1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2016 Intel Corporation.
7 The i40e PMD (**librte_net_i40e**) provides poll mode driver support for
8 10/25/40 Gbps Intel® Ethernet 700 Series Network Adapters based on
9 the Intel Ethernet Controller X710/XL710/XXV710 and Intel Ethernet
10 Connection X722 (only support part of features).
16 Features of the i40e PMD are:
18 - Multiple queues for TX and RX
19 - Receiver Side Scaling (RSS)
21 - Packet type information
25 - VLAN/QinQ stripping and inserting
29 - Port hardware statistics
31 - Link state information
33 - Mirror on port, VLAN and VSI
34 - Interrupt mode for RX
35 - Scattered and gather for TX and RX
36 - Vector Poll mode driver
41 - IEEE1588/802.1AS timestamping
42 - VF Daemon (VFD) - EXPERIMENTAL
43 - Dynamic Device Personalization (DDP)
44 - Queue region configuration
45 - Virtual Function Port Representors
46 - Malicious Device Drive event catch and notify
52 - Identifying your adapter using `Intel Support
53 <http://www.intel.com/support>`_ and get the latest NVM/FW images.
55 - Follow the DPDK :ref:`Getting Started Guide for Linux <linux_gsg>` to setup the basic DPDK environment.
57 - To get better performance on Intel platforms, please follow the "How to get best performance with NICs on Intel platforms"
58 section of the :ref:`Getting Started Guide for Linux <linux_gsg>`.
60 - Upgrade the NVM/FW version following the `Intel® Ethernet NVM Update Tool Quick Usage Guide for Linux
61 <https://www-ssl.intel.com/content/www/us/en/embedded/products/networking/nvm-update-tool-quick-linux-usage-guide.html>`_ and `Intel® Ethernet NVM Update Tool: Quick Usage Guide for EFI <https://www.intel.com/content/www/us/en/embedded/products/networking/nvm-update-tool-quick-efi-usage-guide.html>`_ if needed.
63 - For information about supported media, please refer to this document: `Intel® Ethernet Controller X710/XXV710/XL710 Feature Support Matrix
64 <http://www.intel.com/content/dam/www/public/us/en/documents/release-notes/xl710-ethernet-controller-feature-matrix.pdf>`_.
68 * Some adapters based on the Intel(R) Ethernet Controller 700 Series only
69 support Intel Ethernet Optics modules. On these adapters, other modules are not
70 supported and will not function.
72 * For connections based on Intel(R) Ethernet Controller 700 Series,
73 support is dependent on your system board. Please see your vendor for details.
75 * In all cases Intel recommends using Intel Ethernet Optics; other modules
76 may function but are not validated by Intel. Contact Intel for supported media types.
81 - Follow the :doc:`guide for Windows <../windows_gsg/run_apps>`
82 to setup the basic DPDK environment.
84 - Identify the Intel® Ethernet adapter and get the latest NVM/FW version.
86 - To access any Intel® Ethernet hardware, load the NetUIO driver in place of existing built-in (inbox) driver.
88 - To load NetUIO driver, follow the steps mentioned in `dpdk-kmods repository
89 <https://git.dpdk.org/dpdk-kmods/tree/windows/netuio/README.rst>`_.
91 Recommended Matching List
92 -------------------------
94 It is highly recommended to upgrade the i40e kernel driver and firmware to
95 avoid the compatibility issues with i40e PMD. Here is the suggested matching
96 list which has been tested and verified. The detailed information can refer
97 to chapter Tested Platforms/Tested NICs in release notes.
99 For X710/XL710/XXV710,
101 +--------------+-----------------------+------------------+
102 | DPDK version | Kernel driver version | Firmware version |
103 +==============+=======================+==================+
104 | 21.02 | 2.14.13 | 8.00 |
105 +--------------+-----------------------+------------------+
106 | 20.11 | 2.14.13 | 8.00 |
107 +--------------+-----------------------+------------------+
108 | 20.08 | 2.12.6 | 7.30 |
109 +--------------+-----------------------+------------------+
110 | 20.05 | 2.11.27 | 7.30 |
111 +--------------+-----------------------+------------------+
112 | 20.02 | 2.10.19 | 7.20 |
113 +--------------+-----------------------+------------------+
114 | 19.11 | 2.9.21 | 7.00 |
115 +--------------+-----------------------+------------------+
116 | 19.08 | 2.8.43 | 7.00 |
117 +--------------+-----------------------+------------------+
118 | 19.05 | 2.7.29 | 6.80 |
119 +--------------+-----------------------+------------------+
120 | 19.02 | 2.7.26 | 6.80 |
121 +--------------+-----------------------+------------------+
122 | 18.11 | 2.4.6 | 6.01 |
123 +--------------+-----------------------+------------------+
124 | 18.08 | 2.4.6 | 6.01 |
125 +--------------+-----------------------+------------------+
126 | 18.05 | 2.4.6 | 6.01 |
127 +--------------+-----------------------+------------------+
128 | 18.02 | 2.4.3 | 6.01 |
129 +--------------+-----------------------+------------------+
130 | 17.11 | 2.1.26 | 6.01 |
131 +--------------+-----------------------+------------------+
132 | 17.08 | 2.0.19 | 6.01 |
133 +--------------+-----------------------+------------------+
134 | 17.05 | 1.5.23 | 5.05 |
135 +--------------+-----------------------+------------------+
136 | 17.02 | 1.5.23 | 5.05 |
137 +--------------+-----------------------+------------------+
138 | 16.11 | 1.5.23 | 5.05 |
139 +--------------+-----------------------+------------------+
140 | 16.07 | 1.4.25 | 5.04 |
141 +--------------+-----------------------+------------------+
142 | 16.04 | 1.4.25 | 5.02 |
143 +--------------+-----------------------+------------------+
148 +--------------+-----------------------+------------------+
149 | DPDK version | Kernel driver version | Firmware version |
150 +==============+=======================+==================+
151 | 21.02 | 2.14.13 | 5.00 |
152 +--------------+-----------------------+------------------+
153 | 20.11 | 2.13.10 | 5.00 |
154 +--------------+-----------------------+------------------+
155 | 20.08 | 2.12.6 | 4.11 |
156 +--------------+-----------------------+------------------+
157 | 20.05 | 2.11.27 | 4.11 |
158 +--------------+-----------------------+------------------+
159 | 20.02 | 2.10.19 | 4.11 |
160 +--------------+-----------------------+------------------+
161 | 19.11 | 2.9.21 | 4.10 |
162 +--------------+-----------------------+------------------+
163 | 19.08 | 2.9.21 | 4.10 |
164 +--------------+-----------------------+------------------+
165 | 19.05 | 2.7.29 | 3.33 |
166 +--------------+-----------------------+------------------+
167 | 19.02 | 2.7.26 | 3.33 |
168 +--------------+-----------------------+------------------+
169 | 18.11 | 2.4.6 | 3.33 |
170 +--------------+-----------------------+------------------+
173 Pre-Installation Configuration
174 ------------------------------
179 The following options can be modified in the ``config/rte_config.h`` file.
181 - ``RTE_LIBRTE_I40E_QUEUE_NUM_PER_PF`` (default ``64``)
183 Number of queues reserved for PF.
185 - ``RTE_LIBRTE_I40E_QUEUE_NUM_PER_VM`` (default ``4``)
187 Number of queues reserved for each VMDQ Pool.
189 Runtime Config Options
190 ~~~~~~~~~~~~~~~~~~~~~~
192 - ``Reserved number of Queues per VF`` (default ``4``)
194 The number of reserved queue per VF is determined by its host PF. If the
195 PCI address of an i40e PF is aaaa:bb.cc, the number of reserved queues per
196 VF can be configured with EAL parameter like -a aaaa:bb.cc,queue-num-per-vf=n.
197 The value n can be 1, 2, 4, 8 or 16. If no such parameter is configured, the
198 number of reserved queues per VF is 4 by default. If VF request more than
199 reserved queues per VF, PF will able to allocate max to 16 queues after a VF
203 - ``Support multiple driver`` (default ``disable``)
205 There was a multiple driver support issue during use of 700 series Ethernet
206 Adapter with both Linux kernel and DPDK PMD. To fix this issue, ``devargs``
207 parameter ``support-multi-driver`` is introduced, for example::
209 -a 84:00.0,support-multi-driver=1
211 With the above configuration, DPDK PMD will not change global registers, and
212 will switch PF interrupt from IntN to Int0 to avoid interrupt conflict between
213 DPDK and Linux Kernel.
215 - ``Support VF Port Representor`` (default ``not enabled``)
217 The i40e PF PMD supports the creation of VF port representors for the control
218 and monitoring of i40e virtual function devices. Each port representor
219 corresponds to a single virtual function of that device. Using the ``devargs``
220 option ``representor`` the user can specify which virtual functions to create
221 port representors for on initialization of the PF PMD by passing the VF IDs of
222 the VFs which are required.::
224 -a DBDF,representor=[0,1,4]
226 Currently hot-plugging of representor ports is not supported so all required
227 representors must be specified on the creation of the PF.
229 - ``Enable validation for VF message`` (default ``not enabled``)
231 The PF counts messages from each VF. If in any period of seconds the message
232 statistic from a VF exceeds maximal limitation, the PF will ignore any new message
233 from that VF for some seconds.
234 Format -- "maximal-message@period-seconds:ignore-seconds"
237 -a 84:00.0,vf_msg_cfg=80@120:180
239 Vector RX Pre-conditions
240 ~~~~~~~~~~~~~~~~~~~~~~~~
241 For Vector RX it is assumed that the number of descriptor rings will be a power
242 of 2. With this pre-condition, the ring pointer can easily scroll back to the
243 head after hitting the tail without a conditional check. In addition Vector RX
244 can use this assumption to do a bit mask using ``ring_size - 1``.
246 Driver compilation and testing
247 ------------------------------
249 Refer to the document :ref:`compiling and testing a PMD for a NIC <pmd_build_and_test>`
253 SR-IOV: Prerequisites and sample Application Notes
254 --------------------------------------------------
256 #. Load the kernel module:
258 .. code-block:: console
262 Check the output in dmesg:
264 .. code-block:: console
266 i40e 0000:83:00.1 ens802f0: renamed from eth0
268 #. Bring up the PF ports:
270 .. code-block:: console
274 #. Create VF device(s):
276 Echo the number of VFs to be created into the ``sriov_numvfs`` sysfs entry
281 .. code-block:: console
283 echo 2 > /sys/devices/pci0000:00/0000:00:03.0/0000:81:00.0/sriov_numvfs
286 #. Assign VF MAC address:
288 Assign MAC address to the VF using iproute2 utility. The syntax is:
290 .. code-block:: console
292 ip link set <PF netdev id> vf <VF id> mac <macaddr>
296 .. code-block:: console
298 ip link set ens802f0 vf 0 mac a0:b0:c0:d0:e0:f0
300 #. Assign VF to VM, and bring up the VM.
301 Please see the documentation for the *I40E/IXGBE/IGB Virtual Function Driver*.
305 Follow instructions available in the document
306 :ref:`compiling and testing a PMD for a NIC <pmd_build_and_test>`
311 .. code-block:: console
314 EAL: PCI device 0000:83:00.0 on NUMA socket 1
315 EAL: probe driver: 8086:1572 rte_i40e_pmd
316 EAL: PCI memory mapped at 0x7f7f80000000
317 EAL: PCI memory mapped at 0x7f7f80800000
318 PMD: eth_i40e_dev_init(): FW 5.0 API 1.5 NVM 05.00.02 eetrack 8000208a
319 Interactive-mode selected
320 Configuring Port 0 (socket 0)
323 PMD: i40e_dev_rx_queue_setup(): Rx Burst Bulk Alloc Preconditions are
324 satisfied.Rx Burst Bulk Alloc function will be used on port=0, queue=0.
327 Port 0: 68:05:CA:26:85:84
328 Checking link statuses...
329 Port 0 Link Up - speed 10000 Mbps - full-duplex
335 Sample Application Notes
336 ------------------------
341 Vlan filter only works when Promiscuous mode is off.
343 To start ``testpmd``, and add vlan 10 to port 0:
345 .. code-block:: console
347 ./<build_dir>/app/dpdk-testpmd -l 0-15 -n 4 -- -i --forward-mode=mac
350 testpmd> set promisc 0 off
351 testpmd> rx_vlan add 10 0
357 The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
358 The Flow Director filters can match the different fields for different type of packet: flow type, specific input set per flow type and the flexible payload.
360 The default input set of each flow type is::
362 ipv4-other : src_ip_address, dst_ip_address
363 ipv4-frag : src_ip_address, dst_ip_address
364 ipv4-tcp : src_ip_address, dst_ip_address, src_port, dst_port
365 ipv4-udp : src_ip_address, dst_ip_address, src_port, dst_port
366 ipv4-sctp : src_ip_address, dst_ip_address, src_port, dst_port,
368 ipv6-other : src_ip_address, dst_ip_address
369 ipv6-frag : src_ip_address, dst_ip_address
370 ipv6-tcp : src_ip_address, dst_ip_address, src_port, dst_port
371 ipv6-udp : src_ip_address, dst_ip_address, src_port, dst_port
372 ipv6-sctp : src_ip_address, dst_ip_address, src_port, dst_port,
374 l2_payload : ether_type
376 The flex payload is selected from offset 0 to 15 of packet's payload by default, while it is masked out from matching.
378 Start ``testpmd`` with ``--disable-rss`` and ``--pkt-filter-mode=perfect``:
380 .. code-block:: console
382 ./<build_dir>/app/dpdk-testpmd -l 0-15 -n 4 -- -i --disable-rss \
383 --pkt-filter-mode=perfect --rxq=8 --txq=8 --nb-cores=8 \
386 Add a rule to direct ``ipv4-udp`` packet whose ``dst_ip=2.2.2.5, src_ip=2.2.2.3, src_port=32, dst_port=32`` to queue 1:
388 .. code-block:: console
390 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.3 \
391 dst is 2.2.2.5 / udp src is 32 dst is 32 / end \
392 actions mark id 1 / queue index 1 / end
394 Check the flow director status:
396 .. code-block:: console
398 testpmd> show port fdir 0
400 ######################## FDIR infos for port 0 ####################
402 SUPPORTED FLOW TYPE: ipv4-frag ipv4-tcp ipv4-udp ipv4-sctp ipv4-other
403 ipv6-frag ipv6-tcp ipv6-udp ipv6-sctp ipv6-other
406 max_len: 16 payload_limit: 480
407 payload_unit: 2 payload_seg: 3
408 bitmask_unit: 2 bitmask_num: 2
411 src_ipv4: 0x00000000,
412 dst_ipv4: 0x00000000,
415 src_ipv6: 0x00000000,0x00000000,0x00000000,0x00000000,
416 dst_ipv6: 0x00000000,0x00000000,0x00000000,0x00000000
417 FLEX PAYLOAD SRC OFFSET:
418 L2_PAYLOAD: 0 1 2 3 4 5 6 ...
419 L3_PAYLOAD: 0 1 2 3 4 5 6 ...
420 L4_PAYLOAD: 0 1 2 3 4 5 6 ...
422 ipv4-udp: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
423 ipv4-tcp: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
424 ipv4-sctp: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
425 ipv4-other: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
426 ipv4-frag: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
427 ipv6-udp: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
428 ipv6-tcp: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
429 ipv6-sctp: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
430 ipv6-other: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
431 ipv6-frag: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
432 l2_payload: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
433 guarant_count: 1 best_count: 0
434 guarant_space: 512 best_space: 7168
444 The Intel® Ethernet 700 Series support a feature called
447 A Virtual Ethernet Bridge (VEB) is an IEEE Edge Virtual Bridging (EVB) term
448 for functionality that allows local switching between virtual endpoints within
449 a physical endpoint and also with an external bridge/network.
451 A "Floating" VEB doesn't have an uplink connection to the outside world so all
452 switching is done internally and remains within the host. As such, this
453 feature provides security benefits.
455 In addition, a Floating VEB overcomes a limitation of normal VEBs where they
456 cannot forward packets when the physical link is down. Floating VEBs don't need
457 to connect to the NIC port so they can still forward traffic from VF to VF
458 even when the physical link is down.
460 Therefore, with this feature enabled VFs can be limited to communicating with
461 each other but not an outside network, and they can do so even when there is
462 no physical uplink on the associated NIC port.
464 To enable this feature, the user should pass a ``devargs`` parameter to the
467 -a 84:00.0,enable_floating_veb=1
469 In this configuration the PMD will use the floating VEB feature for all the
470 VFs created by this PF device.
472 Alternatively, the user can specify which VFs need to connect to this floating
473 VEB using the ``floating_veb_list`` argument::
475 -a 84:00.0,enable_floating_veb=1,floating_veb_list=1;3-4
477 In this example ``VF1``, ``VF3`` and ``VF4`` connect to the floating VEB,
478 while other VFs connect to the normal VEB.
480 The current implementation only supports one floating VEB and one regular
481 VEB. VFs can connect to a floating VEB or a regular VEB according to the
482 configuration passed on the EAL command line.
484 The floating VEB functionality requires a NIC firmware version of 5.0
487 Dynamic Device Personalization (DDP)
488 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
490 The Intel® Ethernet 700 Series except for the Intel Ethernet Connection
491 X722 support a feature called "Dynamic Device Personalization (DDP)",
492 which is used to configure hardware by downloading a profile to support
493 protocols/filters which are not supported by default. The DDP
494 functionality requires a NIC firmware version of 6.0 or greater.
496 Current implementation supports GTP-C/GTP-U/PPPoE/PPPoL2TP/ESP,
497 steering can be used with rte_flow API.
499 GTPv1 package is released, and it can be downloaded from
500 https://downloadcenter.intel.com/download/27587.
502 PPPoE package is released, and it can be downloaded from
503 https://downloadcenter.intel.com/download/28040.
505 ESP-AH package is released, and it can be downloaded from
506 https://downloadcenter.intel.com/download/29446.
508 Load a profile which supports GTP and store backup profile:
510 .. code-block:: console
512 testpmd> ddp add 0 ./gtp.pkgo,./backup.pkgo
514 Delete a GTP profile and restore backup profile:
516 .. code-block:: console
518 testpmd> ddp del 0 ./backup.pkgo
520 Get loaded DDP package info list:
522 .. code-block:: console
524 testpmd> ddp get list 0
526 Display information about a GTP profile:
528 .. code-block:: console
530 testpmd> ddp get info ./gtp.pkgo
532 Input set configuration
533 ~~~~~~~~~~~~~~~~~~~~~~~
534 Input set for any PCTYPE can be configured with user defined configuration,
535 For example, to use only 48bit prefix for IPv6 src address for IPv6 TCP RSS:
537 .. code-block:: console
539 testpmd> port config 0 pctype 43 hash_inset clear all
540 testpmd> port config 0 pctype 43 hash_inset set field 13
541 testpmd> port config 0 pctype 43 hash_inset set field 14
542 testpmd> port config 0 pctype 43 hash_inset set field 15
544 Queue region configuration
545 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
546 The Intel® Ethernet 700 Series supports a feature of queue regions
547 configuration for RSS in the PF, so that different traffic classes or
548 different packet classification types can be separated to different
549 queues in different queue regions. There is an API for configuration
550 of queue regions in RSS with a command line. It can parse the parameters
551 of the region index, queue number, queue start index, user priority, traffic
552 classes and so on. Depending on commands from the command line, it will call
553 i40e private APIs and start the process of setting or flushing the queue
554 region configuration. As this feature is specific for i40e only private
555 APIs are used. These new ``test_pmd`` commands are as shown below. For
556 details please refer to :doc:`../testpmd_app_ug/index`.
558 .. code-block:: console
560 testpmd> set port (port_id) queue-region region_id (value) \
561 queue_start_index (value) queue_num (value)
562 testpmd> set port (port_id) queue-region region_id (value) flowtype (value)
563 testpmd> set port (port_id) queue-region UP (value) region_id (value)
564 testpmd> set port (port_id) queue-region flush (on|off)
565 testpmd> show port (port_id) queue-region
572 RSS Flow supports to set hash input set, hash function, enable hash
573 and configure queues.
575 Configure queues as queue 0, 1, 2, 3.
577 .. code-block:: console
579 testpmd> flow create 0 ingress pattern end actions rss types end \
580 queues 0 1 2 3 end / end
582 Enable hash and set input set for ipv4-tcp.
584 .. code-block:: console
586 testpmd> flow create 0 ingress pattern eth / ipv4 / tcp / end \
587 actions rss types ipv4-tcp l3-src-only end queues end / end
589 Set symmetric hash enable for flow type ipv4-tcp.
591 .. code-block:: console
593 testpmd> flow create 0 ingress pattern eth / ipv4 / tcp / end \
594 actions rss types ipv4-tcp end queues end func symmetric_toeplitz / end
596 Set hash function as simple xor.
598 .. code-block:: console
600 testpmd> flow create 0 ingress pattern end actions rss types end \
601 queues end func simple_xor / end
603 Limitations or Known issues
604 ---------------------------
606 MPLS packet classification
607 ~~~~~~~~~~~~~~~~~~~~~~~~~~
609 For firmware versions prior to 5.0, MPLS packets are not recognized by the NIC.
610 The L2 Payload flow type in flow director can be used to classify MPLS packet
611 by using a command in testpmd like:
613 testpmd> flow_director_filter 0 mode IP add flow l2_payload ether \
614 0x8847 flexbytes () fwd pf queue <N> fd_id <M>
616 With the NIC firmware version 5.0 or greater, some limited MPLS support
617 is added: Native MPLS (MPLS in Ethernet) skip is implemented, while no
618 new packet type, no classification or offload are possible. With this change,
619 L2 Payload flow type in flow director cannot be used to classify MPLS packet
620 as with previous firmware versions. Meanwhile, the Ethertype filter can be
621 used to classify MPLS packet by using a command in testpmd like:
623 testpmd> flow create 0 ingress pattern eth type is 0x8847 / end \
624 actions queue index <M> / end
626 16 Byte RX Descriptor setting on DPDK VF
627 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
629 Currently the VF's RX descriptor mode is decided by PF. There's no PF-VF
630 interface for VF to request the RX descriptor mode, also no interface to notify
631 VF its own RX descriptor mode.
632 For all available versions of the i40e driver, these drivers don't support 16
633 byte RX descriptor. If the Linux i40e kernel driver is used as host driver,
634 while DPDK i40e PMD is used as the VF driver, DPDK cannot choose 16 byte receive
635 descriptor. The reason is that the RX descriptor is already set to 32 byte by
636 the i40e kernel driver.
637 In the future, if the Linux i40e driver supports 16 byte RX descriptor, user
638 should make sure the DPDK VF uses the same RX descriptor mode, 16 byte or 32
639 byte, as the PF driver.
641 The same rule for DPDK PF + DPDK VF. The PF and VF should use the same RX
642 descriptor mode. Or the VF RX will not work.
644 Receive packets with Ethertype 0x88A8
645 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
647 Due to the FW limitation, PF can receive packets with Ethertype 0x88A8
648 only when floating VEB is disabled.
650 Incorrect Rx statistics when packet is oversize
651 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
653 When a packet is over maximum frame size, the packet is dropped.
654 However, the Rx statistics, when calling `rte_eth_stats_get` incorrectly
655 shows it as received.
657 RX/TX statistics may be incorrect when register overflowed
658 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
660 The rx_bytes/tx_bytes statistics register is 48 bit length.
661 Although this limitation is enlarged to 64 bit length on the software side,
662 but there is no way to detect if the overflow occurred more than once.
663 So rx_bytes/tx_bytes statistics data is correct when statistics are
664 updated at least once between two overflows.
666 VF & TC max bandwidth setting
667 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
669 The per VF max bandwidth and per TC max bandwidth cannot be enabled in parallel.
670 The behavior is different when handling per VF and per TC max bandwidth setting.
671 When enabling per VF max bandwidth, SW will check if per TC max bandwidth is
672 enabled. If so, return failure.
673 When enabling per TC max bandwidth, SW will check if per VF max bandwidth
674 is enabled. If so, disable per VF max bandwidth and continue with per TC max
677 TC TX scheduling mode setting
678 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
680 There are 2 TX scheduling modes for TCs, round robin and strict priority mode.
681 If a TC is set to strict priority mode, it can consume unlimited bandwidth.
682 It means if APP has set the max bandwidth for that TC, it comes to no
684 It's suggested to set the strict priority mode for a TC that is latency
685 sensitive but no consuming much bandwidth.
687 VF performance is impacted by PCI extended tag setting
688 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
690 To reach maximum NIC performance in the VF the PCI extended tag must be
691 enabled. The DPDK i40e PF driver will set this feature during initialization,
692 but the kernel PF driver does not. So when running traffic on a VF which is
693 managed by the kernel PF driver, a significant NIC performance downgrade has
694 been observed (for 64 byte packets, there is about 25% line-rate downgrade for
695 a 25GbE device and about 35% for a 40GbE device).
697 For kernel version >= 4.11, the kernel's PCI driver will enable the extended
698 tag if it detects that the device supports it. So by default, this is not an
699 issue. For kernels <= 4.11 or when the PCI extended tag is disabled it can be
700 enabled using the steps below.
702 #. Get the current value of the PCI configure register::
704 setpci -s <XX:XX.X> a8.w
708 value = value | 0x100
710 #. Set the PCI configure register with new value::
712 setpci -s <XX:XX.X> a8.w=<value>
717 The VF vlan strip function is only supported in the i40e kernel driver >= 2.1.26.
722 DCB works only when RSS is enabled.
724 Global configuration warning
725 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
727 I40E PMD will set some global registers to enable some function or set some
728 configure. Then when using different ports of the same NIC with Linux kernel
729 and DPDK, the port with Linux kernel will be impacted by the port with DPDK.
730 For example, register I40E_GL_SWT_L2TAGCTRL is used to control L2 tag, i40e
731 PMD uses I40E_GL_SWT_L2TAGCTRL to set vlan TPID. If setting TPID in port A
732 with DPDK, then the configuration will also impact port B in the NIC with
733 kernel driver, which don't want to use the TPID.
734 So PMD reports warning to clarify what is changed by writing global register.
739 When programming cloud filters for IPv4/6_UDP/TCP/SCTP with SRC port only or DST port only,
740 it will make any cloud filter using inner_vlan or tunnel key invalid. Default configuration will be
741 recovered only by NIC core reset.
743 Mirror rule limitation for X722
744 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
746 Due to firmware restriction of X722, the same VSI cannot have more than one mirror rule.
748 High Performance of Small Packets on 40GbE NIC
749 ----------------------------------------------
751 As there might be firmware fixes for performance enhancement in latest version
752 of firmware image, the firmware update might be needed for getting high performance.
753 Check the Intel support website for the latest firmware updates.
754 Users should consult the release notes specific to a DPDK release to identify
755 the validated firmware version for a NIC using the i40e driver.
757 Use 16 Bytes RX Descriptor Size
758 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
760 As i40e PMD supports both 16 and 32 bytes RX descriptor sizes, and 16 bytes size can provide helps to high performance of small packets.
761 In ``config/rte_config.h`` set the following to use 16 bytes size RX descriptors::
763 #define RTE_LIBRTE_I40E_16BYTE_RX_DESC 1
765 Input set requirement of each pctype for FDIR
766 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
768 Each PCTYPE can only have one specific FDIR input set at one time.
769 For example, if creating 2 rte_flow rules with different input set for one PCTYPE,
770 it will fail and return the info "Conflict with the first rule's input set",
771 which means the current rule's input set conflicts with the first rule's.
772 Remove the first rule if want to change the input set of the PCTYPE.
774 Example of getting best performance with l3fwd example
775 ------------------------------------------------------
777 The following is an example of running the DPDK ``l3fwd`` sample application to get high performance with a
778 server with Intel Xeon processors and Intel Ethernet CNA XL710.
780 The example scenario is to get best performance with two Intel Ethernet CNA XL710 40GbE ports.
781 See :numref:`figure_intel_perf_test_setup` for the performance test setup.
783 .. _figure_intel_perf_test_setup:
785 .. figure:: img/intel_perf_test_setup.*
787 Performance Test Setup
790 1. Add two Intel Ethernet CNA XL710 to the platform, and use one port per card to get best performance.
791 The reason for using two NICs is to overcome a PCIe v3.0 limitation since it cannot provide 80GbE bandwidth
792 for two 40GbE ports, but two different PCIe v3.0 x8 slot can.
793 Refer to the sample NICs output above, then we can select ``82:00.0`` and ``85:00.0`` as test ports::
795 82:00.0 Ethernet [0200]: Intel XL710 for 40GbE QSFP+ [8086:1583]
796 85:00.0 Ethernet [0200]: Intel XL710 for 40GbE QSFP+ [8086:1583]
798 2. Connect the ports to the traffic generator. For high speed testing, it's best to use a hardware traffic generator.
800 3. Check the PCI devices numa node (socket id) and get the cores number on the exact socket id.
801 In this case, ``82:00.0`` and ``85:00.0`` are both in socket 1, and the cores on socket 1 in the referenced platform
803 Note: Don't use 2 logical cores on the same core (e.g core18 has 2 logical cores, core18 and core54), instead, use 2 logical
804 cores from different cores (e.g core18 and core19).
806 4. Bind these two ports to igb_uio.
808 5. As to Intel Ethernet CNA XL710 40GbE port, we need at least two queue pairs to achieve best performance, then two queues per port
809 will be required, and each queue pair will need a dedicated CPU core for receiving/transmitting packets.
811 6. The DPDK sample application ``l3fwd`` will be used for performance testing, with using two ports for bi-directional forwarding.
812 Compile the ``l3fwd sample`` with the default lpm mode.
814 7. The command line of running l3fwd would be something like the following::
816 ./dpdk-l3fwd -l 18-21 -n 4 -a 82:00.0 -a 85:00.0 \
817 -- -p 0x3 --config '(0,0,18),(0,1,19),(1,0,20),(1,1,21)'
819 This means that the application uses core 18 for port 0, queue pair 0 forwarding, core 19 for port 0, queue pair 1 forwarding,
820 core 20 for port 1, queue pair 0 forwarding, and core 21 for port 1, queue pair 1 forwarding.
822 8. Configure the traffic at a traffic generator.
824 * Start creating a stream on packet generator.
826 * Set the Ethernet II type to 0x0800.
828 Tx bytes affected by the link status change
829 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
831 For firmware versions prior to 6.01 for X710 series and 3.33 for X722 series, the tx_bytes statistics data is affected by
832 the link down event. Each time the link status changes to down, the tx_bytes decreases 110 bytes.