02a7b55657 ("net/hns3: support Rx interrupt")
Cc: stable@dpdk.org
Signed-off-by: Hao Chen <chenhao164@huawei.com>
Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
Signed-off-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
This patch adds TCP segment offload support for hns3 PMD driver.
Signed-off-by: Hongbo Zheng <zhenghongbo3@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Remove function ice_clear_queues, since all equivalent code
has already been executed during ice_rx|tx_queue_stop.
Also function ice_rx|tx_queue_release_mbufs simply wrapped a
function pointer call which is not necessary, remove them.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
gcc 10.0.1 reports:
../drivers/net/hinic/base/hinic_pmd_hwdev.c: In function ‘print_cable_info’:
../drivers/net/hinic/base/hinic_pmd_hwdev.c:1398:3:
warning:
‘snprintf’ argument 4 may overlap destination object ‘tmp_str’
[-Wrestrict]
1398 | snprintf(tmp_str + strlen(tmp_str), (sizeof(tmp_str) - 1),
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1399 | "%s, Temperature: %u", tmp_str,
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1400 | info->cable_temp);
| ~~~~~~~~~~~~~~~~~
The warning is that tmp_str is in both src and dest.
Anyway, the current code is incorrect and because of the +strlen
the existing string will be repeated twice and max length
does not limit to the end of the string.
Fix by removing tmp_str from the src of snprintf and adding the
correct max length.
Fixes: d9ce191794 ("net/hinic/base: add hardware operation")
Cc: stable@dpdk.org
Signed-off-by: Kevin Traynor <ktraynor@redhat.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
This patch fixes the situation when there is already connected pair of
memif interfaces and another slave tries to initiate the connection with
(already occupied) master. Expected behavior is that the second slave
is refused and gets disconnect message with reason: "Already connected",
while old connection remains functional.
Fixes: 09c7e63a71 ("net/memif: introduce memory interface PMD")
Cc: stable@dpdk.org
Signed-off-by: Július Milan <jmilan.dev@gmail.com>
Reviewed-by: Jakub Grajciar <jgrajcia@cisco.com>
With this patch it is possible to connect 2 DPDK memifs into loopback,
i.e. when they have the same id and different roles, as for example:
"--vdev=net_memif0,role=master,id=0"
"--vdev=net_memif1,role=slave,id=0"
Signed-off-by: Július Milan <jmilan.dev@gmail.com>
Reviewed-by: Jakub Grajciar <jgrajcia@cisco.com>
When setting the CONFIG_RTE_BUILD_SHARED_LIB to y,
and build the mlx5 pmd, there is a building error.
To fix it, add RTE_IBVERBS_LINK_DLOPEN to include
relative codes.
mlx5_common.o: In function `mlx5_glue_init':
drivers/common/mlx5/mlx5_common.c:324: undefined reference to `dlclose'
Fixes: 7b4f1e6bd3 ("common/mlx5: introduce common library")
Cc: stable@dpdk.org
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Tested-by: Ali Alnubani <alialnu@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
This patch adds to MLX5 PMD the support of matching on IPv4 TTL and
IPv6 hoplimit.
Support is valid when using DV flow engine (Verbs flow engine doesn't
support matching on this field).
Signed-off-by: Eli Britstein <elibr@mellanox.com>
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
The nic_mask struct in function mlx5_flow_validate_item_ipv6()
includes hop_limits field.
This is redundant since matching on this item is not supported using
Verbs flow engine (dv_flow_en=0).
(Separate patch will be sent to handle this issue for DV flow engine.)
Setting a rule including match on this field will fail:
testpmd> flow create 0 ingress pattern eth / ipv6 hop is 1 / end actions
drop / end
port_flow_complain(): Caught PMD error type 1 (cause unspecified):
hardware refuses to create flow: Operation not supported
This patch removes the redundant field from nic_mask, to ensure that
such flow rules will be rejected by PMD validation:
testpmd> flow create 0 ingress pattern eth / ipv6 hop is 1 / end actions
drop / end
port_flow_complain(): Caught PMD error type 13 (specific pattern item):
cause: 0x7fffffffa2c8, mask enables non supported bits: Operation not
supported
The related lines in function flow_verbs_translate_item_ipv6() are
removed as well, since they are redundant.
Fixes: 23c1d42c71 ("net/mlx5: split flow validation to dedicated function")
Fixes: 84c406e745 ("net/mlx5: add flow translate function")
Cc: stable@dpdk.org
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Remove CONFIG_RTE_LIBRTE_ICE_RX_ALLOW_BULK_ALLOC with below
consideration:
1. A default Rx path can always be selected by setting a proper
rx_free_thresh value at runtime, see
ice_check_rx_burst_bulk_alloc_preconditions.
2. Its not a big deal to always reserve more space for desc ring.
"ring_size = (uint16_t)(rxq->nb_rx_desc + ICE_RX_MAX_BURST);"
3. Fixes a potential invalid memory access in ice_reset_rx_queue.
If CONFIG_RTE_LIBRTE_ICE_RX_ALLOW_BULK_ALLOC is turned on while
ice_check_rx_burst_bulk_alloc_preconditions return fail.
Below code will have problem.
for (i = 0; i < ICE_RX_MAX_BURST; ++i)
rxq->sw_ring[rxq->nb_rx_desc + i].mbuf = &rxq->fake_mbuf;
Fixes: 50370662b7 ("net/ice: support device and queue ops")
Cc: stable@dpdk.org
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
This is needed to reuse pattern processing engine for MAE.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
As a preparation step, generalise flow start and stop path using
callbacks.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Generalise flow attribute parsing function in regard to transfer attribute.
Add a method table and factor out VNIC-specific parsing code as a callback.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Riverhead boards maintain support for MAE, a low-level Match-Action Engine.
The upcoming patches will bring support for MAE to RTE flow implementation.
A follow-up patch will introduce MAE-specific specification cleanup method.
In order to prepare for the patch, introduce flow allocation and free path.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Add the concept of a flow specification type.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
A backend which a driver employs to handle flow rules of a given
type depends on the underlying NIC flow engine. The driver in
question in its current state is tailored to support the only
flow engine, VNIC filtering. As the need arises to add support
for transfer rules, the driver has to be reworked so that it
becomes possible to introduce yet another backend.
As a preparation step, make the flow list shared
between different engines.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
The flow specification structure will be transformed to a generic one, and
its current contents will be fenced off to form a VNIC-specific parameters.
Flow RSS details do not belong to the said specification currently, albeit
being VNIC-specific. This patch addresses this issue as a preparation step.
Signed-off-by: Ivan Malov <ivan.malov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Implement eth_dev_ops uc_hash_table_set and uc_all_hash_table_set
Signed-off-by: Chandu Babu N <chandu@amd.com>
Acked-by: Ravi Kumar <ravi1.kumar@amd.com>
Supports adding MAC addresses to enable whitelist filtering to
accept packets
implement eth_dev_ops mac_addr_set, mac_addr_add, mac_addr_remove and
set_mc_addr_list
Signed-off-by: Chandu Babu N <chandu@amd.com>
Acked-by: Ravi Kumar <ravi1.kumar@amd.com>
Stopping LLDP is necessary for DPDK, but it will cause DCB init failure.
For kernel shared code, the prerequisite for successful initialization
of DCB is that LLDP is enabled. This patch starts LLDP before DCB init
and stops it when init finishes.
Fixes: a8e84b22bd ("net/i40e/base: support persistent LLDP")
Cc: stable@dpdk.org
Signed-off-by: Shougang Wang <shougangx.wang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
NIC deinitialization should not happen after a failed NIC
initialization.
Fixes: 91831d4068 ("net/sfc: estimate available resources")
Cc: stable@dpdk.org
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
When calling to setup RSS on v4 API, ESX will expect
IPv4/6 TCP RSS to be set/requested mandatory.
This patch will:
- Set IPv4/6 TCP RSS when these have not been set. A warning
message is thrown to make sure we warn the application we are
setting IPv4/6 TCP RSS when not set.
- An additional check has been added to dodge RSS configuration
altogether unless MQ_RSS has been requested, similar to v3.
The alternative (returning error) was considered, the intent
is to ease the task of setting up and running vmxnet3 in situations
where it's supposed to be most straightforward (testpmd, pktgen).
Bugzilla ID: 400
Fixes: 643fba7707 ("net/vmxnet3: add v4 boot and guest UDP RSS config")
Cc: stable@dpdk.org
Signed-off-by: Eduard Serra <eserra@vmware.com>
Acked-by: Yong Wang <yongwang@vmware.com>
Requested promiscuous or all-multicast mode may be dropped on port
start if FW denies to enable it (e.g. because of no permission
in the case of VF). Return applied value on get.
Fixes: f5258439ee ("net/sfc: avoid failure on port start if Rx mode is rejected")
Cc: stable@dpdk.org
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Remove unnecessary variable "meta" in ice_flow_create and
ice_flow_validate, it should be defined when really needed:
its ice_parse_engine_create and ice_parse_engine_validate.
A meta data created by parser->parse_pattern_action should be
freed inside parser->engine->create.
During a validate operation, since parser->engine_>create will not
be invoked, it is unnecessary to create a meta during parser, so
NULL will passed to engine->parse_pattern_action and all parser's
parse_pattern_action need to be modified to handle meta = NULL
properly.
With above implementation, the patch also fixes a potential memory
leak in ice_parse_engine_validate, since meta may not be freed.
BTW, an engine without a create op should be regarded as a bug. So
use RTE_ASSERT to replace runtime engine->create == NULL check in
ice_parse_engine_create.
Fixes: 4e27d3ed02 ("net/ice: fix flow API framework")
Cc: stable@dpdk.org
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
This was left to its default value. With the patch transactions are:
- coherent,
- do not allocate in downstream cache (there is none on LS1028a),
- merge surrounding data for BD writes,
- overwrite surrounding data for frame data writes.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Gagandeep Singh <g.singh@nxp.com>
LS1028A does not have platform cache so any reads following a hardware
write will go directly to DDR. Latency of such a read is in excess of
100 core cycles, so try to prefetch more in advance to mitigate this.
How much is worth prefetching really depends on traffic conditions.
With congested Rx this could go up to 4 cache lines or so. But if
software keeps up with hardware and follows behind Rx PI by a cache line
then it's harmful in terms of performance to cache more. We would only
prefetch data that's yet to be written by ENETC, which will be evicted
again anyway.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Gagandeep Singh <g.singh@nxp.com>
Use rte_pktmbuf_free_bulk to release all mbufs at once. This is flagged
as obsolete/not yet stable in DPDK but seems to be functional.
Don't count the released frames, it's no longer needed in the caller.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Gagandeep Singh <g.singh@nxp.com>
Since we know in advance that we're going to fill in multiple
descriptors it's convenient to allocate the buffers in batches.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Gagandeep Singh <g.singh@nxp.com>
Reorder the members of the structure so that the ones used on datapath
fit in a single cache line, to slightly reduce pressure on cache and
miss rate.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Gagandeep Singh <g.singh@nxp.com>
Move from doing batch refill of Rx ring from bundles of 8 to once per
enetc_clean_rx_ring call. One benefit is that we're cleaning up all the
BDs that we just processed, which should still be cached. The other is
that hardware Rx index stays a little back and doesn't cause contention
on the BDs processed in the Rx loop.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Gagandeep Singh <g.singh@nxp.com>
The default value in hardware for the Rx MAC FIFO (@) is higher than it
should be and can lead to Rx lock-up under traffic.
Set it to the value recommended by hardware team, 1.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Signed-off-by: Gagandeep Singh <g.singh@nxp.com>
Each call to enetc_clean_tx_ring will cost at least 150-200 CPU cycles
even if no clean-up is done, due to the CI register read.
We're only calling it once at the end of the function, on the assumption
that software is slower than hardware and hardware completed sending
older frames out by now.
We're also cleaning up the ring before kicking off Tx for the new batch
to minimize chances of contention on the Tx ring.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Gagandeep Singh <g.singh@nxp.com>
We don't need barriers here since this read doesn't have to be strictly
serialized in relation to other surrounding memory/register accesses.
We only want a reasonably recent value out of hardware so we know how
much we can clean.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Gagandeep Singh <g.singh@nxp.com>
Don't read the hardware CI register in a loop, read it once, clean up
and exit.
The issue with reading the register in a loop is that we're stalling
here trying to catch up with hardware which keeps sending traffic as
long as it has traffic to send, so in effect we could be waiting here
for the Tx ring to be drained by hardware, instead of us doing Rx in
that meantime.
At the time we return the function there may be new BDs in the ring that
could be cleaned, we're just leaving those there for the next time.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Gagandeep Singh <g.singh@nxp.com>
Driver destroys the VNIC when the port is brought down.
Port HW filter setting such as default MAC address and
unicast MAC filters will be applied when port is started.
Fixed to return success silently for these callbacks
when port is stopped.
Fixes: 39b88344e3 ("net/bnxt: fix enable/disable VLAN filtering")
Cc: stable@dpdk.org
Reported-by: Stephen Hemminger <stephen@networkplumber.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>
There is a race condition when port is closed while error recovery
is happening in driver. Fixed this by canceling the recovery threads
during port close.
Fixes: df6cd7c1f7 ("net/bnxt: handle reset notify async event from FW")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Santoshkumar Karanappa Rastapur <santosh.rastapur@broadcom.com>
Commit 'dc36bd5dfdeb' mistakenly initializes ret as -rte_errno, then
return ret if ice_search_pattern_match_item fails, if rte_errno happens
to be 0 in this case, it would cause an unrecognized flow rule to be
marked as recognition success. Later, when the code tries to parse the
flow rule, a segmentation fault will occur due to null pointer
reference.
Fixes: dc36bd5dfd ("net/ice: fix flow FDIR/switch memory leak")
Cc: stable@dpdk.org
Signed-off-by: Tao Zhu <taox.zhu@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
Lack of GL_SWR_PM_UP_THR value for mac type 4x10G X722 caused it can't
reach line rate in one queue one core forwarding case. It's about 1%
performance drop. This patch adds value as:
Device-ID Value Comments
0x37D0 0x03030303 4x10G FPK
This value is hardware specific.
Fixes: 3320d4a240 ("net/i40e: workaround performance degradation")
Cc: stable@dpdk.org
Signed-off-by: Qiming Yang <qiming.yang@intel.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
Since qpairs is part of the vdev arguments, there is no need to
limit it to 16. The queue arrays can be dynamically sized based
on the requested parameters.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: John W. Linville <linville@tuxdriver.com>
The VLAN 16 bits tci field contains both values of PCP and VID. When
extracting any one of them - it is required not to affect the other one.
Previous to this commit in routine flow_dev_get_vlan_info_from_items()
we calculated the PCP as follows:
(1) vlan->vlan_tci &= MLX5DV_FLOW_VLAN_PCP_MASK;
(2) vlan->vlan_tci |= <3 bits value of PCP>
In line (1) we should have used the negated mask ('~' operator) such
that only the PCP bits will be nullified before ORing them with the
updated PCP value.
Fixes: 9aee7a8418 ("net/mlx5: support push flow action on VLAN header")
Cc: stable@dpdk.org
Signed-off-by: Ophir Munk <ophirmu@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
If configuring VF promiscuous mode is not supported, return '-ENOTSUP'
error code in '.promiscuous_enable/disable' dev_ops.
This is to fix the port start during config restore, where if
'.promiscuous_enable/disable' dev_ops exists and return any value other
than '-ENOTSUP', start will fail.
Same is done for '.allmulticast_enable/disable' dev_ops.
Fixes: ddc7cb0d94 ("net/i40e: re-program promiscuous mode on VF interface")
Cc: stable@dpdk.org
Signed-off-by: Xiao Zhang <xiao.zhang@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
Add support for .set_link_up/down() eth ops to bring
link up and down.
Signed-off-by: Harman Kalra <hkalra@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Loopback devices are exposed as ethdev device in octeontx2.
This patch fixes the link information for loopback ports.
Fixes: 38f566280a ("net/octeontx2: add link stats operations")
Cc: stable@dpdk.org
Signed-off-by: Ashish Gupta <ashishg@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Since packet is talking to a network device (probably tap)
it is possible for the device driver to change the hardware address
by making an ioctl to kernel.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
The devices of the family ConnectX may have two letters as suffix.
Such suffix is preceded with a space and the second x is lowercase:
- ConnectX-4 Lx
- ConnectX-5 Ex
- ConnectX-6 Dx
Uppercase of the device family name BlueField is also fixed.
The lists of supported devices are fixed.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
The routine sending packets with Multi-Packet Write method assigns
the wqe_last variable with transmit descriptor (WQE - work queue entry)
being built. If send queue is close to full state, the WQE has no data
yet (trying to put the first packet) and there is no enough space
in descriptor for the next packet the WQE is discarded and the stored
wqe_last value becomes invalid - points to the discarded WQE.
The mlx5_tx_burst_request_completion() routine might set the completion
request flags in the WQE pointed by wqe_last, it is safe, but the next
mlx5_tx_burst call uses the WQE as the first free one and request
completion flags might be overwritten and completion request will be
lost causing the transmit datapath malfunction.
Fixes: 8b581c690a ("net/mlx5: move Tx complete request routine")
Cc: stable@dpdk.org
Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
In order to move the mbuf metadata from the WQE to the FDB steering
domain, the PMD add for each NIC TX flow a new action to copy the
metadata register REG_A to REG_C_0.
This copy action is considered as modify header action from HW
perspective.
The HW doesn't support to do modify header action after ant
encapsulation action.
The split metadata function wrongly added the copy action in the end of
the original actions list, hence, NIC egress flow with encapsulation
action failed when the PMD worked with dv_xmeta_en mode.
Move the copy action to be before and back to back with the
encapsulation action for the aforementioned case.
Fixes: 71e254bc02 ("net/mlx5: split Rx flows to provide metadata copy")
Cc: stable@dpdk.org
Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
When running mlx5_dev_start in mlx5_ethdev the function calls
mlx5_dev_configure_rss_reta in order to configure the rxq's.
Before mlx5_dev_configure_rss_reta there isn't a check whether
there are rxq's and if rxq's are 0 the function fails.
For example, this command:
/build/app/test-pmd/testpmd -n 4 -w 0000:08:00.0,rx_vec_en=0
-- --burst=64 --mbcache=512 -i --nb-cores=27 --txd=2048 --rxd=2048
--vxlan-gpe-port=6081 --mp-alloc=xbuf --rxq 0 --forward-mode=txonly
would fail.
In order to fix this issue, we should call mlx5_dev_configure_rss_reta
only if we have rxq's.
Fixes: 63bd16292c ("net/mlx5: support RSS on hairpin")
Cc: stable@dpdk.org
Reported-by: Sylvain Rodon <srn@nexatech.fr>
Signed-off-by: Shiri Kuzin <shirik@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
HW supports match on one Ethertype, the Ethertype following the last
VLAN tag of the packet (see PRM).
Previous patch added specific handling for packets with VLAN tag,
after setting match on Ethertype.
This patch moves the handling of packets with VLAN tag, to be done
before and instead of setting match on Ethertype.
Previous patch also added, as part of specific handling for packets
with VLAN tag, the setting of cvlan_tag mask bit in translation of
L3 items.
In case of L3 tunnel there is no inner L2 header, so setting this
mask bit is wrong and causes match failures.
This patch adds check to make sure L2 header exists before setting
cvlan_tag mask bit for L3 items.
Fixes: 00f75a4057 ("net/mlx5: fix VLAN match for DV mode")
Cc: stable@dpdk.org
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Tested-by: Raslan Darawsheh <rasland@mellanox.com>
Meter suffix subflow only has the port id and tag match item, if VLAN
push and set VLAN id actions exist in the suffix subflow, the user
defined VLAN items is required for the actions to set a correct VLAN
id.
Currently, the VLAN item stays in the meter prefix subflow. Without
the VLAN item, VLAN id or pcp will not be inherited.
Actions require the VLAN item as below:
RTE_FLOW_ACTION_TYPE_OF_PUSH_VLAN
RTE_FLOW_ACTION_TYPE_OF_SET_VLAN_VID
Add a private VLAN item to copy the user defined VLAN item to the meter
suffix subflow, so the suffix subflow will have the chance to get the
correct original VLAN id and pcp value from the VLAN item.
Fixes: 9ea9b049a9 ("net/mlx5: split meter flow")
Cc: stable@dpdk.org
Signed-off-by: Suanming Mou <suanmingm@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
As meter flows are split into three subflows each, the prefix subflow
with meter action color the packet, the meter subflow filters out the
colored packets, the suffix subflow applies all the remaining actions
to the passed packets.
Currently, all the user defined items are matched in the prefix flow.
Flow id tag match item is the only item added to the meter suffix
subflow. Some of the remaining actions to be applied in the suffix
subflow require more information in the match item, or the suffix
subflow will not be created successfully.
Actions require the L3/L4 type in the match items as below:
RTE_FLOW_ACTION_TYPE_SET_TP_SRC
RTE_FLOW_ACTION_TYPE_SET_TP_DST
RTE_FLOW_ACTION_TYPE_DEC_TTL
RTE_FLOW_ACTION_TYPE_SET_TTL
RTE_FLOW_ACTION_TYPE_RSS
RTE_FLOW_ACTION_TYPE_QUEUE
Inherit the match item flags from meter prefix subflow to make actions
in suffix subflow get sufficient information.
Fixes: 9ea9b049a9 ("net/mlx5: split meter flow")
Cc: stable@dpdk.org
Signed-off-by: Suanming Mou <suanmingm@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Metadata suffix subflow inherits the RSS needed hash_fields from the
prefix subflow as the suffix subflow only has the tag match item unable
to generate the full original hash_fields for RSS action.
Unfortunately, hash_fields will only be generated if flow has RSS
action. So it means the prefix flow won't generate the hash_fields as
the RSS action has been split to the suffix flow.
Copy the layer flags from prefix subflow to suffix subflow to help the
suffix subflow to generate the correct hash_fields itself.
Fixes: 71e254bc02 ("net/mlx5: split Rx flows to provide metadata copy")
Cc: stable@dpdk.org
Signed-off-by: Suanming Mou <suanmingm@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Currently, for header modify actions after tunnel decapsulation, the
header modify actions will still select the old outermost layer type.
It will cause header modify actions get the wrong layer type.
Add the tunnel decap flag to indicate the layer type to get for header
modify actions.
Fixes: 4bb14c83df ("net/mlx5: support modify header using Direct Verbs")
Cc: stable@dpdk.org
Signed-off-by: Suanming Mou <suanmingm@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Currently, the flow validate function only validate the outermost layer
with the header modify actions. If there is decapsulation action before
the header modify action, the validation should choose the inner layer
for validation.
Add decapsulation check when validate with the header modify actions.
Choose the inner layer once there is decapsulation action.
Fixes: 4bb14c83df ("net/mlx5: support modify header using Direct Verbs")
Cc: stable@dpdk.org
Signed-off-by: Suanming Mou <suanmingm@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Currently, the set VLAN ID header modify action has already got the
action pointer with offset from the header modify action array, but
the configuration saves the detail to the memory of action with the
offset again. It causes double offset to set the VLAN ID action to
the wrong place in the header modify array.
Remove the offset when get the action pointer to fix that issue.
Fixes: 5f163d520c ("net/mlx5: support modify VLAN ID on existing VLAN header")
Cc: stable@dpdk.org
Signed-off-by: Suanming Mou <suanmingm@mellanox.com>
Acked-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Fix a race condition between port stop and error recovery task. When io
forwarding is not started on Stingray devices, driver will not receive
the async event from FW when there is a FW reset. While exiting testpmd,
as part of port stop driver sees this event and this in turn causes a
race between port stop and error recovery task.
Fixed this by ignoring the fatal/non-fatal async event from FW while
stopping port.
Fixes: df6cd7c1f7 ("net/bnxt: handle reset notify async event from FW")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Rahul Gupta <rahul.gupta@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Check for bp->hwrm_cmd_resp_addr before using it in HWRM_PREP to avoid
segmentation fault when stop port and meanwhile receive events from FW.
Fixes: df6cd7c1f7 ("net/bnxt: handle reset notify async event from FW")
Cc: stable@dpdk.org
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Operator '<<' should be used instead of '<' for shifting value to be
set in MSI-X configuration register.
Old code adds 1 on even MSI-X interrupt vector index used by queue,
resulting in interrupt mapping error.
Fixes: 65dfc889d8 ("net/ice: support Rx queue interruption")
Cc: stable@dpdk.org
Signed-off-by: Tao Zhu <taox.zhu@intel.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
'fm10k_rxtx_vec.c' is SSE vector instructions implementation and should
be compiled only for x86.
The vector PMD support controlled by
'CONFIG_RTE_LIBRTE_FM10K_INC_VECTOR' config option, but it is enabled by
default, safer to add x86 checks too.
Fixes: 10eb9ce8dc ("fm10k: allow to disable vector driver")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Xiao Wang <xiao.w.wang@intel.com>
The format specifier for the 'size_t' format should be '%z'.
Also this fix enables compiling PMD for 32bit architecture.
Fixes: ba641f2076 ("net/sfc: add init on attach")
Fixes: 82faef5076 ("net/sfc: set RSS key and hash types config")
Fixes: 7803554a9e ("net/sfc: process RSS settings on Rx configure step")
Cc: stable@dpdk.org
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
The RSS expansion feature was introduced to split RSS flows according to
the adjustment between the RSS types and the flow items.
The expansion function gets an item tree for the above adjustment from
the caller which reflects the HW needs.
The standard vxlan header next protocol is always Ethernet while there
are some Mellanox customers who use their own method to allow L3
headers after the vxlan tunnel header.
The expansion tree of mlx5 PMD didn't expect to get L3 headers after the
vxlan header what caused a failure in flow creation when inner RSS is
requested on L3 after vxlan flow.
Add IPV4 and IPV6 as optional headers after vxlan in the RSS expansion
tree to allow L3 tunnel support for vxlan.
Fixes: f4f06e3615 ("net/mlx5: add flow VXLAN item")
Cc: stable@dpdk.org
Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
The hairpin TX/RX queue depth and packet size is fixed in the past.
When the firmware has some fix or improvement, the PMD will not
make full use of it. And also, 32 packets for a single queue will not
guarantee a good performance for hairpin flows. It will make the
stride size larger and for small packets, it is a waste of memory.
The recommended stride size is 64B now.
The parameter of hairpin queue setup needs to be adjusted.
1. A proper buffer size should support the standard jumbo frame with
9KB, and also more than 1 jumbo frame packet for performance.
2. Number of packets of a single queue should be the maximum
supported value (total buffer size / stride size).
There is no need to support the max capacity of total buffer size
because the memory consumption should also be taken into
consideration.
Fixes: e79c9be915 ("net/mlx5: support Rx hairpin queues")
Cc: stable@dpdk.org
Signed-off-by: Bing Zhao <bingz@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
The new Mellanox devices support GENEVE tunnel stateless offloads by
native HW parsers, FW doesn't need to configure flex parser to support
it.
The FW capability for GENEVE flex parser is not set for the above
devices, hence the GENEVE flow validation failed for them.
Remove the flex parser validation for GENEVE, stay only with the direct
FW capability for GENEVE support.
Fixes: e59a5dbcfd ("net/mlx5: add flow match on GENEVE item")
Cc: stable@dpdk.org
Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
The RTE_FLOW_ITEM_TYPE_GRE_KEY means that GRE key should
be present on the flow and in addition can explicitly match on the GRE
key field itself.
When no value is set in this item any GRE key should match the flow.
The match on GRE key wrongly was not added if the item fields are NULL
while at least the GRE key should be present in the flow.
Add the GRE key present match when RTE_FLOW_ITEM_TYPE_GRE_KEY is in
the flow item list.
Fixes: a7a0365565 ("net/mlx5: match GRE key and present bits")
Cc: stable@dpdk.org
Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
ICMPv6 header contains the checksum field, that includes the IP
addresses from IP header.
The ConnectX-5 hardware does not support ICMPv6 checksum calculation
and in case of IP header modification the checksum will contain
the wrong value.
The patch rejects the header modification actions related to IP
addresses if ICMPv6 protocol is detected in pattern.
This issue was also partly fixed in commit below which rejects
flows with the pattern "ipv6 proto is 58".
Fixes: 0e2ceb639d ("net/mlx5: fix ICMPv6 header rewrite actions")
Cc: stable@dpdk.org
Signed-off-by: Shiri Kuzin <shirik@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
Changed the ABI version to 20.0.1.
Fixes: a72cf98cc9 ("net/ionic: add skeleton")
Reported-by: David Marchand <david.marchand@redhat.com>
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Ethdev's .dev_configure callback can be called multiple
time during a device life-time, but Vhost makes the
wrong assumption that it is not the case and try to
setup again the device on reconfiguration.
This patch ensures the device hasn't been already setup
before proceeding.
Fixes: 3d01b759d2 ("net/vhost: delay driver setup")
Cc: stable@dpdk.org
Reported-by: Yinan Wang <yinan.wang@intel.com>
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Tested-by: Yinan Wang <yinan.wang@intel.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
If for some reason vhost_driver_setup() fails, the list
element for the device may be freed without being removed
from the internal list of devices.
This patch fixes all the error paths, by unregistering the
device from Vhost library it has been registered, remove
the device from the list, reset device vring_state pointer
from the global table and only free vring state if it had
been allocated.
Fixes: 3d01b759d2 ("net/vhost: delay driver setup")
Cc: stable@dpdk.org
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
When the port restarts, the flow ctrl register will be reset,
we need to make sure it can be configured the same as previous setting,
otherwise a register read error would occur. This patch fixes this
issue.
Fixes: af75078fec ("first public release")
Cc: stable@dpdk.org
Signed-off-by: Guinan Sun <guinanx.sun@intel.com>
Tested-by: Tomasz Konieczny <tomaszx.konieczny@intel.com>
Reviewed-by: Xiaolong Ye <xiaolong.ye@intel.com>
Interrupt mapping should be 1:n queue(s).This patch fixes the
logic of interrupt bind by code reconstruction.
Fixes: 6a6cf5f88b ("net/i40e: enable multi-queue Rx interrupt for VF")
Signed-off-by: Shougang Wang <shougangx.wang@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
The hardware can recognize and mark the layer 4 protocol type for TCP,
UDP and IPSec non-fragmented packets. For all the fragmented packets,
L4 type will be considered as None. This can be used when creating a
flow with L4 matching, then hops number will be reduced and a better
performance could be gained.
But for ICMP packets, it cannot be recognized correctly because it is
not a L4 protocol in the stack, even if the packet format is similar.
All the fragmented and non-fragmented ICMP will have the None L4 type.
Fragmented packets with incomplete headers could not hit the flow,
even for the first fragment. Because then it will make it complex to
defragment for both HW and SW. For other types, the implicit rules
could be used directly and all the fragments will miss the flow.
For ICMP packets, this should be done explicitly because all packets
have None type. The first fragment will still hit the flow if there is
no explicit rule.
All ICMP fragments will still hit the rules like ETH, ETH + IP, and
ETH + IP + "ICMP protocol" only since they are wildcard rules, and
there is no next layer protocol specified field in such rules.
Fixes: d53aa89aea ("net/mlx5: support matching on ICMP/ICMP6")
Cc: stable@dpdk.org
Signed-off-by: Bing Zhao <bingz@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
This adds new device id to the list of Mellanox devices
that runs mlx5 PMD.
- BlueField-2 integrated ConnectX-6 Dx network controller
This device is not ready yet, it is in development stage.
Signed-off-by: Raslan Darawsheh <rasland@mellanox.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
Acked-by: Matan Azrad <matan@mellanox.com>
The PMD manages internally the priority of the flows in addition to the
user configured priority.
So, 2 flows with the same user priority may get different priority.
The method:
As much as the flow is more specific it gets higher priority
(higher means first to be matched).
In addition, When the user creates a RSS flow the PMD splits the flows
according to the flow RSS layers as the HW requests for RSS TIR.
The internal priority for each flow is decided by the flow last layer.
L2, L3 and L4 (L2 low and L4 high).
The tunnel layer was wrongly decided to be L4 all the time, even when
the flow is configured with inner-RSS.
Hence, the first RSS split which takes the tunnel layer priority all the
time will be matched before the more specific splits.
Change the priority of tunnel layer to be L2 when inner-RSS is
configured.
Fixes: d4a405186b ("net/mlx5: support tunnel RSS level")
Cc: stable@dpdk.org
Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Ori Kam <orika@mellanox.com>
For ixgbe, there is restriction that data buffers of any transmitted
packet must include at least 12 bytes of the src/dst Ethernet MAC
addresses as well as 2 bytes of the Type/Len field, otherwise, tx hang
would happen.
This patch adds check for those illegal packets and protects TX from
hanging.
Fixes: 7829b8d52b ("net/ixgbe: add Tx preparation")
Cc: stable@dpdk.org
Signed-off-by: Xiao Zhang <xiao.zhang@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
Stopping an already disabled vport leads to firmware
assert. Stop the vport only if it was started.
Fixes: 2ea6f76aff ("qede: add core driver")
Cc: stable@dpdk.org
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: Rasesh Mody <rmody@marvell.com>
