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.
78 Recommended Matching List
79 -------------------------
81 It is highly recommended to upgrade the i40e kernel driver and firmware to
82 avoid the compatibility issues with i40e PMD. Here is the suggested matching
83 list which has been tested and verified. The detailed information can refer
84 to chapter Tested Platforms/Tested NICs in release notes.
86 For X710/XL710/XXV710,
88 +--------------+-----------------------+------------------+
89 | DPDK version | Kernel driver version | Firmware version |
90 +==============+=======================+==================+
91 | 20.11 | 2.13.10 | 8.00 |
92 +--------------+-----------------------+------------------+
93 | 20.08 | 2.12.6 | 7.30 |
94 +--------------+-----------------------+------------------+
95 | 20.05 | 2.11.27 | 7.30 |
96 +--------------+-----------------------+------------------+
97 | 20.02 | 2.10.19 | 7.20 |
98 +--------------+-----------------------+------------------+
99 | 19.11 | 2.9.21 | 7.00 |
100 +--------------+-----------------------+------------------+
101 | 19.08 | 2.8.43 | 7.00 |
102 +--------------+-----------------------+------------------+
103 | 19.05 | 2.7.29 | 6.80 |
104 +--------------+-----------------------+------------------+
105 | 19.02 | 2.7.26 | 6.80 |
106 +--------------+-----------------------+------------------+
107 | 18.11 | 2.4.6 | 6.01 |
108 +--------------+-----------------------+------------------+
109 | 18.08 | 2.4.6 | 6.01 |
110 +--------------+-----------------------+------------------+
111 | 18.05 | 2.4.6 | 6.01 |
112 +--------------+-----------------------+------------------+
113 | 18.02 | 2.4.3 | 6.01 |
114 +--------------+-----------------------+------------------+
115 | 17.11 | 2.1.26 | 6.01 |
116 +--------------+-----------------------+------------------+
117 | 17.08 | 2.0.19 | 6.01 |
118 +--------------+-----------------------+------------------+
119 | 17.05 | 1.5.23 | 5.05 |
120 +--------------+-----------------------+------------------+
121 | 17.02 | 1.5.23 | 5.05 |
122 +--------------+-----------------------+------------------+
123 | 16.11 | 1.5.23 | 5.05 |
124 +--------------+-----------------------+------------------+
125 | 16.07 | 1.4.25 | 5.04 |
126 +--------------+-----------------------+------------------+
127 | 16.04 | 1.4.25 | 5.02 |
128 +--------------+-----------------------+------------------+
133 +--------------+-----------------------+------------------+
134 | DPDK version | Kernel driver version | Firmware version |
135 +==============+=======================+==================+
136 | 20.11 | 2.13.10 | 5.00 |
137 +--------------+-----------------------+------------------+
138 | 20.08 | 2.12.6 | 4.11 |
139 +--------------+-----------------------+------------------+
140 | 20.05 | 2.11.27 | 4.11 |
141 +--------------+-----------------------+------------------+
142 | 20.02 | 2.10.19 | 4.11 |
143 +--------------+-----------------------+------------------+
144 | 19.11 | 2.9.21 | 4.10 |
145 +--------------+-----------------------+------------------+
146 | 19.08 | 2.9.21 | 4.10 |
147 +--------------+-----------------------+------------------+
148 | 19.05 | 2.7.29 | 3.33 |
149 +--------------+-----------------------+------------------+
150 | 19.02 | 2.7.26 | 3.33 |
151 +--------------+-----------------------+------------------+
152 | 18.11 | 2.4.6 | 3.33 |
153 +--------------+-----------------------+------------------+
156 Pre-Installation Configuration
157 ------------------------------
162 The following options can be modified in the ``config/rte_config.h`` file.
164 - ``RTE_LIBRTE_I40E_QUEUE_NUM_PER_PF`` (default ``64``)
166 Number of queues reserved for PF.
168 - ``RTE_LIBRTE_I40E_QUEUE_NUM_PER_VM`` (default ``4``)
170 Number of queues reserved for each VMDQ Pool.
172 Runtime Config Options
173 ~~~~~~~~~~~~~~~~~~~~~~
175 - ``Reserved number of Queues per VF`` (default ``4``)
177 The number of reserved queue per VF is determined by its host PF. If the
178 PCI address of an i40e PF is aaaa:bb.cc, the number of reserved queues per
179 VF can be configured with EAL parameter like -a aaaa:bb.cc,queue-num-per-vf=n.
180 The value n can be 1, 2, 4, 8 or 16. If no such parameter is configured, the
181 number of reserved queues per VF is 4 by default. If VF request more than
182 reserved queues per VF, PF will able to allocate max to 16 queues after a VF
186 - ``Support multiple driver`` (default ``disable``)
188 There was a multiple driver support issue during use of 700 series Ethernet
189 Adapter with both Linux kernel and DPDK PMD. To fix this issue, ``devargs``
190 parameter ``support-multi-driver`` is introduced, for example::
192 -a 84:00.0,support-multi-driver=1
194 With the above configuration, DPDK PMD will not change global registers, and
195 will switch PF interrupt from IntN to Int0 to avoid interrupt conflict between
196 DPDK and Linux Kernel.
198 - ``Support VF Port Representor`` (default ``not enabled``)
200 The i40e PF PMD supports the creation of VF port representors for the control
201 and monitoring of i40e virtual function devices. Each port representor
202 corresponds to a single virtual function of that device. Using the ``devargs``
203 option ``representor`` the user can specify which virtual functions to create
204 port representors for on initialization of the PF PMD by passing the VF IDs of
205 the VFs which are required.::
207 -a DBDF,representor=[0,1,4]
209 Currently hot-plugging of representor ports is not supported so all required
210 representors must be specified on the creation of the PF.
212 - ``Use latest supported vector`` (default ``disable``)
214 Latest supported vector path may not always get the best perf so vector path was
215 recommended to use only on later platform. But users may want the latest vector path
216 since it can get better perf in some real work loading cases. So ``devargs`` param
217 ``use-latest-supported-vec`` is introduced, for example::
219 -a 84:00.0,use-latest-supported-vec=1
221 - ``Enable validation for VF message`` (default ``not enabled``)
223 The PF counts messages from each VF. If in any period of seconds the message
224 statistic from a VF exceeds maximal limitation, the PF will ignore any new message
225 from that VF for some seconds.
226 Format -- "maximal-message@period-seconds:ignore-seconds"
229 -a 84:00.0,vf_msg_cfg=80@120:180
231 Vector RX Pre-conditions
232 ~~~~~~~~~~~~~~~~~~~~~~~~
233 For Vector RX it is assumed that the number of descriptor rings will be a power
234 of 2. With this pre-condition, the ring pointer can easily scroll back to the
235 head after hitting the tail without a conditional check. In addition Vector RX
236 can use this assumption to do a bit mask using ``ring_size - 1``.
238 Driver compilation and testing
239 ------------------------------
241 Refer to the document :ref:`compiling and testing a PMD for a NIC <pmd_build_and_test>`
245 SR-IOV: Prerequisites and sample Application Notes
246 --------------------------------------------------
248 #. Load the kernel module:
250 .. code-block:: console
254 Check the output in dmesg:
256 .. code-block:: console
258 i40e 0000:83:00.1 ens802f0: renamed from eth0
260 #. Bring up the PF ports:
262 .. code-block:: console
266 #. Create VF device(s):
268 Echo the number of VFs to be created into the ``sriov_numvfs`` sysfs entry
273 .. code-block:: console
275 echo 2 > /sys/devices/pci0000:00/0000:00:03.0/0000:81:00.0/sriov_numvfs
278 #. Assign VF MAC address:
280 Assign MAC address to the VF using iproute2 utility. The syntax is:
282 .. code-block:: console
284 ip link set <PF netdev id> vf <VF id> mac <macaddr>
288 .. code-block:: console
290 ip link set ens802f0 vf 0 mac a0:b0:c0:d0:e0:f0
292 #. Assign VF to VM, and bring up the VM.
293 Please see the documentation for the *I40E/IXGBE/IGB Virtual Function Driver*.
297 Follow instructions available in the document
298 :ref:`compiling and testing a PMD for a NIC <pmd_build_and_test>`
303 .. code-block:: console
306 EAL: PCI device 0000:83:00.0 on NUMA socket 1
307 EAL: probe driver: 8086:1572 rte_i40e_pmd
308 EAL: PCI memory mapped at 0x7f7f80000000
309 EAL: PCI memory mapped at 0x7f7f80800000
310 PMD: eth_i40e_dev_init(): FW 5.0 API 1.5 NVM 05.00.02 eetrack 8000208a
311 Interactive-mode selected
312 Configuring Port 0 (socket 0)
315 PMD: i40e_dev_rx_queue_setup(): Rx Burst Bulk Alloc Preconditions are
316 satisfied.Rx Burst Bulk Alloc function will be used on port=0, queue=0.
319 Port 0: 68:05:CA:26:85:84
320 Checking link statuses...
321 Port 0 Link Up - speed 10000 Mbps - full-duplex
327 Sample Application Notes
328 ------------------------
333 Vlan filter only works when Promiscuous mode is off.
335 To start ``testpmd``, and add vlan 10 to port 0:
337 .. code-block:: console
339 ./<build_dir>/app/dpdk-testpmd -l 0-15 -n 4 -- -i --forward-mode=mac
342 testpmd> set promisc 0 off
343 testpmd> rx_vlan add 10 0
349 The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
350 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.
352 The default input set of each flow type is::
354 ipv4-other : src_ip_address, dst_ip_address
355 ipv4-frag : src_ip_address, dst_ip_address
356 ipv4-tcp : src_ip_address, dst_ip_address, src_port, dst_port
357 ipv4-udp : src_ip_address, dst_ip_address, src_port, dst_port
358 ipv4-sctp : src_ip_address, dst_ip_address, src_port, dst_port,
360 ipv6-other : src_ip_address, dst_ip_address
361 ipv6-frag : src_ip_address, dst_ip_address
362 ipv6-tcp : src_ip_address, dst_ip_address, src_port, dst_port
363 ipv6-udp : src_ip_address, dst_ip_address, src_port, dst_port
364 ipv6-sctp : src_ip_address, dst_ip_address, src_port, dst_port,
366 l2_payload : ether_type
368 The flex payload is selected from offset 0 to 15 of packet's payload by default, while it is masked out from matching.
370 Start ``testpmd`` with ``--disable-rss`` and ``--pkt-filter-mode=perfect``:
372 .. code-block:: console
374 ./<build_dir>/app/dpdk-testpmd -l 0-15 -n 4 -- -i --disable-rss \
375 --pkt-filter-mode=perfect --rxq=8 --txq=8 --nb-cores=8 \
378 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:
380 .. code-block:: console
382 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.3 \
383 dst is 2.2.2.5 / udp src is 32 dst is 32 / end \
384 actions mark id 1 / queue index 1 / end
386 Check the flow director status:
388 .. code-block:: console
390 testpmd> show port fdir 0
392 ######################## FDIR infos for port 0 ####################
394 SUPPORTED FLOW TYPE: ipv4-frag ipv4-tcp ipv4-udp ipv4-sctp ipv4-other
395 ipv6-frag ipv6-tcp ipv6-udp ipv6-sctp ipv6-other
398 max_len: 16 payload_limit: 480
399 payload_unit: 2 payload_seg: 3
400 bitmask_unit: 2 bitmask_num: 2
403 src_ipv4: 0x00000000,
404 dst_ipv4: 0x00000000,
407 src_ipv6: 0x00000000,0x00000000,0x00000000,0x00000000,
408 dst_ipv6: 0x00000000,0x00000000,0x00000000,0x00000000
409 FLEX PAYLOAD SRC OFFSET:
410 L2_PAYLOAD: 0 1 2 3 4 5 6 ...
411 L3_PAYLOAD: 0 1 2 3 4 5 6 ...
412 L4_PAYLOAD: 0 1 2 3 4 5 6 ...
414 ipv4-udp: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
415 ipv4-tcp: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
416 ipv4-sctp: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
417 ipv4-other: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
418 ipv4-frag: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
419 ipv6-udp: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
420 ipv6-tcp: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
421 ipv6-sctp: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
422 ipv6-other: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
423 ipv6-frag: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
424 l2_payload: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
425 guarant_count: 1 best_count: 0
426 guarant_space: 512 best_space: 7168
436 The Intel® Ethernet 700 Series support a feature called
439 A Virtual Ethernet Bridge (VEB) is an IEEE Edge Virtual Bridging (EVB) term
440 for functionality that allows local switching between virtual endpoints within
441 a physical endpoint and also with an external bridge/network.
443 A "Floating" VEB doesn't have an uplink connection to the outside world so all
444 switching is done internally and remains within the host. As such, this
445 feature provides security benefits.
447 In addition, a Floating VEB overcomes a limitation of normal VEBs where they
448 cannot forward packets when the physical link is down. Floating VEBs don't need
449 to connect to the NIC port so they can still forward traffic from VF to VF
450 even when the physical link is down.
452 Therefore, with this feature enabled VFs can be limited to communicating with
453 each other but not an outside network, and they can do so even when there is
454 no physical uplink on the associated NIC port.
456 To enable this feature, the user should pass a ``devargs`` parameter to the
459 -a 84:00.0,enable_floating_veb=1
461 In this configuration the PMD will use the floating VEB feature for all the
462 VFs created by this PF device.
464 Alternatively, the user can specify which VFs need to connect to this floating
465 VEB using the ``floating_veb_list`` argument::
467 -a 84:00.0,enable_floating_veb=1,floating_veb_list=1;3-4
469 In this example ``VF1``, ``VF3`` and ``VF4`` connect to the floating VEB,
470 while other VFs connect to the normal VEB.
472 The current implementation only supports one floating VEB and one regular
473 VEB. VFs can connect to a floating VEB or a regular VEB according to the
474 configuration passed on the EAL command line.
476 The floating VEB functionality requires a NIC firmware version of 5.0
479 Dynamic Device Personalization (DDP)
480 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
482 The Intel® Ethernet 700 Series except for the Intel Ethernet Connection
483 X722 support a feature called "Dynamic Device Personalization (DDP)",
484 which is used to configure hardware by downloading a profile to support
485 protocols/filters which are not supported by default. The DDP
486 functionality requires a NIC firmware version of 6.0 or greater.
488 Current implementation supports GTP-C/GTP-U/PPPoE/PPPoL2TP/ESP,
489 steering can be used with rte_flow API.
491 GTPv1 package is released, and it can be downloaded from
492 https://downloadcenter.intel.com/download/27587.
494 PPPoE package is released, and it can be downloaded from
495 https://downloadcenter.intel.com/download/28040.
497 ESP-AH package is released, and it can be downloaded from
498 https://downloadcenter.intel.com/download/29446.
500 Load a profile which supports GTP and store backup profile:
502 .. code-block:: console
504 testpmd> ddp add 0 ./gtp.pkgo,./backup.pkgo
506 Delete a GTP profile and restore backup profile:
508 .. code-block:: console
510 testpmd> ddp del 0 ./backup.pkgo
512 Get loaded DDP package info list:
514 .. code-block:: console
516 testpmd> ddp get list 0
518 Display information about a GTP profile:
520 .. code-block:: console
522 testpmd> ddp get info ./gtp.pkgo
524 Input set configuration
525 ~~~~~~~~~~~~~~~~~~~~~~~
526 Input set for any PCTYPE can be configured with user defined configuration,
527 For example, to use only 48bit prefix for IPv6 src address for IPv6 TCP RSS:
529 .. code-block:: console
531 testpmd> port config 0 pctype 43 hash_inset clear all
532 testpmd> port config 0 pctype 43 hash_inset set field 13
533 testpmd> port config 0 pctype 43 hash_inset set field 14
534 testpmd> port config 0 pctype 43 hash_inset set field 15
536 Queue region configuration
537 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
538 The Intel® Ethernet 700 Series supports a feature of queue regions
539 configuration for RSS in the PF, so that different traffic classes or
540 different packet classification types can be separated to different
541 queues in different queue regions. There is an API for configuration
542 of queue regions in RSS with a command line. It can parse the parameters
543 of the region index, queue number, queue start index, user priority, traffic
544 classes and so on. Depending on commands from the command line, it will call
545 i40e private APIs and start the process of setting or flushing the queue
546 region configuration. As this feature is specific for i40e only private
547 APIs are used. These new ``test_pmd`` commands are as shown below. For
548 details please refer to :doc:`../testpmd_app_ug/index`.
550 .. code-block:: console
552 testpmd> set port (port_id) queue-region region_id (value) \
553 queue_start_index (value) queue_num (value)
554 testpmd> set port (port_id) queue-region region_id (value) flowtype (value)
555 testpmd> set port (port_id) queue-region UP (value) region_id (value)
556 testpmd> set port (port_id) queue-region flush (on|off)
557 testpmd> show port (port_id) queue-region
564 RSS Flow supports to set hash input set, hash function, enable hash
565 and configure queues.
567 Configure queues as queue 0, 1, 2, 3.
569 .. code-block:: console
571 testpmd> flow create 0 ingress pattern end actions rss types end \
572 queues 0 1 2 3 end / end
574 Enable hash and set input set for ipv4-tcp.
576 .. code-block:: console
578 testpmd> flow create 0 ingress pattern eth / ipv4 / tcp / end \
579 actions rss types ipv4-tcp l3-src-only end queues end / end
581 Set symmetric hash enable for flow type ipv4-tcp.
583 .. code-block:: console
585 testpmd> flow create 0 ingress pattern eth / ipv4 / tcp / end \
586 actions rss types ipv4-tcp end queues end func symmetric_toeplitz / end
588 Set hash function as simple xor.
590 .. code-block:: console
592 testpmd> flow create 0 ingress pattern end actions rss types end \
593 queues end func simple_xor / end
595 Limitations or Known issues
596 ---------------------------
598 MPLS packet classification
599 ~~~~~~~~~~~~~~~~~~~~~~~~~~
601 For firmware versions prior to 5.0, MPLS packets are not recognized by the NIC.
602 The L2 Payload flow type in flow director can be used to classify MPLS packet
603 by using a command in testpmd like:
605 testpmd> flow_director_filter 0 mode IP add flow l2_payload ether \
606 0x8847 flexbytes () fwd pf queue <N> fd_id <M>
608 With the NIC firmware version 5.0 or greater, some limited MPLS support
609 is added: Native MPLS (MPLS in Ethernet) skip is implemented, while no
610 new packet type, no classification or offload are possible. With this change,
611 L2 Payload flow type in flow director cannot be used to classify MPLS packet
612 as with previous firmware versions. Meanwhile, the Ethertype filter can be
613 used to classify MPLS packet by using a command in testpmd like:
615 testpmd> flow create 0 ingress pattern eth type is 0x8847 / end \
616 actions queue index <M> / end
618 16 Byte RX Descriptor setting on DPDK VF
619 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
621 Currently the VF's RX descriptor mode is decided by PF. There's no PF-VF
622 interface for VF to request the RX descriptor mode, also no interface to notify
623 VF its own RX descriptor mode.
624 For all available versions of the i40e driver, these drivers don't support 16
625 byte RX descriptor. If the Linux i40e kernel driver is used as host driver,
626 while DPDK i40e PMD is used as the VF driver, DPDK cannot choose 16 byte receive
627 descriptor. The reason is that the RX descriptor is already set to 32 byte by
628 the i40e kernel driver.
629 In the future, if the Linux i40e driver supports 16 byte RX descriptor, user
630 should make sure the DPDK VF uses the same RX descriptor mode, 16 byte or 32
631 byte, as the PF driver.
633 The same rule for DPDK PF + DPDK VF. The PF and VF should use the same RX
634 descriptor mode. Or the VF RX will not work.
636 Receive packets with Ethertype 0x88A8
637 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
639 Due to the FW limitation, PF can receive packets with Ethertype 0x88A8
640 only when floating VEB is disabled.
642 Incorrect Rx statistics when packet is oversize
643 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
645 When a packet is over maximum frame size, the packet is dropped.
646 However, the Rx statistics, when calling `rte_eth_stats_get` incorrectly
647 shows it as received.
649 RX/TX statistics may be incorrect when register overflowed
650 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
652 The rx_bytes/tx_bytes statistics register is 48 bit length.
653 Although this limitation is enlarged to 64 bit length on the software side,
654 but there is no way to detect if the overflow occurred more than once.
655 So rx_bytes/tx_bytes statistics data is correct when statistics are
656 updated at least once between two overflows.
658 VF & TC max bandwidth setting
659 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
661 The per VF max bandwidth and per TC max bandwidth cannot be enabled in parallel.
662 The behavior is different when handling per VF and per TC max bandwidth setting.
663 When enabling per VF max bandwidth, SW will check if per TC max bandwidth is
664 enabled. If so, return failure.
665 When enabling per TC max bandwidth, SW will check if per VF max bandwidth
666 is enabled. If so, disable per VF max bandwidth and continue with per TC max
669 TC TX scheduling mode setting
670 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
672 There are 2 TX scheduling modes for TCs, round robin and strict priority mode.
673 If a TC is set to strict priority mode, it can consume unlimited bandwidth.
674 It means if APP has set the max bandwidth for that TC, it comes to no
676 It's suggested to set the strict priority mode for a TC that is latency
677 sensitive but no consuming much bandwidth.
679 VF performance is impacted by PCI extended tag setting
680 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
682 To reach maximum NIC performance in the VF the PCI extended tag must be
683 enabled. The DPDK i40e PF driver will set this feature during initialization,
684 but the kernel PF driver does not. So when running traffic on a VF which is
685 managed by the kernel PF driver, a significant NIC performance downgrade has
686 been observed (for 64 byte packets, there is about 25% line-rate downgrade for
687 a 25GbE device and about 35% for a 40GbE device).
689 For kernel version >= 4.11, the kernel's PCI driver will enable the extended
690 tag if it detects that the device supports it. So by default, this is not an
691 issue. For kernels <= 4.11 or when the PCI extended tag is disabled it can be
692 enabled using the steps below.
694 #. Get the current value of the PCI configure register::
696 setpci -s <XX:XX.X> a8.w
700 value = value | 0x100
702 #. Set the PCI configure register with new value::
704 setpci -s <XX:XX.X> a8.w=<value>
709 The VF vlan strip function is only supported in the i40e kernel driver >= 2.1.26.
714 DCB works only when RSS is enabled.
716 Global configuration warning
717 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
719 I40E PMD will set some global registers to enable some function or set some
720 configure. Then when using different ports of the same NIC with Linux kernel
721 and DPDK, the port with Linux kernel will be impacted by the port with DPDK.
722 For example, register I40E_GL_SWT_L2TAGCTRL is used to control L2 tag, i40e
723 PMD uses I40E_GL_SWT_L2TAGCTRL to set vlan TPID. If setting TPID in port A
724 with DPDK, then the configuration will also impact port B in the NIC with
725 kernel driver, which don't want to use the TPID.
726 So PMD reports warning to clarify what is changed by writing global register.
731 When programming cloud filters for IPv4/6_UDP/TCP/SCTP with SRC port only or DST port only,
732 it will make any cloud filter using inner_vlan or tunnel key invalid. Default configuration will be
733 recovered only by NIC core reset.
735 High Performance of Small Packets on 40GbE NIC
736 ----------------------------------------------
738 As there might be firmware fixes for performance enhancement in latest version
739 of firmware image, the firmware update might be needed for getting high performance.
740 Check the Intel support website for the latest firmware updates.
741 Users should consult the release notes specific to a DPDK release to identify
742 the validated firmware version for a NIC using the i40e driver.
744 Use 16 Bytes RX Descriptor Size
745 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
747 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.
748 In ``config/rte_config.h`` set the following to use 16 bytes size RX descriptors::
750 #define RTE_LIBRTE_I40E_16BYTE_RX_DESC 1
752 Input set requirement of each pctype for FDIR
753 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
755 Each PCTYPE can only have one specific FDIR input set at one time.
756 For example, if creating 2 rte_flow rules with different input set for one PCTYPE,
757 it will fail and return the info "Conflict with the first rule's input set",
758 which means the current rule's input set conflicts with the first rule's.
759 Remove the first rule if want to change the input set of the PCTYPE.
761 Example of getting best performance with l3fwd example
762 ------------------------------------------------------
764 The following is an example of running the DPDK ``l3fwd`` sample application to get high performance with a
765 server with Intel Xeon processors and Intel Ethernet CNA XL710.
767 The example scenario is to get best performance with two Intel Ethernet CNA XL710 40GbE ports.
768 See :numref:`figure_intel_perf_test_setup` for the performance test setup.
770 .. _figure_intel_perf_test_setup:
772 .. figure:: img/intel_perf_test_setup.*
774 Performance Test Setup
777 1. Add two Intel Ethernet CNA XL710 to the platform, and use one port per card to get best performance.
778 The reason for using two NICs is to overcome a PCIe v3.0 limitation since it cannot provide 80GbE bandwidth
779 for two 40GbE ports, but two different PCIe v3.0 x8 slot can.
780 Refer to the sample NICs output above, then we can select ``82:00.0`` and ``85:00.0`` as test ports::
782 82:00.0 Ethernet [0200]: Intel XL710 for 40GbE QSFP+ [8086:1583]
783 85:00.0 Ethernet [0200]: Intel XL710 for 40GbE QSFP+ [8086:1583]
785 2. Connect the ports to the traffic generator. For high speed testing, it's best to use a hardware traffic generator.
787 3. Check the PCI devices numa node (socket id) and get the cores number on the exact socket id.
788 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
790 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
791 cores from different cores (e.g core18 and core19).
793 4. Bind these two ports to igb_uio.
795 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
796 will be required, and each queue pair will need a dedicated CPU core for receiving/transmitting packets.
798 6. The DPDK sample application ``l3fwd`` will be used for performance testing, with using two ports for bi-directional forwarding.
799 Compile the ``l3fwd sample`` with the default lpm mode.
801 7. The command line of running l3fwd would be something like the following::
803 ./dpdk-l3fwd -l 18-21 -n 4 -a 82:00.0 -a 85:00.0 \
804 -- -p 0x3 --config '(0,0,18),(0,1,19),(1,0,20),(1,1,21)'
806 This means that the application uses core 18 for port 0, queue pair 0 forwarding, core 19 for port 0, queue pair 1 forwarding,
807 core 20 for port 1, queue pair 0 forwarding, and core 21 for port 1, queue pair 1 forwarding.
809 8. Configure the traffic at a traffic generator.
811 * Start creating a stream on packet generator.
813 * Set the Ethernet II type to 0x0800.
815 Tx bytes affected by the link status change
816 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
818 For firmware versions prior to 6.01 for X710 series and 3.33 for X722 series, the tx_bytes statistics data is affected by
819 the link down event. Each time the link status changes to down, the tx_bytes decreases 110 bytes.