1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright 2015 6WIND S.A.
3 Copyright 2015 Mellanox Technologies, Ltd
8 The MLX5 poll mode driver library (**librte_pmd_mlx5**) provides support
9 for **Mellanox ConnectX-4**, **Mellanox ConnectX-4 Lx** and **Mellanox
10 ConnectX-5** families of 10/25/40/50/100 Gb/s adapters as well as their
11 virtual functions (VF) in SR-IOV context.
13 Information and documentation about these adapters can be found on the
14 `Mellanox website <http://www.mellanox.com>`__. Help is also provided by the
15 `Mellanox community <http://community.mellanox.com/welcome>`__.
17 There is also a `section dedicated to this poll mode driver
18 <http://www.mellanox.com/page/products_dyn?product_family=209&mtag=pmd_for_dpdk>`__.
22 Due to external dependencies, this driver is disabled by default. It must
23 be enabled manually by setting ``CONFIG_RTE_LIBRTE_MLX5_PMD=y`` and
26 Implementation details
27 ----------------------
29 Besides its dependency on libibverbs (that implies libmlx5 and associated
30 kernel support), librte_pmd_mlx5 relies heavily on system calls for control
31 operations such as querying/updating the MTU and flow control parameters.
33 For security reasons and robustness, this driver only deals with virtual
34 memory addresses. The way resources allocations are handled by the kernel
35 combined with hardware specifications that allow it to handle virtual memory
36 addresses directly ensure that DPDK applications cannot access random
37 physical memory (or memory that does not belong to the current process).
39 This capability allows the PMD to coexist with kernel network interfaces
40 which remain functional, although they stop receiving unicast packets as
41 long as they share the same MAC address.
42 This means legacy linux control tools (for example: ethtool, ifconfig and
43 more) can operate on the same network interfaces that owned by the DPDK
46 Enabling librte_pmd_mlx5 causes DPDK applications to be linked against
52 - Multi arch support: x86_64, POWER8, ARMv8.
53 - Multiple TX and RX queues.
54 - Support for scattered TX and RX frames.
55 - IPv4, IPv6, TCPv4, TCPv6, UDPv4 and UDPv6 RSS on any number of queues.
56 - Several RSS hash keys, one for each flow type.
57 - Configurable RETA table.
58 - Support for multiple MAC addresses.
62 - RX CRC stripping configuration.
64 - Multicast promiscuous mode.
65 - Hardware checksum offloads.
66 - Flow director (RTE_FDIR_MODE_PERFECT, RTE_FDIR_MODE_PERFECT_MAC_VLAN and
70 - KVM and VMware ESX SR-IOV modes are supported.
71 - RSS hash result is supported.
73 - Hardware checksum TX offload for VXLAN and GRE.
75 - Statistics query including Basic, Extended and per queue.
81 - Inner RSS for VXLAN frames is not supported yet.
82 - Hardware checksum RX offloads for VXLAN inner header are not supported yet.
83 - For secondary process:
85 - Forked secondary process not supported.
86 - All mempools must be initialized before rte_eth_dev_start().
88 - Flow pattern without any specific vlan will match for vlan packets as well:
90 When VLAN spec is not specified in the pattern, the matching rule will be created with VLAN as a wild card.
91 Meaning, the flow rule::
93 flow create 0 ingress pattern eth / vlan vid is 3 / ipv4 / end ...
95 Will only match vlan packets with vid=3. and the flow rules::
97 flow create 0 ingress pattern eth / ipv4 / end ...
101 flow create 0 ingress pattern eth / vlan / ipv4 / end ...
103 Will match any ipv4 packet (VLAN included).
105 - A multi segment packet must have less than 6 segments in case the Tx burst function
106 is set to multi-packet send or Enhanced multi-packet send. Otherwise it must have
107 less than 50 segments.
108 - Count action for RTE flow is **only supported in Mellanox OFED**.
109 - Flows with a VXLAN Network Identifier equal (or ends to be equal)
110 to 0 are not supported.
111 - VXLAN TSO and checksum offloads are not supported on VM.
112 - VF: flow rules created on VF devices can only match traffic targeted at the
113 configured MAC addresses (see ``rte_eth_dev_mac_addr_add()``).
117 MAC addresses not already present in the bridge table of the associated
118 kernel network device will be added and cleaned up by the PMD when closing
119 the device. In case of ungraceful program termination, some entries may
120 remain present and should be removed manually by other means.
125 MLX5 supports various of methods to report statistics:
127 Port statistics can be queried using ``rte_eth_stats_get()``. The port statistics are through SW only and counts the number of packets received or sent successfully by the PMD.
129 Extended statistics can be queried using ``rte_eth_xstats_get()``. The extended statistics expose a wider set of counters counted by the device. The extended port statistics counts the number of packets received or sent successfully by the port. As Mellanox NICs are using the :ref:`Bifurcated Linux Driver <linux_gsg_linux_drivers>` those counters counts also packet received or sent by the Linux kernel. The counters with ``_phy`` suffix counts the total events on the physical port, therefore not valid for VF.
131 Finally per-flow statistics can by queried using ``rte_flow_query`` when attaching a count action for specific flow. The flow counter counts the number of packets received successfully by the port and match the specific flow.
139 These options can be modified in the ``.config`` file.
141 - ``CONFIG_RTE_LIBRTE_MLX5_PMD`` (default **n**)
143 Toggle compilation of librte_pmd_mlx5 itself.
145 - ``CONFIG_RTE_LIBRTE_MLX5_DLOPEN_DEPS`` (default **n**)
147 Build PMD with additional code to make it loadable without hard
148 dependencies on **libibverbs** nor **libmlx5**, which may not be installed
149 on the target system.
151 In this mode, their presence is still required for it to run properly,
152 however their absence won't prevent a DPDK application from starting (with
153 ``CONFIG_RTE_BUILD_SHARED_LIB`` disabled) and they won't show up as
154 missing with ``ldd(1)``.
156 It works by moving these dependencies to a purpose-built rdma-core "glue"
157 plug-in which must either be installed in a directory whose name is based
158 on ``CONFIG_RTE_EAL_PMD_PATH`` suffixed with ``-glue`` if set, or in a
159 standard location for the dynamic linker (e.g. ``/lib``) if left to the
160 default empty string (``""``).
162 This option has no performance impact.
164 - ``CONFIG_RTE_LIBRTE_MLX5_DEBUG`` (default **n**)
166 Toggle debugging code and stricter compilation flags. Enabling this option
167 adds additional run-time checks and debugging messages at the cost of
170 - ``CONFIG_RTE_LIBRTE_MLX5_TX_MP_CACHE`` (default **8**)
172 Maximum number of cached memory pools (MPs) per TX queue. Each MP from
173 which buffers are to be transmitted must be associated to memory regions
174 (MRs). This is a slow operation that must be cached.
176 This value is always 1 for RX queues since they use a single MP.
178 Environment variables
179 ~~~~~~~~~~~~~~~~~~~~~
183 A list of directories in which to search for the rdma-core "glue" plug-in,
184 separated by colons or semi-colons.
186 Only matters when compiled with ``CONFIG_RTE_LIBRTE_MLX5_DLOPEN_DEPS``
187 enabled and most useful when ``CONFIG_RTE_EAL_PMD_PATH`` is also set,
188 since ``LD_LIBRARY_PATH`` has no effect in this case.
190 - ``MLX5_PMD_ENABLE_PADDING``
192 Enables HW packet padding in PCI bus transactions.
194 When packet size is cache aligned and CRC stripping is enabled, 4 fewer
195 bytes are written to the PCI bus. Enabling padding makes such packets
198 In cases where PCI bandwidth is the bottleneck, padding can improve
201 This is disabled by default since this can also decrease performance for
202 unaligned packet sizes.
204 - ``MLX5_SHUT_UP_BF``
206 Configures HW Tx doorbell register as IO-mapped.
208 By default, the HW Tx doorbell is configured as a write-combining register.
209 The register would be flushed to HW usually when the write-combining buffer
210 becomes full, but it depends on CPU design.
212 Except for vectorized Tx burst routines, a write memory barrier is enforced
213 after updating the register so that the update can be immediately visible to
216 When vectorized Tx burst is called, the barrier is set only if the burst size
217 is not aligned to MLX5_VPMD_TX_MAX_BURST. However, setting this environmental
218 variable will bring better latency even though the maximum throughput can
221 Run-time configuration
222 ~~~~~~~~~~~~~~~~~~~~~~
224 - librte_pmd_mlx5 brings kernel network interfaces up during initialization
225 because it is affected by their state. Forcing them down prevents packets
228 - **ethtool** operations on related kernel interfaces also affect the PMD.
230 - ``rxq_cqe_comp_en`` parameter [int]
232 A nonzero value enables the compression of CQE on RX side. This feature
233 allows to save PCI bandwidth and improve performance. Enabled by default.
237 - x86_64 with ConnectX-4, ConnectX-4 LX and ConnectX-5.
238 - POWER8 and ARMv8 with ConnectX-4 LX and ConnectX-5.
240 - ``txq_inline`` parameter [int]
242 Amount of data to be inlined during TX operations. Improves latency.
243 Can improve PPS performance when PCI back pressure is detected and may be
244 useful for scenarios involving heavy traffic on many queues.
246 Because additional software logic is necessary to handle this mode, this
247 option should be used with care, as it can lower performance when back
248 pressure is not expected.
250 - ``txqs_min_inline`` parameter [int]
252 Enable inline send only when the number of TX queues is greater or equal
255 This option should be used in combination with ``txq_inline`` above.
257 On ConnectX-4, ConnectX-4 LX and ConnectX-5 without Enhanced MPW:
259 - Disabled by default.
260 - In case ``txq_inline`` is set recommendation is 4.
262 On ConnectX-5 with Enhanced MPW:
264 - Set to 8 by default.
266 - ``txq_mpw_en`` parameter [int]
268 A nonzero value enables multi-packet send (MPS) for ConnectX-4 Lx and
269 enhanced multi-packet send (Enhanced MPS) for ConnectX-5. MPS allows the
270 TX burst function to pack up multiple packets in a single descriptor
271 session in order to save PCI bandwidth and improve performance at the
272 cost of a slightly higher CPU usage. When ``txq_inline`` is set along
273 with ``txq_mpw_en``, TX burst function tries to copy entire packet data
274 on to TX descriptor instead of including pointer of packet only if there
275 is enough room remained in the descriptor. ``txq_inline`` sets
276 per-descriptor space for either pointers or inlined packets. In addition,
277 Enhanced MPS supports hybrid mode - mixing inlined packets and pointers
278 in the same descriptor.
280 This option cannot be used with certain offloads such as ``DEV_TX_OFFLOAD_TCP_TSO,
281 DEV_TX_OFFLOAD_VXLAN_TNL_TSO, DEV_TX_OFFLOAD_GRE_TNL_TSO, DEV_TX_OFFLOAD_VLAN_INSERT``.
282 When those offloads are requested the MPS send function will not be used.
284 It is currently only supported on the ConnectX-4 Lx and ConnectX-5
285 families of adapters. Enabled by default.
287 - ``txq_mpw_hdr_dseg_en`` parameter [int]
289 A nonzero value enables including two pointers in the first block of TX
290 descriptor. This can be used to lessen CPU load for memory copy.
292 Effective only when Enhanced MPS is supported. Disabled by default.
294 - ``txq_max_inline_len`` parameter [int]
296 Maximum size of packet to be inlined. This limits the size of packet to
297 be inlined. If the size of a packet is larger than configured value, the
298 packet isn't inlined even though there's enough space remained in the
299 descriptor. Instead, the packet is included with pointer.
301 Effective only when Enhanced MPS is supported. The default value is 256.
303 - ``tx_vec_en`` parameter [int]
305 A nonzero value enables Tx vector on ConnectX-5 only NIC if the number of
306 global Tx queues on the port is lesser than MLX5_VPMD_MIN_TXQS.
308 This option cannot be used with certain offloads such as ``DEV_TX_OFFLOAD_TCP_TSO,
309 DEV_TX_OFFLOAD_VXLAN_TNL_TSO, DEV_TX_OFFLOAD_GRE_TNL_TSO, DEV_TX_OFFLOAD_VLAN_INSERT``.
310 When those offloads are requested the MPS send function will not be used.
312 Enabled by default on ConnectX-5.
314 - ``rx_vec_en`` parameter [int]
316 A nonzero value enables Rx vector if the port is not configured in
317 multi-segment otherwise this parameter is ignored.
321 - ``vf_nl_en`` parameter [int]
323 A nonzero value enables Netlink requests from the VF to add/remove MAC
324 addresses or/and enable/disable promiscuous/all multicast on the Netdevice.
325 Otherwise the relevant configuration must be run with Linux iproute2 tools.
326 This is a prerequisite to receive this kind of traffic.
328 Enabled by default, valid only on VF devices ignored otherwise.
333 This driver relies on external libraries and kernel drivers for resources
334 allocations and initialization. The following dependencies are not part of
335 DPDK and must be installed separately:
339 User space Verbs framework used by librte_pmd_mlx5. This library provides
340 a generic interface between the kernel and low-level user space drivers
343 It allows slow and privileged operations (context initialization, hardware
344 resources allocations) to be managed by the kernel and fast operations to
345 never leave user space.
349 Low-level user space driver library for Mellanox ConnectX-4/ConnectX-5
350 devices, it is automatically loaded by libibverbs.
352 This library basically implements send/receive calls to the hardware
357 They provide the kernel-side Verbs API and low level device drivers that
358 manage actual hardware initialization and resources sharing with user
361 Unlike most other PMDs, these modules must remain loaded and bound to
364 - mlx5_core: hardware driver managing Mellanox ConnectX-4/ConnectX-5
365 devices and related Ethernet kernel network devices.
366 - mlx5_ib: InifiniBand device driver.
367 - ib_uverbs: user space driver for Verbs (entry point for libibverbs).
369 - **Firmware update**
371 Mellanox OFED releases include firmware updates for ConnectX-4/ConnectX-5
374 Because each release provides new features, these updates must be applied to
375 match the kernel modules and libraries they come with.
379 Both libraries are BSD and GPL licensed. Linux kernel modules are GPL
385 Either RDMA Core library with a recent enough Linux kernel release
386 (recommended) or Mellanox OFED, which provides compatibility with older
389 RMDA Core with Linux Kernel
390 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
392 - Minimal kernel version : v4.14 or the most recent 4.14-rc (see `Linux installation documentation`_)
393 - Minimal rdma-core version: v15+ commit 0c5f5765213a ("Merge pull request #227 from yishaih/tm")
394 (see `RDMA Core installation documentation`_)
396 .. _`Linux installation documentation`: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git/plain/Documentation/admin-guide/README.rst
397 .. _`RDMA Core installation documentation`: https://raw.githubusercontent.com/linux-rdma/rdma-core/master/README.md
402 - Mellanox OFED version: **4.2, 4.3**.
405 - ConnectX-4: **12.21.1000** and above.
406 - ConnectX-4 Lx: **14.21.1000** and above.
407 - ConnectX-5: **16.21.1000** and above.
408 - ConnectX-5 Ex: **16.21.1000** and above.
410 While these libraries and kernel modules are available on OpenFabrics
411 Alliance's `website <https://www.openfabrics.org/>`__ and provided by package
412 managers on most distributions, this PMD requires Ethernet extensions that
413 may not be supported at the moment (this is a work in progress).
416 <http://www.mellanox.com/page/products_dyn?product_family=26&mtag=linux>`__
417 includes the necessary support and should be used in the meantime. For DPDK,
418 only libibverbs, libmlx5, mlnx-ofed-kernel packages and firmware updates are
419 required from that distribution.
423 Several versions of Mellanox OFED are available. Installing the version
424 this DPDK release was developed and tested against is strongly
425 recommended. Please check the `prerequisites`_.
430 * Mellanox(R) ConnectX(R)-4 10G MCX4111A-XCAT (1x10G)
431 * Mellanox(R) ConnectX(R)-4 10G MCX4121A-XCAT (2x10G)
432 * Mellanox(R) ConnectX(R)-4 25G MCX4111A-ACAT (1x25G)
433 * Mellanox(R) ConnectX(R)-4 25G MCX4121A-ACAT (2x25G)
434 * Mellanox(R) ConnectX(R)-4 40G MCX4131A-BCAT (1x40G)
435 * Mellanox(R) ConnectX(R)-4 40G MCX413A-BCAT (1x40G)
436 * Mellanox(R) ConnectX(R)-4 40G MCX415A-BCAT (1x40G)
437 * Mellanox(R) ConnectX(R)-4 50G MCX4131A-GCAT (1x50G)
438 * Mellanox(R) ConnectX(R)-4 50G MCX413A-GCAT (1x50G)
439 * Mellanox(R) ConnectX(R)-4 50G MCX414A-BCAT (2x50G)
440 * Mellanox(R) ConnectX(R)-4 50G MCX415A-GCAT (2x50G)
441 * Mellanox(R) ConnectX(R)-4 50G MCX416A-BCAT (2x50G)
442 * Mellanox(R) ConnectX(R)-4 50G MCX416A-GCAT (2x50G)
443 * Mellanox(R) ConnectX(R)-4 50G MCX415A-CCAT (1x100G)
444 * Mellanox(R) ConnectX(R)-4 100G MCX416A-CCAT (2x100G)
445 * Mellanox(R) ConnectX(R)-4 Lx 10G MCX4121A-XCAT (2x10G)
446 * Mellanox(R) ConnectX(R)-4 Lx 25G MCX4121A-ACAT (2x25G)
447 * Mellanox(R) ConnectX(R)-5 100G MCX556A-ECAT (2x100G)
448 * Mellanox(R) ConnectX(R)-5 Ex EN 100G MCX516A-CDAT (2x100G)
450 Quick Start Guide on OFED
451 -------------------------
453 1. Download latest Mellanox OFED. For more info check the `prerequisites`_.
456 2. Install the required libraries and kernel modules either by installing
457 only the required set, or by installing the entire Mellanox OFED:
459 .. code-block:: console
461 ./mlnxofedinstall --upstream-libs --dpdk
463 3. Verify the firmware is the correct one:
465 .. code-block:: console
469 4. Verify all ports links are set to Ethernet:
471 .. code-block:: console
473 mlxconfig -d <mst device> query | grep LINK_TYPE
477 Link types may have to be configured to Ethernet:
479 .. code-block:: console
481 mlxconfig -d <mst device> set LINK_TYPE_P1/2=1/2/3
483 * LINK_TYPE_P1=<1|2|3> , 1=Infiniband 2=Ethernet 3=VPI(auto-sense)
485 For hypervisors verify SR-IOV is enabled on the NIC:
487 .. code-block:: console
489 mlxconfig -d <mst device> query | grep SRIOV_EN
492 If needed, set enable the set the relevant fields:
494 .. code-block:: console
496 mlxconfig -d <mst device> set SRIOV_EN=1 NUM_OF_VFS=16
497 mlxfwreset -d <mst device> reset
499 5. Restart the driver:
501 .. code-block:: console
503 /etc/init.d/openibd restart
507 .. code-block:: console
509 service openibd restart
511 If link type was changed, firmware must be reset as well:
513 .. code-block:: console
515 mlxfwreset -d <mst device> reset
517 For hypervisors, after reset write the sysfs number of virtual functions
520 To dynamically instantiate a given number of virtual functions (VFs):
522 .. code-block:: console
524 echo [num_vfs] > /sys/class/infiniband/mlx5_0/device/sriov_numvfs
526 6. Compile DPDK and you are ready to go. See instructions on
527 :ref:`Development Kit Build System <Development_Kit_Build_System>`
532 1. Configure aggressive CQE Zipping for maximum performance:
534 .. code-block:: console
536 mlxconfig -d <mst device> s CQE_COMPRESSION=1
538 To set it back to the default CQE Zipping mode use:
540 .. code-block:: console
542 mlxconfig -d <mst device> s CQE_COMPRESSION=0
544 2. In case of virtualization:
546 - Make sure that hypervisor kernel is 3.16 or newer.
547 - Configure boot with ``iommu=pt``.
549 - Make sure to allocate a VM on huge pages.
550 - Make sure to set CPU pinning.
552 3. Use the CPU near local NUMA node to which the PCIe adapter is connected,
553 for better performance. For VMs, verify that the right CPU
554 and NUMA node are pinned according to the above. Run:
556 .. code-block:: console
560 to identify the NUMA node to which the PCIe adapter is connected.
562 4. If more than one adapter is used, and root complex capabilities allow
563 to put both adapters on the same NUMA node without PCI bandwidth degradation,
564 it is recommended to locate both adapters on the same NUMA node.
565 This in order to forward packets from one to the other without
566 NUMA performance penalty.
568 5. Disable pause frames:
570 .. code-block:: console
572 ethtool -A <netdev> rx off tx off
574 6. Verify IO non-posted prefetch is disabled by default. This can be checked
575 via the BIOS configuration. Please contact you server provider for more
576 information about the settings.
580 On some machines, depends on the machine integrator, it is beneficial
581 to set the PCI max read request parameter to 1K. This can be
582 done in the following way:
584 To query the read request size use:
586 .. code-block:: console
588 setpci -s <NIC PCI address> 68.w
590 If the output is different than 3XXX, set it by:
592 .. code-block:: console
594 setpci -s <NIC PCI address> 68.w=3XXX
596 The XXX can be different on different systems. Make sure to configure
597 according to the setpci output.
602 Compared to librte_pmd_mlx4 that implements a single RSS configuration per
603 port, librte_pmd_mlx5 supports per-protocol RSS configuration.
605 Since ``testpmd`` defaults to IP RSS mode and there is currently no
606 command-line parameter to enable additional protocols (UDP and TCP as well
607 as IP), the following commands must be entered from its CLI to get the same
608 behavior as librte_pmd_mlx4:
610 .. code-block:: console
613 > port config all rss all
619 This section demonstrates how to launch **testpmd** with Mellanox
620 ConnectX-4/ConnectX-5 devices managed by librte_pmd_mlx5.
622 #. Load the kernel modules:
624 .. code-block:: console
626 modprobe -a ib_uverbs mlx5_core mlx5_ib
628 Alternatively if MLNX_OFED is fully installed, the following script can
631 .. code-block:: console
633 /etc/init.d/openibd restart
637 User space I/O kernel modules (uio and igb_uio) are not used and do
638 not have to be loaded.
640 #. Make sure Ethernet interfaces are in working order and linked to kernel
641 verbs. Related sysfs entries should be present:
643 .. code-block:: console
645 ls -d /sys/class/net/*/device/infiniband_verbs/uverbs* | cut -d / -f 5
649 .. code-block:: console
656 #. Optionally, retrieve their PCI bus addresses for whitelisting:
658 .. code-block:: console
661 for intf in eth2 eth3 eth4 eth5;
663 (cd "/sys/class/net/${intf}/device/" && pwd -P);
666 sed -n 's,.*/\(.*\),-w \1,p'
670 .. code-block:: console
677 #. Request huge pages:
679 .. code-block:: console
681 echo 1024 > /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages/nr_hugepages
683 #. Start testpmd with basic parameters:
685 .. code-block:: console
687 testpmd -l 8-15 -n 4 -w 05:00.0 -w 05:00.1 -w 06:00.0 -w 06:00.1 -- --rxq=2 --txq=2 -i
691 .. code-block:: console
694 EAL: PCI device 0000:05:00.0 on NUMA socket 0
695 EAL: probe driver: 15b3:1013 librte_pmd_mlx5
696 PMD: librte_pmd_mlx5: PCI information matches, using device "mlx5_0" (VF: false)
697 PMD: librte_pmd_mlx5: 1 port(s) detected
698 PMD: librte_pmd_mlx5: port 1 MAC address is e4:1d:2d:e7:0c:fe
699 EAL: PCI device 0000:05:00.1 on NUMA socket 0
700 EAL: probe driver: 15b3:1013 librte_pmd_mlx5
701 PMD: librte_pmd_mlx5: PCI information matches, using device "mlx5_1" (VF: false)
702 PMD: librte_pmd_mlx5: 1 port(s) detected
703 PMD: librte_pmd_mlx5: port 1 MAC address is e4:1d:2d:e7:0c:ff
704 EAL: PCI device 0000:06:00.0 on NUMA socket 0
705 EAL: probe driver: 15b3:1013 librte_pmd_mlx5
706 PMD: librte_pmd_mlx5: PCI information matches, using device "mlx5_2" (VF: false)
707 PMD: librte_pmd_mlx5: 1 port(s) detected
708 PMD: librte_pmd_mlx5: port 1 MAC address is e4:1d:2d:e7:0c:fa
709 EAL: PCI device 0000:06:00.1 on NUMA socket 0
710 EAL: probe driver: 15b3:1013 librte_pmd_mlx5
711 PMD: librte_pmd_mlx5: PCI information matches, using device "mlx5_3" (VF: false)
712 PMD: librte_pmd_mlx5: 1 port(s) detected
713 PMD: librte_pmd_mlx5: port 1 MAC address is e4:1d:2d:e7:0c:fb
714 Interactive-mode selected
715 Configuring Port 0 (socket 0)
716 PMD: librte_pmd_mlx5: 0x8cba80: TX queues number update: 0 -> 2
717 PMD: librte_pmd_mlx5: 0x8cba80: RX queues number update: 0 -> 2
718 Port 0: E4:1D:2D:E7:0C:FE
719 Configuring Port 1 (socket 0)
720 PMD: librte_pmd_mlx5: 0x8ccac8: TX queues number update: 0 -> 2
721 PMD: librte_pmd_mlx5: 0x8ccac8: RX queues number update: 0 -> 2
722 Port 1: E4:1D:2D:E7:0C:FF
723 Configuring Port 2 (socket 0)
724 PMD: librte_pmd_mlx5: 0x8cdb10: TX queues number update: 0 -> 2
725 PMD: librte_pmd_mlx5: 0x8cdb10: RX queues number update: 0 -> 2
726 Port 2: E4:1D:2D:E7:0C:FA
727 Configuring Port 3 (socket 0)
728 PMD: librte_pmd_mlx5: 0x8ceb58: TX queues number update: 0 -> 2
729 PMD: librte_pmd_mlx5: 0x8ceb58: RX queues number update: 0 -> 2
730 Port 3: E4:1D:2D:E7:0C:FB
731 Checking link statuses...
732 Port 0 Link Up - speed 40000 Mbps - full-duplex
733 Port 1 Link Up - speed 40000 Mbps - full-duplex
734 Port 2 Link Up - speed 10000 Mbps - full-duplex
735 Port 3 Link Up - speed 10000 Mbps - full-duplex