1.. BSD LICENSE 2 Copyright(c) 2010-2016 Intel Corporation. All rights reserved. 3 All rights reserved. 4 5 Redistribution and use in source and binary forms, with or without 6 modification, are permitted provided that the following conditions 7 are met: 8 9 * Redistributions of source code must retain the above copyright 10 notice, this list of conditions and the following disclaimer. 11 * Redistributions in binary form must reproduce the above copyright 12 notice, this list of conditions and the following disclaimer in 13 the documentation and/or other materials provided with the 14 distribution. 15 * Neither the name of Intel Corporation nor the names of its 16 contributors may be used to endorse or promote products derived 17 from this software without specific prior written permission. 18 19 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS 20 "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT 21 LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR 22 A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT 23 OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, 24 SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT 25 LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 26 DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 27 THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 28 (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE 29 OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 30 31.. _testpmd_runtime: 32 33Testpmd Runtime Functions 34========================= 35 36Where the testpmd application is started in interactive mode, (``-i|--interactive``), 37it displays a prompt that can be used to start and stop forwarding, 38configure the application, display statistics (including the extended NIC 39statistics aka xstats) , set the Flow Director and other tasks:: 40 41 testpmd> 42 43The testpmd prompt has some, limited, readline support. 44Common bash command-line functions such as ``Ctrl+a`` and ``Ctrl+e`` to go to the start and end of the prompt line are supported 45as well as access to the command history via the up-arrow. 46 47There is also support for tab completion. 48If you type a partial command and hit ``<TAB>`` you get a list of the available completions: 49 50.. code-block:: console 51 52 testpmd> show port <TAB> 53 54 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap X 55 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap all 56 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap X 57 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap all 58 ... 59 60 61.. note:: 62 63 Some examples in this document are too long to fit on one line are are shown wrapped at `"\\"` for display purposes:: 64 65 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \ 66 (pause_time) (send_xon) (port_id) 67 68In the real ``testpmd>`` prompt these commands should be on a single line. 69 70Help Functions 71-------------- 72 73The testpmd has on-line help for the functions that are available at runtime. 74These are divided into sections and can be accessed using help, help section or help all: 75 76.. code-block:: console 77 78 testpmd> help 79 80 help control : Start and stop forwarding. 81 help display : Displaying port, stats and config information. 82 help config : Configuration information. 83 help ports : Configuring ports. 84 help registers : Reading and setting port registers. 85 help filters : Filters configuration help. 86 help all : All of the above sections. 87 88 89Control Functions 90----------------- 91 92start 93~~~~~ 94 95Start packet forwarding with current configuration:: 96 97 testpmd> start 98 99start tx_first 100~~~~~~~~~~~~~~ 101 102Start packet forwarding with current configuration after sending specified number of bursts of packets:: 103 104 testpmd> start tx_first (""|burst_num) 105 106The default burst number is 1 when ``burst_num`` not presented. 107 108stop 109~~~~ 110 111Stop packet forwarding, and display accumulated statistics:: 112 113 testpmd> stop 114 115quit 116~~~~ 117 118Quit to prompt:: 119 120 testpmd> quit 121 122 123Display Functions 124----------------- 125 126The functions in the following sections are used to display information about the 127testpmd configuration or the NIC status. 128 129show port 130~~~~~~~~~ 131 132Display information for a given port or all ports:: 133 134 testpmd> show port (info|stats|xstats|fdir|stat_qmap|dcb_tc|cap) (port_id|all) 135 136The available information categories are: 137 138* ``info``: General port information such as MAC address. 139 140* ``stats``: RX/TX statistics. 141 142* ``xstats``: RX/TX extended NIC statistics. 143 144* ``fdir``: Flow Director information and statistics. 145 146* ``stat_qmap``: Queue statistics mapping. 147 148* ``dcb_tc``: DCB information such as TC mapping. 149 150* ``cap``: Supported offload capabilities. 151 152For example: 153 154.. code-block:: console 155 156 testpmd> show port info 0 157 158 ********************* Infos for port 0 ********************* 159 160 MAC address: XX:XX:XX:XX:XX:XX 161 Connect to socket: 0 162 memory allocation on the socket: 0 163 Link status: up 164 Link speed: 40000 Mbps 165 Link duplex: full-duplex 166 Promiscuous mode: enabled 167 Allmulticast mode: disabled 168 Maximum number of MAC addresses: 64 169 Maximum number of MAC addresses of hash filtering: 0 170 VLAN offload: 171 strip on 172 filter on 173 qinq(extend) off 174 Redirection table size: 512 175 Supported flow types: 176 ipv4-frag 177 ipv4-tcp 178 ipv4-udp 179 ipv4-sctp 180 ipv4-other 181 ipv6-frag 182 ipv6-tcp 183 ipv6-udp 184 ipv6-sctp 185 ipv6-other 186 l2_payload 187 port 188 vxlan 189 geneve 190 nvgre 191 192show port rss reta 193~~~~~~~~~~~~~~~~~~ 194 195Display the rss redirection table entry indicated by masks on port X:: 196 197 testpmd> show port (port_id) rss reta (size) (mask0, mask1...) 198 199size is used to indicate the hardware supported reta size 200 201show port rss-hash 202~~~~~~~~~~~~~~~~~~ 203 204Display the RSS hash functions and RSS hash key of a port:: 205 206 testpmd> show port (port_id) rss-hash ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2-payload|ipv6-ex|ipv6-tcp-ex|ipv6-udp-ex [key] 207 208clear port 209~~~~~~~~~~ 210 211Clear the port statistics for a given port or for all ports:: 212 213 testpmd> clear port (info|stats|xstats|fdir|stat_qmap) (port_id|all) 214 215For example:: 216 217 testpmd> clear port stats all 218 219show (rxq|txq) 220~~~~~~~~~~~~~~ 221 222Display information for a given port's RX/TX queue:: 223 224 testpmd> show (rxq|txq) info (port_id) (queue_id) 225 226show config 227~~~~~~~~~~~ 228 229Displays the configuration of the application. 230The configuration comes from the command-line, the runtime or the application defaults:: 231 232 testpmd> show config (rxtx|cores|fwd|txpkts) 233 234The available information categories are: 235 236* ``rxtx``: RX/TX configuration items. 237 238* ``cores``: List of forwarding cores. 239 240* ``fwd``: Packet forwarding configuration. 241 242* ``txpkts``: Packets to TX configuration. 243 244For example: 245 246.. code-block:: console 247 248 testpmd> show config rxtx 249 250 io packet forwarding - CRC stripping disabled - packets/burst=16 251 nb forwarding cores=2 - nb forwarding ports=1 252 RX queues=1 - RX desc=128 - RX free threshold=0 253 RX threshold registers: pthresh=8 hthresh=8 wthresh=4 254 TX queues=1 - TX desc=512 - TX free threshold=0 255 TX threshold registers: pthresh=36 hthresh=0 wthresh=0 256 TX RS bit threshold=0 - TXQ flags=0x0 257 258set fwd 259~~~~~~~ 260 261Set the packet forwarding mode:: 262 263 testpmd> set fwd (io|mac|macswap|flowgen| \ 264 rxonly|txonly|csum|icmpecho) (""|retry) 265 266``retry`` can be specified for forwarding engines except ``rx_only``. 267 268The available information categories are: 269 270* ``io``: Forwards packets "as-is" in I/O mode. 271 This is the fastest possible forwarding operation as it does not access packets data. 272 This is the default mode. 273 274* ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them. 275 Default application behaviour is to set source Ethernet address to that of the transmitting interface, and destination 276 address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or 277 'eth-peer-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address. 278 279* ``macswap``: MAC swap forwarding mode. 280 Swaps the source and the destination Ethernet addresses of packets before forwarding them. 281 282* ``flowgen``: Multi-flow generation mode. 283 Originates a number of flows (with varying destination IP addresses), and terminate receive traffic. 284 285* ``rxonly``: Receives packets but doesn't transmit them. 286 287* ``txonly``: Generates and transmits packets without receiving any. 288 289* ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet. 290 291* ``icmpecho``: Receives a burst of packets, lookup for IMCP echo requests and, if any, send back ICMP echo replies. 292 293* ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX. Requires ``CONFIG_RTE_LIBRTE_IEEE1588=y``. 294 295Note: TX timestamping is only available in the "Full Featured" TX path. To force ``testpmd`` into this mode set ``--txqflags=0``. 296 297Example:: 298 299 testpmd> set fwd rxonly 300 301 Set rxonly packet forwarding mode 302 303 304read rxd 305~~~~~~~~ 306 307Display an RX descriptor for a port RX queue:: 308 309 testpmd> read rxd (port_id) (queue_id) (rxd_id) 310 311For example:: 312 313 testpmd> read rxd 0 0 4 314 0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180 315 316read txd 317~~~~~~~~ 318 319Display a TX descriptor for a port TX queue:: 320 321 testpmd> read txd (port_id) (queue_id) (txd_id) 322 323For example:: 324 325 testpmd> read txd 0 0 4 326 0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C 327 328 329Configuration Functions 330----------------------- 331 332The testpmd application can be configured from the runtime as well as from the command-line. 333 334This section details the available configuration functions that are available. 335 336.. note:: 337 338 Configuration changes only become active when forwarding is started/restarted. 339 340set default 341~~~~~~~~~~~ 342 343Reset forwarding to the default configuration:: 344 345 testpmd> set default 346 347set verbose 348~~~~~~~~~~~ 349 350Set the debug verbosity level:: 351 352 testpmd> set verbose (level) 353 354Currently the only available levels are 0 (silent except for error) and 1 (fully verbose). 355 356set nbport 357~~~~~~~~~~ 358 359Set the number of ports used by the application: 360 361set nbport (num) 362 363This is equivalent to the ``--nb-ports`` command-line option. 364 365set nbcore 366~~~~~~~~~~ 367 368Set the number of cores used by the application:: 369 370 testpmd> set nbcore (num) 371 372This is equivalent to the ``--nb-cores`` command-line option. 373 374.. note:: 375 376 The number of cores used must not be greater than number of ports used multiplied by the number of queues per port. 377 378set coremask 379~~~~~~~~~~~~ 380 381Set the forwarding cores hexadecimal mask:: 382 383 testpmd> set coremask (mask) 384 385This is equivalent to the ``--coremask`` command-line option. 386 387.. note:: 388 389 The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding. 390 391set portmask 392~~~~~~~~~~~~ 393 394Set the forwarding ports hexadecimal mask:: 395 396 testpmd> set portmask (mask) 397 398This is equivalent to the ``--portmask`` command-line option. 399 400set burst 401~~~~~~~~~ 402 403Set number of packets per burst:: 404 405 testpmd> set burst (num) 406 407This is equivalent to the ``--burst command-line`` option. 408 409When retry is enabled, the transmit delay time and number of retries can also be set:: 410 411 testpmd> set burst tx delay (microseconds) retry (num) 412 413set txpkts 414~~~~~~~~~~ 415 416Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode:: 417 418 testpmd> set txpkts (x[,y]*) 419 420Where x[,y]* represents a CSV list of values, without white space. 421 422set txsplit 423~~~~~~~~~~~ 424 425Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes:: 426 427 testpmd> set txsplit (off|on|rand) 428 429Where: 430 431* ``off`` disable packet copy & split for CSUM mode. 432 433* ``on`` split outgoing packet into multiple segments. Size of each segment 434 and number of segments per packet is determined by ``set txpkts`` command 435 (see above). 436 437* ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments. 438 439set corelist 440~~~~~~~~~~~~ 441 442Set the list of forwarding cores:: 443 444 testpmd> set corelist (x[,y]*) 445 446For example, to change the forwarding cores: 447 448.. code-block:: console 449 450 testpmd> set corelist 3,1 451 testpmd> show config fwd 452 453 io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled 454 Logical Core 3 (socket 0) forwards packets on 1 streams: 455 RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01 456 Logical Core 1 (socket 0) forwards packets on 1 streams: 457 RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00 458 459.. note:: 460 461 The cores are used in the same order as specified on the command line. 462 463set portlist 464~~~~~~~~~~~~ 465 466Set the list of forwarding ports:: 467 468 testpmd> set portlist (x[,y]*) 469 470For example, to change the port forwarding: 471 472.. code-block:: console 473 474 testpmd> set portlist 0,2,1,3 475 testpmd> show config fwd 476 477 io packet forwarding - ports=4 - cores=1 - streams=4 478 Logical Core 3 (socket 0) forwards packets on 4 streams: 479 RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01 480 RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00 481 RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03 482 RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02 483 484set tx loopback 485~~~~~~~~~~~~~~~ 486 487Enable/disable tx loopback:: 488 489 testpmd> set tx loopback (port_id) (on|off) 490 491set drop enable 492~~~~~~~~~~~~~~~ 493 494set drop enable bit for all queues:: 495 496 testpmd> set all queues drop (port_id) (on|off) 497 498set split drop enable (for VF) 499~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 500 501set split drop enable bit for VF from PF:: 502 503 testpmd> set vf split drop (port_id) (vf_id) (on|off) 504 505set mac antispoof (for VF) 506~~~~~~~~~~~~~~~~~~~~~~~~~~ 507 508Set mac antispoof for a VF from the PF:: 509 510 testpmd> set vf mac antispoof (port_id) (vf_id) (on|off) 511 512set macsec offload 513~~~~~~~~~~~~~~~~~~ 514 515Enable/disable MACsec offload:: 516 517 testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off) 518 testpmd> set macsec offload (port_id) off 519 520set macsec sc 521~~~~~~~~~~~~~ 522 523Configure MACsec secure connection (SC):: 524 525 testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi) 526 527.. note:: 528 529 The pi argument is ignored for tx. 530 Check the NIC Datasheet for hardware limits. 531 532set macsec sa 533~~~~~~~~~~~~~ 534 535Configure MACsec secure association (SA):: 536 537 testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key) 538 539.. note:: 540 541 The IDX value must be 0 or 1. 542 Check the NIC Datasheet for hardware limits. 543 544set broadcast mode (for VF) 545~~~~~~~~~~~~~~~~~~~~~~~~~~~ 546 547Set broadcast mode for a VF from the PF:: 548 549 testpmd> set vf broadcast (port_id) (vf_id) (on|off) 550 551vlan set strip 552~~~~~~~~~~~~~~ 553 554Set the VLAN strip on a port:: 555 556 testpmd> vlan set strip (on|off) (port_id) 557 558vlan set stripq 559~~~~~~~~~~~~~~~ 560 561Set the VLAN strip for a queue on a port:: 562 563 testpmd> vlan set stripq (on|off) (port_id,queue_id) 564 565vlan set stripq (for VF) 566~~~~~~~~~~~~~~~~~~~~~~~~ 567 568Set VLAN strip for all queues in a pool for a VF from the PF:: 569 570 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off) 571 572vlan set insert (for VF) 573~~~~~~~~~~~~~~~~~~~~~~~~ 574 575Set VLAN insert for a VF from the PF:: 576 577 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id) 578 579vlan set tag (for VF) 580~~~~~~~~~~~~~~~~~~~~~ 581 582Set VLAN tag for a VF from the PF:: 583 584 testpmd> set vf vlan tag (port_id) (vf_id) (on|off) 585 586vlan set antispoof (for VF) 587~~~~~~~~~~~~~~~~~~~~~~~~~~~ 588 589Set VLAN antispoof for a VF from the PF:: 590 591 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off) 592 593vlan set filter 594~~~~~~~~~~~~~~~ 595 596Set the VLAN filter on a port:: 597 598 testpmd> vlan set filter (on|off) (port_id) 599 600vlan set qinq 601~~~~~~~~~~~~~ 602 603Set the VLAN QinQ (extended queue in queue) on for a port:: 604 605 testpmd> vlan set qinq (on|off) (port_id) 606 607vlan set tpid 608~~~~~~~~~~~~~ 609 610Set the inner or outer VLAN TPID for packet filtering on a port:: 611 612 testpmd> vlan set (inner|outer) tpid (value) (port_id) 613 614.. note:: 615 616 TPID value must be a 16-bit number (value <= 65536). 617 618rx_vlan add 619~~~~~~~~~~~ 620 621Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID:: 622 623 testpmd> rx_vlan add (vlan_id|all) (port_id) 624 625.. note:: 626 627 VLAN filter must be set on that port. VLAN ID < 4096. 628 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries 629 in VFTA table. This is important if enabling all vlan_ids. 630 631rx_vlan rm 632~~~~~~~~~~ 633 634Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID:: 635 636 testpmd> rx_vlan rm (vlan_id|all) (port_id) 637 638rx_vlan add (for VF) 639~~~~~~~~~~~~~~~~~~~~ 640 641Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID:: 642 643 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask) 644 645rx_vlan rm (for VF) 646~~~~~~~~~~~~~~~~~~~ 647 648Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID:: 649 650 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask) 651 652tunnel_filter add 653~~~~~~~~~~~~~~~~~ 654 655Add a tunnel filter on a port:: 656 657 testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \ 658 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\ 659 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id) 660 661The available information categories are: 662 663* ``vxlan``: Set tunnel type as VXLAN. 664 665* ``nvgre``: Set tunnel type as NVGRE. 666 667* ``ipingre``: Set tunnel type as IP-in-GRE. 668 669* ``imac-ivlan``: Set filter type as Inner MAC and VLAN. 670 671* ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID. 672 673* ``imac-tenid``: Set filter type as Inner MAC and tenant ID. 674 675* ``imac``: Set filter type as Inner MAC. 676 677* ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID. 678 679* ``oip``: Set filter type as Outer IP. 680 681* ``iip``: Set filter type as Inner IP. 682 683Example:: 684 685 testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \ 686 192.168.2.2 0 ipingre oip 1 1 687 688 Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP. 689 690tunnel_filter remove 691~~~~~~~~~~~~~~~~~~~~ 692 693Remove a tunnel filter on a port:: 694 695 testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \ 696 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\ 697 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id) 698 699rx_vxlan_port add 700~~~~~~~~~~~~~~~~~ 701 702Add an UDP port for VXLAN packet filter on a port:: 703 704 testpmd> rx_vxlan_port add (udp_port) (port_id) 705 706rx_vxlan_port remove 707~~~~~~~~~~~~~~~~~~~~ 708 709Remove an UDP port for VXLAN packet filter on a port:: 710 711 testpmd> rx_vxlan_port rm (udp_port) (port_id) 712 713tx_vlan set 714~~~~~~~~~~~ 715 716Set hardware insertion of VLAN IDs in packets sent on a port:: 717 718 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer] 719 720For example, set a single VLAN ID (5) insertion on port 0:: 721 722 tx_vlan set 0 5 723 724Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1:: 725 726 tx_vlan set 1 2 3 727 728 729tx_vlan set pvid 730~~~~~~~~~~~~~~~~ 731 732Set port based hardware insertion of VLAN ID in packets sent on a port:: 733 734 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off) 735 736tx_vlan reset 737~~~~~~~~~~~~~ 738 739Disable hardware insertion of a VLAN header in packets sent on a port:: 740 741 testpmd> tx_vlan reset (port_id) 742 743csum set 744~~~~~~~~ 745 746Select hardware or software calculation of the checksum when 747transmitting a packet using the ``csum`` forwarding engine:: 748 749 testpmd> csum set (ip|udp|tcp|sctp|outer-ip) (hw|sw) (port_id) 750 751Where: 752 753* ``ip|udp|tcp|sctp`` always relate to the inner layer. 754 755* ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized 756 as a tunnel packet by the forwarding engine (vxlan, gre and ipip are 757 supported). See also the ``csum parse-tunnel`` command. 758 759.. note:: 760 761 Check the NIC Datasheet for hardware limits. 762 763csum parse-tunnel 764~~~~~~~~~~~~~~~~~ 765 766Define how tunneled packets should be handled by the csum forward 767engine:: 768 769 testpmd> csum parse-tunnel (on|off) (tx_port_id) 770 771If enabled, the csum forward engine will try to recognize supported 772tunnel headers (vxlan, gre, ipip). 773 774If disabled, treat tunnel packets as non-tunneled packets (a inner 775header is handled as a packet payload). 776 777.. note:: 778 779 The port argument is the TX port like in the ``csum set`` command. 780 781Example: 782 783Consider a packet in packet like the following:: 784 785 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in 786 787* If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set`` 788 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the 789 ``outer-ip parameter`` relates to the outer headers (here ``ipv4_out``). 790 791* If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set`` 792 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``. 793 794csum show 795~~~~~~~~~ 796 797Display tx checksum offload configuration:: 798 799 testpmd> csum show (port_id) 800 801tso set 802~~~~~~~ 803 804Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine:: 805 806 testpmd> tso set (segsize) (port_id) 807 808.. note:: 809 810 Check the NIC datasheet for hardware limits. 811 812tso show 813~~~~~~~~ 814 815Display the status of TCP Segmentation Offload:: 816 817 testpmd> tso show (port_id) 818 819mac_addr add 820~~~~~~~~~~~~ 821 822Add an alternative MAC address to a port:: 823 824 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX) 825 826mac_addr remove 827~~~~~~~~~~~~~~~ 828 829Remove a MAC address from a port:: 830 831 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX) 832 833mac_addr add (for VF) 834~~~~~~~~~~~~~~~~~~~~~ 835 836Add an alternative MAC address for a VF to a port:: 837 838 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX) 839 840mac_addr set 841~~~~~~~~~~~~ 842 843Set the default MAC address for a port:: 844 845 testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX) 846 847mac_addr set (for VF) 848~~~~~~~~~~~~~~~~~~~~~ 849 850Set the MAC address for a VF from the PF:: 851 852 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX) 853 854set port-uta 855~~~~~~~~~~~~ 856 857Set the unicast hash filter(s) on/off for a port:: 858 859 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off) 860 861set promisc 862~~~~~~~~~~~ 863 864Set the promiscuous mode on for a port or for all ports. 865In promiscuous mode packets are not dropped if they aren't for the specified MAC address:: 866 867 testpmd> set promisc (port_id|all) (on|off) 868 869set allmulti 870~~~~~~~~~~~~ 871 872Set the allmulti mode for a port or for all ports:: 873 874 testpmd> set allmulti (port_id|all) (on|off) 875 876Same as the ifconfig (8) option. Controls how multicast packets are handled. 877 878set promisc (for VF) 879~~~~~~~~~~~~~~~~~~~~ 880 881Set the unicast promiscuous mode for a VF from PF. 882It's supported by Intel i40e NICs now. 883In promiscuous mode packets are not dropped if they aren't for the specified MAC address:: 884 885 testpmd> set vf promisc (port_id) (vf_id) (on|off) 886 887set allmulticast (for VF) 888~~~~~~~~~~~~~~~~~~~~~~~~~ 889 890Set the multicast promiscuous mode for a VF from PF. 891It's supported by Intel i40e NICs now. 892In promiscuous mode packets are not dropped if they aren't for the specified MAC address:: 893 894 testpmd> set vf allmulti (port_id) (vf_id) (on|off) 895 896set tx max bandwidth (for VF) 897~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 898 899Set TX max absolute bandwidth (Mbps) for a VF from PF:: 900 901 testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth) 902 903set tc tx min bandwidth (for VF) 904~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 905 906Set all TCs' TX min relative bandwidth (%) for a VF from PF:: 907 908 testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...) 909 910set tc tx max bandwidth (for VF) 911~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 912 913Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF:: 914 915 testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth) 916 917set tc strict link priority mode 918~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 919 920Set some TCs' strict link priority mode on a physical port:: 921 922 testpmd> set tx strict-link-priority (port_id) (tc_bitmap) 923 924set tc tx min bandwidth 925~~~~~~~~~~~~~~~~~~~~~~~ 926 927Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs:: 928 929 testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...) 930 931set flow_ctrl rx 932~~~~~~~~~~~~~~~~ 933 934Set the link flow control parameter on a port:: 935 936 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \ 937 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \ 938 autoneg (on|off) (port_id) 939 940Where: 941 942* ``high_water`` (integer): High threshold value to trigger XOFF. 943 944* ``low_water`` (integer): Low threshold value to trigger XON. 945 946* ``pause_time`` (integer): Pause quota in the Pause frame. 947 948* ``send_xon`` (0/1): Send XON frame. 949 950* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames. 951 952* ``autoneg``: Change the auto-negotiation parameter. 953 954set pfc_ctrl rx 955~~~~~~~~~~~~~~~ 956 957Set the priority flow control parameter on a port:: 958 959 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \ 960 (pause_time) (priority) (port_id) 961 962Where: 963 964* ``high_water`` (integer): High threshold value. 965 966* ``low_water`` (integer): Low threshold value. 967 968* ``pause_time`` (integer): Pause quota in the Pause frame. 969 970* ``priority`` (0-7): VLAN User Priority. 971 972set stat_qmap 973~~~~~~~~~~~~~ 974 975Set statistics mapping (qmapping 0..15) for RX/TX queue on port:: 976 977 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping) 978 979For example, to set rx queue 2 on port 0 to mapping 5:: 980 981 testpmd>set stat_qmap rx 0 2 5 982 983set port - rx/tx (for VF) 984~~~~~~~~~~~~~~~~~~~~~~~~~ 985 986Set VF receive/transmit from a port:: 987 988 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off) 989 990set port - mac address filter (for VF) 991~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 992 993Add/Remove unicast or multicast MAC addr filter for a VF:: 994 995 testpmd> set port (port_id) vf (vf_id) (mac_addr) \ 996 (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off) 997 998set port - rx mode(for VF) 999~~~~~~~~~~~~~~~~~~~~~~~~~~ 1000 1001Set the VF receive mode of a port:: 1002 1003 testpmd> set port (port_id) vf (vf_id) \ 1004 rxmode (AUPE|ROPE|BAM|MPE) (on|off) 1005 1006The available receive modes are: 1007 1008* ``AUPE``: Accepts untagged VLAN. 1009 1010* ``ROPE``: Accepts unicast hash. 1011 1012* ``BAM``: Accepts broadcast packets. 1013 1014* ``MPE``: Accepts all multicast packets. 1015 1016set port - tx_rate (for Queue) 1017~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1018 1019Set TX rate limitation for a queue on a port:: 1020 1021 testpmd> set port (port_id) queue (queue_id) rate (rate_value) 1022 1023set port - tx_rate (for VF) 1024~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1025 1026Set TX rate limitation for queues in VF on a port:: 1027 1028 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask) 1029 1030set port - mirror rule 1031~~~~~~~~~~~~~~~~~~~~~~ 1032 1033Set pool or vlan type mirror rule for a port:: 1034 1035 testpmd> set port (port_id) mirror-rule (rule_id) \ 1036 (pool-mirror-up|pool-mirror-down|vlan-mirror) \ 1037 (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off) 1038 1039Set link mirror rule for a port:: 1040 1041 testpmd> set port (port_id) mirror-rule (rule_id) \ 1042 (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off) 1043 1044For example to enable mirror traffic with vlan 0,1 to pool 0:: 1045 1046 set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on 1047 1048reset port - mirror rule 1049~~~~~~~~~~~~~~~~~~~~~~~~ 1050 1051Reset a mirror rule for a port:: 1052 1053 testpmd> reset port (port_id) mirror-rule (rule_id) 1054 1055set flush_rx 1056~~~~~~~~~~~~ 1057 1058Set the flush on RX streams before forwarding. 1059The default is flush ``on``. 1060Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams:: 1061 1062 testpmd> set flush_rx off 1063 1064set bypass mode 1065~~~~~~~~~~~~~~~ 1066 1067Set the bypass mode for the lowest port on bypass enabled NIC:: 1068 1069 testpmd> set bypass mode (normal|bypass|isolate) (port_id) 1070 1071set bypass event 1072~~~~~~~~~~~~~~~~ 1073 1074Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled:: 1075 1076 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \ 1077 mode (normal|bypass|isolate) (port_id) 1078 1079Where: 1080 1081* ``timeout``: Enable bypass after watchdog timeout. 1082 1083* ``os_on``: Enable bypass when OS/board is powered on. 1084 1085* ``os_off``: Enable bypass when OS/board is powered off. 1086 1087* ``power_on``: Enable bypass when power supply is turned on. 1088 1089* ``power_off``: Enable bypass when power supply is turned off. 1090 1091 1092set bypass timeout 1093~~~~~~~~~~~~~~~~~~ 1094 1095Set the bypass watchdog timeout to ``n`` seconds where 0 = instant:: 1096 1097 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32) 1098 1099show bypass config 1100~~~~~~~~~~~~~~~~~~ 1101 1102Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC:: 1103 1104 testpmd> show bypass config (port_id) 1105 1106set link up 1107~~~~~~~~~~~ 1108 1109Set link up for a port:: 1110 1111 testpmd> set link-up port (port id) 1112 1113set link down 1114~~~~~~~~~~~~~ 1115 1116Set link down for a port:: 1117 1118 testpmd> set link-down port (port id) 1119 1120E-tag set 1121~~~~~~~~~ 1122 1123Enable E-tag insertion for a VF on a port:: 1124 1125 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id) 1126 1127Disable E-tag insertion for a VF on a port:: 1128 1129 testpmd> E-tag set insertion off port (port_id) vf (vf_id) 1130 1131Enable/disable E-tag stripping on a port:: 1132 1133 testpmd> E-tag set stripping (on|off) port (port_id) 1134 1135Enable/disable E-tag based forwarding on a port:: 1136 1137 testpmd> E-tag set forwarding (on|off) port (port_id) 1138 1139Add an E-tag forwarding filter on a port:: 1140 1141 testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id) 1142 1143Delete an E-tag forwarding filter on a port:: 1144 testpmd> E-tag set filter del e-tag-id (value) port (port_id) 1145 1146 1147Port Functions 1148-------------- 1149 1150The following sections show functions for configuring ports. 1151 1152.. note:: 1153 1154 Port configuration changes only become active when forwarding is started/restarted. 1155 1156port attach 1157~~~~~~~~~~~ 1158 1159Attach a port specified by pci address or virtual device args:: 1160 1161 testpmd> port attach (identifier) 1162 1163To attach a new pci device, the device should be recognized by kernel first. 1164Then it should be moved under DPDK management. 1165Finally the port can be attached to testpmd. 1166 1167For example, to move a pci device using ixgbe under DPDK management: 1168 1169.. code-block:: console 1170 1171 # Check the status of the available devices. 1172 ./usertools/dpdk-devbind.py --status 1173 1174 Network devices using DPDK-compatible driver 1175 ============================================ 1176 <none> 1177 1178 Network devices using kernel driver 1179 =================================== 1180 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused= 1181 1182 1183 # Bind the device to igb_uio. 1184 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0 1185 1186 1187 # Recheck the status of the devices. 1188 ./usertools/dpdk-devbind.py --status 1189 Network devices using DPDK-compatible driver 1190 ============================================ 1191 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused= 1192 1193To attach a port created by virtual device, above steps are not needed. 1194 1195For example, to attach a port whose pci address is 0000:0a:00.0. 1196 1197.. code-block:: console 1198 1199 testpmd> port attach 0000:0a:00.0 1200 Attaching a new port... 1201 EAL: PCI device 0000:0a:00.0 on NUMA socket -1 1202 EAL: probe driver: 8086:10fb rte_ixgbe_pmd 1203 EAL: PCI memory mapped at 0x7f83bfa00000 1204 EAL: PCI memory mapped at 0x7f83bfa80000 1205 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5 1206 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb 1207 Port 0 is attached. Now total ports is 1 1208 Done 1209 1210For example, to attach a port created by pcap PMD. 1211 1212.. code-block:: console 1213 1214 testpmd> port attach net_pcap0 1215 Attaching a new port... 1216 PMD: Initializing pmd_pcap for net_pcap0 1217 PMD: Creating pcap-backed ethdev on numa socket 0 1218 Port 0 is attached. Now total ports is 1 1219 Done 1220 1221In this case, identifier is ``net_pcap0``. 1222This identifier format is the same as ``--vdev`` format of DPDK applications. 1223 1224For example, to re-attach a bonded port which has been previously detached, 1225the mode and slave parameters must be given. 1226 1227.. code-block:: console 1228 1229 testpmd> port attach net_bond_0,mode=0,slave=1 1230 Attaching a new port... 1231 EAL: Initializing pmd_bond for net_bond_0 1232 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0. 1233 Port 0 is attached. Now total ports is 1 1234 Done 1235 1236 1237port detach 1238~~~~~~~~~~~ 1239 1240Detach a specific port:: 1241 1242 testpmd> port detach (port_id) 1243 1244Before detaching a port, the port should be stopped and closed. 1245 1246For example, to detach a pci device port 0. 1247 1248.. code-block:: console 1249 1250 testpmd> port stop 0 1251 Stopping ports... 1252 Done 1253 testpmd> port close 0 1254 Closing ports... 1255 Done 1256 1257 testpmd> port detach 0 1258 Detaching a port... 1259 EAL: PCI device 0000:0a:00.0 on NUMA socket -1 1260 EAL: remove driver: 8086:10fb rte_ixgbe_pmd 1261 EAL: PCI memory unmapped at 0x7f83bfa00000 1262 EAL: PCI memory unmapped at 0x7f83bfa80000 1263 Done 1264 1265 1266For example, to detach a virtual device port 0. 1267 1268.. code-block:: console 1269 1270 testpmd> port stop 0 1271 Stopping ports... 1272 Done 1273 testpmd> port close 0 1274 Closing ports... 1275 Done 1276 1277 testpmd> port detach 0 1278 Detaching a port... 1279 PMD: Closing pcap ethdev on numa socket 0 1280 Port 'net_pcap0' is detached. Now total ports is 0 1281 Done 1282 1283To remove a pci device completely from the system, first detach the port from testpmd. 1284Then the device should be moved under kernel management. 1285Finally the device can be removed using kernel pci hotplug functionality. 1286 1287For example, to move a pci device under kernel management: 1288 1289.. code-block:: console 1290 1291 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0 1292 1293 ./usertools/dpdk-devbind.py --status 1294 1295 Network devices using DPDK-compatible driver 1296 ============================================ 1297 <none> 1298 1299 Network devices using kernel driver 1300 =================================== 1301 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio 1302 1303To remove a port created by a virtual device, above steps are not needed. 1304 1305port start 1306~~~~~~~~~~ 1307 1308Start all ports or a specific port:: 1309 1310 testpmd> port start (port_id|all) 1311 1312port stop 1313~~~~~~~~~ 1314 1315Stop all ports or a specific port:: 1316 1317 testpmd> port stop (port_id|all) 1318 1319port close 1320~~~~~~~~~~ 1321 1322Close all ports or a specific port:: 1323 1324 testpmd> port close (port_id|all) 1325 1326port start/stop queue 1327~~~~~~~~~~~~~~~~~~~~~ 1328 1329Start/stop a rx/tx queue on a specific port:: 1330 1331 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop) 1332 1333Only take effect when port is started. 1334 1335port config - speed 1336~~~~~~~~~~~~~~~~~~~ 1337 1338Set the speed and duplex mode for all ports or a specific port:: 1339 1340 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \ 1341 duplex (half|full|auto) 1342 1343port config - queues/descriptors 1344~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1345 1346Set number of queues/descriptors for rxq, txq, rxd and txd:: 1347 1348 testpmd> port config all (rxq|txq|rxd|txd) (value) 1349 1350This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options. 1351 1352port config - max-pkt-len 1353~~~~~~~~~~~~~~~~~~~~~~~~~ 1354 1355Set the maximum packet length:: 1356 1357 testpmd> port config all max-pkt-len (value) 1358 1359This is equivalent to the ``--max-pkt-len`` command-line option. 1360 1361port config - CRC Strip 1362~~~~~~~~~~~~~~~~~~~~~~~ 1363 1364Set hardware CRC stripping on or off for all ports:: 1365 1366 testpmd> port config all crc-strip (on|off) 1367 1368CRC stripping is off by default. 1369 1370The ``on`` option is equivalent to the ``--crc-strip`` command-line option. 1371 1372port config - scatter 1373~~~~~~~~~~~~~~~~~~~~~~~ 1374 1375Set RX scatter mode on or off for all ports:: 1376 1377 testpmd> port config all scatter (on|off) 1378 1379RX scatter mode is off by default. 1380 1381The ``on`` option is equivalent to the ``--enable-scatter`` command-line option. 1382 1383port config - TX queue flags 1384~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1385 1386Set a hexadecimal bitmap of TX queue flags for all ports:: 1387 1388 testpmd> port config all txqflags value 1389 1390This command is equivalent to the ``--txqflags`` command-line option. 1391 1392port config - RX Checksum 1393~~~~~~~~~~~~~~~~~~~~~~~~~ 1394 1395Set hardware RX checksum offload to on or off for all ports:: 1396 1397 testpmd> port config all rx-cksum (on|off) 1398 1399Checksum offload is off by default. 1400 1401The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option. 1402 1403port config - VLAN 1404~~~~~~~~~~~~~~~~~~ 1405 1406Set hardware VLAN on or off for all ports:: 1407 1408 testpmd> port config all hw-vlan (on|off) 1409 1410Hardware VLAN is on by default. 1411 1412The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option. 1413 1414port config - VLAN filter 1415~~~~~~~~~~~~~~~~~~~~~~~~~ 1416 1417Set hardware VLAN filter on or off for all ports:: 1418 1419 testpmd> port config all hw-vlan-filter (on|off) 1420 1421Hardware VLAN filter is on by default. 1422 1423The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option. 1424 1425port config - VLAN strip 1426~~~~~~~~~~~~~~~~~~~~~~~~ 1427 1428Set hardware VLAN strip on or off for all ports:: 1429 1430 testpmd> port config all hw-vlan-strip (on|off) 1431 1432Hardware VLAN strip is on by default. 1433 1434The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option. 1435 1436port config - VLAN extend 1437~~~~~~~~~~~~~~~~~~~~~~~~~ 1438 1439Set hardware VLAN extend on or off for all ports:: 1440 1441 testpmd> port config all hw-vlan-extend (on|off) 1442 1443Hardware VLAN extend is off by default. 1444 1445The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option. 1446 1447port config - Drop Packets 1448~~~~~~~~~~~~~~~~~~~~~~~~~~ 1449 1450Set packet drop for packets with no descriptors on or off for all ports:: 1451 1452 testpmd> port config all drop-en (on|off) 1453 1454Packet dropping for packets with no descriptors is off by default. 1455 1456The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option. 1457 1458port config - RSS 1459~~~~~~~~~~~~~~~~~ 1460 1461Set the RSS (Receive Side Scaling) mode on or off:: 1462 1463 testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none) 1464 1465RSS is on by default. 1466 1467The ``none`` option is equivalent to the ``--disable-rss`` command-line option. 1468 1469port config - RSS Reta 1470~~~~~~~~~~~~~~~~~~~~~~ 1471 1472Set the RSS (Receive Side Scaling) redirection table:: 1473 1474 testpmd> port config all rss reta (hash,queue)[,(hash,queue)] 1475 1476port config - DCB 1477~~~~~~~~~~~~~~~~~ 1478 1479Set the DCB mode for an individual port:: 1480 1481 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off) 1482 1483The traffic class should be 4 or 8. 1484 1485port config - Burst 1486~~~~~~~~~~~~~~~~~~~ 1487 1488Set the number of packets per burst:: 1489 1490 testpmd> port config all burst (value) 1491 1492This is equivalent to the ``--burst`` command-line option. 1493 1494port config - Threshold 1495~~~~~~~~~~~~~~~~~~~~~~~ 1496 1497Set thresholds for TX/RX queues:: 1498 1499 testpmd> port config all (threshold) (value) 1500 1501Where the threshold type can be: 1502 1503* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255. 1504 1505* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255. 1506 1507* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255. 1508 1509* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255. 1510 1511* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255. 1512 1513* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255. 1514 1515* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd. 1516 1517* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd. 1518 1519* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd. 1520 1521These threshold options are also available from the command-line. 1522 1523port config - E-tag 1524~~~~~~~~~~~~~~~~~~~ 1525 1526Set the value of ether-type for E-tag:: 1527 1528 testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value) 1529 1530Enable/disable the E-tag support:: 1531 1532 testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable) 1533 1534 1535Link Bonding Functions 1536---------------------- 1537 1538The Link Bonding functions make it possible to dynamically create and 1539manage link bonding devices from within testpmd interactive prompt. 1540 1541create bonded device 1542~~~~~~~~~~~~~~~~~~~~ 1543 1544Create a new bonding device:: 1545 1546 testpmd> create bonded device (mode) (socket) 1547 1548For example, to create a bonded device in mode 1 on socket 0:: 1549 1550 testpmd> create bonded 1 0 1551 created new bonded device (port X) 1552 1553add bonding slave 1554~~~~~~~~~~~~~~~~~ 1555 1556Adds Ethernet device to a Link Bonding device:: 1557 1558 testpmd> add bonding slave (slave id) (port id) 1559 1560For example, to add Ethernet device (port 6) to a Link Bonding device (port 10):: 1561 1562 testpmd> add bonding slave 6 10 1563 1564 1565remove bonding slave 1566~~~~~~~~~~~~~~~~~~~~ 1567 1568Removes an Ethernet slave device from a Link Bonding device:: 1569 1570 testpmd> remove bonding slave (slave id) (port id) 1571 1572For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10):: 1573 1574 testpmd> remove bonding slave 6 10 1575 1576set bonding mode 1577~~~~~~~~~~~~~~~~ 1578 1579Set the Link Bonding mode of a Link Bonding device:: 1580 1581 testpmd> set bonding mode (value) (port id) 1582 1583For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3):: 1584 1585 testpmd> set bonding mode 3 10 1586 1587set bonding primary 1588~~~~~~~~~~~~~~~~~~~ 1589 1590Set an Ethernet slave device as the primary device on a Link Bonding device:: 1591 1592 testpmd> set bonding primary (slave id) (port id) 1593 1594For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10):: 1595 1596 testpmd> set bonding primary 6 10 1597 1598set bonding mac 1599~~~~~~~~~~~~~~~ 1600 1601Set the MAC address of a Link Bonding device:: 1602 1603 testpmd> set bonding mac (port id) (mac) 1604 1605For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01:: 1606 1607 testpmd> set bonding mac 10 00:00:00:00:00:01 1608 1609set bonding xmit_balance_policy 1610~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1611 1612Set the transmission policy for a Link Bonding device when it is in Balance XOR mode:: 1613 1614 testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34) 1615 1616For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports):: 1617 1618 testpmd> set bonding xmit_balance_policy 10 l34 1619 1620 1621set bonding mon_period 1622~~~~~~~~~~~~~~~~~~~~~~ 1623 1624Set the link status monitoring polling period in milliseconds for a bonding device. 1625 1626This adds support for PMD slave devices which do not support link status interrupts. 1627When the mon_period is set to a value greater than 0 then all PMD's which do not support 1628link status ISR will be queried every polling interval to check if their link status has changed:: 1629 1630 testpmd> set bonding mon_period (port_id) (value) 1631 1632For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms:: 1633 1634 testpmd> set bonding mon_period 5 150 1635 1636 1637show bonding config 1638~~~~~~~~~~~~~~~~~~~ 1639 1640Show the current configuration of a Link Bonding device:: 1641 1642 testpmd> show bonding config (port id) 1643 1644For example, 1645to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4) 1646in balance mode with a transmission policy of layer 2+3:: 1647 1648 testpmd> show bonding config 9 1649 Bonding mode: 2 1650 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23 1651 Slaves (3): [1 3 4] 1652 Active Slaves (3): [1 3 4] 1653 Primary: [3] 1654 1655 1656Register Functions 1657------------------ 1658 1659The Register Functions can be used to read from and write to registers on the network card referenced by a port number. 1660This is mainly useful for debugging purposes. 1661Reference should be made to the appropriate datasheet for the network card for details on the register addresses 1662and fields that can be accessed. 1663 1664read reg 1665~~~~~~~~ 1666 1667Display the value of a port register:: 1668 1669 testpmd> read reg (port_id) (address) 1670 1671For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller:: 1672 1673 testpmd> read reg 0 0xEE00 1674 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241) 1675 1676read regfield 1677~~~~~~~~~~~~~ 1678 1679Display a port register bit field:: 1680 1681 testpmd> read regfield (port_id) (address) (bit_x) (bit_y) 1682 1683For example, reading the lowest two bits from the register in the example above:: 1684 1685 testpmd> read regfield 0 0xEE00 0 1 1686 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1) 1687 1688read regbit 1689~~~~~~~~~~~ 1690 1691Display a single port register bit:: 1692 1693 testpmd> read regbit (port_id) (address) (bit_x) 1694 1695For example, reading the lowest bit from the register in the example above:: 1696 1697 testpmd> read regbit 0 0xEE00 0 1698 port 0 PCI register at offset 0xEE00: bit 0=1 1699 1700write reg 1701~~~~~~~~~ 1702 1703Set the value of a port register:: 1704 1705 testpmd> write reg (port_id) (address) (value) 1706 1707For example, to clear a register:: 1708 1709 testpmd> write reg 0 0xEE00 0x0 1710 port 0 PCI register at offset 0xEE00: 0x00000000 (0) 1711 1712write regfield 1713~~~~~~~~~~~~~~ 1714 1715Set bit field of a port register:: 1716 1717 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value) 1718 1719For example, writing to the register cleared in the example above:: 1720 1721 testpmd> write regfield 0 0xEE00 0 1 2 1722 port 0 PCI register at offset 0xEE00: 0x00000002 (2) 1723 1724write regbit 1725~~~~~~~~~~~~ 1726 1727Set single bit value of a port register:: 1728 1729 testpmd> write regbit (port_id) (address) (bit_x) (value) 1730 1731For example, to set the high bit in the register from the example above:: 1732 1733 testpmd> write regbit 0 0xEE00 31 1 1734 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658) 1735 1736 1737Filter Functions 1738---------------- 1739 1740This section details the available filter functions that are available. 1741 1742Note these functions interface the deprecated legacy filtering framework, 1743superseded by *rte_flow*. See `Flow rules management`_. 1744 1745ethertype_filter 1746~~~~~~~~~~~~~~~~~~~~ 1747 1748Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue:: 1749 1750 ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \ 1751 ethertype (ether_type) (drop|fwd) queue (queue_id) 1752 1753The available information parameters are: 1754 1755* ``port_id``: The port which the Ethertype filter assigned on. 1756 1757* ``mac_addr``: Compare destination mac address. 1758 1759* ``mac_ignr``: Ignore destination mac address match. 1760 1761* ``mac_address``: Destination mac address to match. 1762 1763* ``ether_type``: The EtherType value want to match, 1764 for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid. 1765 1766* ``queue_id``: The receive queue associated with this EtherType filter. 1767 It is meaningless when deleting or dropping. 1768 1769Example, to add/remove an ethertype filter rule:: 1770 1771 testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \ 1772 ethertype 0x0806 fwd queue 3 1773 1774 testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \ 1775 ethertype 0x0806 fwd queue 3 1776 17772tuple_filter 1778~~~~~~~~~~~~~~~~~ 1779 1780Add or delete a 2-tuple filter, 1781which identifies packets by specific protocol and destination TCP/UDP port 1782and forwards packets into one of the receive queues:: 1783 1784 2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \ 1785 protocol (protocol_value) mask (mask_value) \ 1786 tcp_flags (tcp_flags_value) priority (prio_value) \ 1787 queue (queue_id) 1788 1789The available information parameters are: 1790 1791* ``port_id``: The port which the 2-tuple filter assigned on. 1792 1793* ``dst_port_value``: Destination port in L4. 1794 1795* ``protocol_value``: IP L4 protocol. 1796 1797* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate. 1798 1799* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP). 1800 1801* ``prio_value``: Priority of this filter. 1802 1803* ``queue_id``: The receive queue associated with this 2-tuple filter. 1804 1805Example, to add/remove an 2tuple filter rule:: 1806 1807 testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \ 1808 tcp_flags 0x02 priority 3 queue 3 1809 1810 testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \ 1811 tcp_flags 0x02 priority 3 queue 3 1812 18135tuple_filter 1814~~~~~~~~~~~~~~~~~ 1815 1816Add or delete a 5-tuple filter, 1817which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port) 1818and routes packets into one of the receive queues:: 1819 1820 5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \ 1821 (src_address) dst_port (dst_port_value) \ 1822 src_port (src_port_value) protocol (protocol_value) \ 1823 mask (mask_value) tcp_flags (tcp_flags_value) \ 1824 priority (prio_value) queue (queue_id) 1825 1826The available information parameters are: 1827 1828* ``port_id``: The port which the 5-tuple filter assigned on. 1829 1830* ``dst_address``: Destination IP address. 1831 1832* ``src_address``: Source IP address. 1833 1834* ``dst_port_value``: TCP/UDP destination port. 1835 1836* ``src_port_value``: TCP/UDP source port. 1837 1838* ``protocol_value``: L4 protocol. 1839 1840* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate 1841 1842* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP). 1843 1844* ``prio_value``: The priority of this filter. 1845 1846* ``queue_id``: The receive queue associated with this 5-tuple filter. 1847 1848Example, to add/remove an 5tuple filter rule:: 1849 1850 testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \ 1851 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \ 1852 flags 0x0 priority 3 queue 3 1853 1854 testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \ 1855 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \ 1856 flags 0x0 priority 3 queue 3 1857 1858syn_filter 1859~~~~~~~~~~ 1860 1861Using the SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue:: 1862 1863 syn_filter (port_id) (add|del) priority (high|low) queue (queue_id) 1864 1865The available information parameters are: 1866 1867* ``port_id``: The port which the SYN filter assigned on. 1868 1869* ``high``: This SYN filter has higher priority than other filters. 1870 1871* ``low``: This SYN filter has lower priority than other filters. 1872 1873* ``queue_id``: The receive queue associated with this SYN filter 1874 1875Example:: 1876 1877 testpmd> syn_filter 0 add priority high queue 3 1878 1879flex_filter 1880~~~~~~~~~~~ 1881 1882With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet 1883and routed into one of the receive queues:: 1884 1885 flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \ 1886 mask (mask_value) priority (prio_value) queue (queue_id) 1887 1888The available information parameters are: 1889 1890* ``port_id``: The port which the Flex filter is assigned on. 1891 1892* ``len_value``: Filter length in bytes, no greater than 128. 1893 1894* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match. 1895 1896* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match. 1897 1898* ``prio_value``: The priority of this filter. 1899 1900* ``queue_id``: The receive queue associated with this Flex filter. 1901 1902Example:: 1903 1904 testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \ 1905 mask 000C priority 3 queue 3 1906 1907 testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \ 1908 mask 000C priority 3 queue 3 1909 1910 1911.. _testpmd_flow_director: 1912 1913flow_director_filter 1914~~~~~~~~~~~~~~~~~~~~ 1915 1916The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues. 1917 1918Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and 1919Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter: 1920 1921* Perfect match filters. 1922 The hardware checks a match between the masked fields of the received packets and the programmed filters. 1923 The masked fields are for IP flow. 1924 1925* Signature filters. 1926 The hardware checks a match between a hash-based signature of the masked fields of the received packet. 1927 1928* Perfect-mac-vlan match filters. 1929 The hardware checks a match between the masked fields of the received packets and the programmed filters. 1930 The masked fields are for MAC VLAN flow. 1931 1932* Perfect-tunnel match filters. 1933 The hardware checks a match between the masked fields of the received packets and the programmed filters. 1934 The masked fields are for tunnel flow. 1935 1936The Flow Director filters can match the different fields for different type of packet: flow type, specific input set 1937per flow type and the flexible payload. 1938 1939The Flow Director can also mask out parts of all of these fields so that filters 1940are only applied to certain fields or parts of the fields. 1941 1942Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information. 1943 1944# Commands to add flow director filters of different flow types:: 1945 1946 flow_director_filter (port_id) mode IP (add|del|update) \ 1947 flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \ 1948 src (src_ip_address) dst (dst_ip_address) \ 1949 tos (tos_value) proto (proto_value) ttl (ttl_value) \ 1950 vlan (vlan_value) flexbytes (flexbytes_value) \ 1951 (drop|fwd) pf|vf(vf_id) queue (queue_id) \ 1952 fd_id (fd_id_value) 1953 1954 flow_director_filter (port_id) mode IP (add|del|update) \ 1955 flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \ 1956 src (src_ip_address) (src_port) \ 1957 dst (dst_ip_address) (dst_port) \ 1958 tos (tos_value) ttl (ttl_value) \ 1959 vlan (vlan_value) flexbytes (flexbytes_value) \ 1960 (drop|fwd) queue pf|vf(vf_id) (queue_id) \ 1961 fd_id (fd_id_value) 1962 1963 flow_director_filter (port_id) mode IP (add|del|update) \ 1964 flow (ipv4-sctp|ipv6-sctp) \ 1965 src (src_ip_address) (src_port) \ 1966 dst (dst_ip_address) (dst_port) \ 1967 tos (tos_value) ttl (ttl_value) \ 1968 tag (verification_tag) vlan (vlan_value) \ 1969 flexbytes (flexbytes_value) (drop|fwd) \ 1970 pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value) 1971 1972 flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \ 1973 ether (ethertype) flexbytes (flexbytes_value) \ 1974 (drop|fwd) pf|vf(vf_id) queue (queue_id) 1975 fd_id (fd_id_value) 1976 1977 flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \ 1978 mac (mac_address) vlan (vlan_value) \ 1979 flexbytes (flexbytes_value) (drop|fwd) \ 1980 queue (queue_id) fd_id (fd_id_value) 1981 1982 flow_director_filter (port_id) mode Tunnel (add|del|update) \ 1983 mac (mac_address) vlan (vlan_value) \ 1984 tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \ 1985 flexbytes (flexbytes_value) (drop|fwd) \ 1986 queue (queue_id) fd_id (fd_id_value) 1987 1988For example, to add an ipv4-udp flow type filter:: 1989 1990 testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \ 1991 dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \ 1992 fwd pf queue 1 fd_id 1 1993 1994For example, add an ipv4-other flow type filter:: 1995 1996 testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \ 1997 dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \ 1998 flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1 1999 2000flush_flow_director 2001~~~~~~~~~~~~~~~~~~~ 2002 2003Flush all flow director filters on a device:: 2004 2005 testpmd> flush_flow_director (port_id) 2006 2007Example, to flush all flow director filter on port 0:: 2008 2009 testpmd> flush_flow_director 0 2010 2011flow_director_mask 2012~~~~~~~~~~~~~~~~~~ 2013 2014Set flow director's input masks:: 2015 2016 flow_director_mask (port_id) mode IP vlan (vlan_value) \ 2017 src_mask (ipv4_src) (ipv6_src) (src_port) \ 2018 dst_mask (ipv4_dst) (ipv6_dst) (dst_port) 2019 2020 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value) 2021 2022 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \ 2023 mac (mac_value) tunnel-type (tunnel_type_value) \ 2024 tunnel-id (tunnel_id_value) 2025 2026Example, to set flow director mask on port 0:: 2027 2028 testpmd> flow_director_mask 0 mode IP vlan 0xefff \ 2029 src_mask 255.255.255.255 \ 2030 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \ 2031 dst_mask 255.255.255.255 \ 2032 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF 2033 2034flow_director_flex_mask 2035~~~~~~~~~~~~~~~~~~~~~~~ 2036 2037set masks of flow director's flexible payload based on certain flow type:: 2038 2039 testpmd> flow_director_flex_mask (port_id) \ 2040 flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2041 ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \ 2042 l2_payload|all) (mask) 2043 2044Example, to set flow director's flex mask for all flow type on port 0:: 2045 2046 testpmd> flow_director_flex_mask 0 flow all \ 2047 (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0) 2048 2049 2050flow_director_flex_payload 2051~~~~~~~~~~~~~~~~~~~~~~~~~~ 2052 2053Configure flexible payload selection:: 2054 2055 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config) 2056 2057For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload:: 2058 2059 testpmd> flow_director_flex_payload 0 l4 \ 2060 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19) 2061 2062get_sym_hash_ena_per_port 2063~~~~~~~~~~~~~~~~~~~~~~~~~ 2064 2065Get symmetric hash enable configuration per port:: 2066 2067 get_sym_hash_ena_per_port (port_id) 2068 2069For example, to get symmetric hash enable configuration of port 1:: 2070 2071 testpmd> get_sym_hash_ena_per_port 1 2072 2073set_sym_hash_ena_per_port 2074~~~~~~~~~~~~~~~~~~~~~~~~~ 2075 2076Set symmetric hash enable configuration per port to enable or disable:: 2077 2078 set_sym_hash_ena_per_port (port_id) (enable|disable) 2079 2080For example, to set symmetric hash enable configuration of port 1 to enable:: 2081 2082 testpmd> set_sym_hash_ena_per_port 1 enable 2083 2084get_hash_global_config 2085~~~~~~~~~~~~~~~~~~~~~~ 2086 2087Get the global configurations of hash filters:: 2088 2089 get_hash_global_config (port_id) 2090 2091For example, to get the global configurations of hash filters of port 1:: 2092 2093 testpmd> get_hash_global_config 1 2094 2095set_hash_global_config 2096~~~~~~~~~~~~~~~~~~~~~~ 2097 2098Set the global configurations of hash filters:: 2099 2100 set_hash_global_config (port_id) (toeplitz|simple_xor|default) \ 2101 (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \ 2102 ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \ 2103 (enable|disable) 2104 2105For example, to enable simple_xor for flow type of ipv6 on port 2:: 2106 2107 testpmd> set_hash_global_config 2 simple_xor ipv6 enable 2108 2109set_hash_input_set 2110~~~~~~~~~~~~~~~~~~ 2111 2112Set the input set for hash:: 2113 2114 set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2115 ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \ 2116 l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \ 2117 ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \ 2118 tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \ 2119 udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \ 2120 fld-8th|none) (select|add) 2121 2122For example, to add source IP to hash input set for flow type of ipv4-udp on port 0:: 2123 2124 testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add 2125 2126set_fdir_input_set 2127~~~~~~~~~~~~~~~~~~ 2128 2129The Flow Director filters can match the different fields for different type of packet, i.e. specific input set 2130on per flow type and the flexible payload. This command can be used to change input set for each flow type. 2131 2132Set the input set for flow director:: 2133 2134 set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2135 ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \ 2136 l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \ 2137 ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \ 2138 tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \ 2139 sctp-dst-port|sctp-veri-tag|none) (select|add) 2140 2141For example to add source IP to FD input set for flow type of ipv4-udp on port 0:: 2142 2143 testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add 2144 2145global_config 2146~~~~~~~~~~~~~ 2147 2148Set different GRE key length for input set:: 2149 2150 global_config (port_id) gre-key-len (number in bytes) 2151 2152For example to set GRE key length for input set to 4 bytes on port 0:: 2153 2154 testpmd> global_config 0 gre-key-len 4 2155 2156 2157.. _testpmd_rte_flow: 2158 2159Flow rules management 2160--------------------- 2161 2162Control of the generic flow API (*rte_flow*) is fully exposed through the 2163``flow`` command (validation, creation, destruction and queries). 2164 2165Considering *rte_flow* overlaps with all `Filter Functions`_, using both 2166features simultaneously may cause undefined side-effects and is therefore 2167not recommended. 2168 2169``flow`` syntax 2170~~~~~~~~~~~~~~~ 2171 2172Because the ``flow`` command uses dynamic tokens to handle the large number 2173of possible flow rules combinations, its behavior differs slightly from 2174other commands, in particular: 2175 2176- Pressing *?* or the *<tab>* key displays contextual help for the current 2177 token, not that of the entire command. 2178 2179- Optional and repeated parameters are supported (provided they are listed 2180 in the contextual help). 2181 2182The first parameter stands for the operation mode. Possible operations and 2183their general syntax are described below. They are covered in detail in the 2184following sections. 2185 2186- Check whether a flow rule can be created:: 2187 2188 flow validate {port_id} 2189 [group {group_id}] [priority {level}] [ingress] [egress] 2190 pattern {item} [/ {item} [...]] / end 2191 actions {action} [/ {action} [...]] / end 2192 2193- Create a flow rule:: 2194 2195 flow create {port_id} 2196 [group {group_id}] [priority {level}] [ingress] [egress] 2197 pattern {item} [/ {item} [...]] / end 2198 actions {action} [/ {action} [...]] / end 2199 2200- Destroy specific flow rules:: 2201 2202 flow destroy {port_id} rule {rule_id} [...] 2203 2204- Destroy all flow rules:: 2205 2206 flow flush {port_id} 2207 2208- Query an existing flow rule:: 2209 2210 flow query {port_id} {rule_id} {action} 2211 2212- List existing flow rules sorted by priority, filtered by group 2213 identifiers:: 2214 2215 flow list {port_id} [group {group_id}] [...] 2216 2217Validating flow rules 2218~~~~~~~~~~~~~~~~~~~~~ 2219 2220``flow validate`` reports whether a flow rule would be accepted by the 2221underlying device in its current state but stops short of creating it. It is 2222bound to ``rte_flow_validate()``:: 2223 2224 flow validate {port_id} 2225 [group {group_id}] [priority {level}] [ingress] [egress] 2226 pattern {item} [/ {item} [...]] / end 2227 actions {action} [/ {action} [...]] / end 2228 2229If successful, it will show:: 2230 2231 Flow rule validated 2232 2233Otherwise it will show an error message of the form:: 2234 2235 Caught error type [...] ([...]): [...] 2236 2237This command uses the same parameters as ``flow create``, their format is 2238described in `Creating flow rules`_. 2239 2240Check whether redirecting any Ethernet packet received on port 0 to RX queue 2241index 6 is supported:: 2242 2243 testpmd> flow validate 0 ingress pattern eth / end 2244 actions queue index 6 / end 2245 Flow rule validated 2246 testpmd> 2247 2248Port 0 does not support TCPv6 rules:: 2249 2250 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end 2251 actions drop / end 2252 Caught error type 9 (specific pattern item): Invalid argument 2253 testpmd> 2254 2255Creating flow rules 2256~~~~~~~~~~~~~~~~~~~ 2257 2258``flow create`` validates and creates the specified flow rule. It is bound 2259to ``rte_flow_create()``:: 2260 2261 flow create {port_id} 2262 [group {group_id}] [priority {level}] [ingress] [egress] 2263 pattern {item} [/ {item} [...]] / end 2264 actions {action} [/ {action} [...]] / end 2265 2266If successful, it will return a flow rule ID usable with other commands:: 2267 2268 Flow rule #[...] created 2269 2270Otherwise it will show an error message of the form:: 2271 2272 Caught error type [...] ([...]): [...] 2273 2274Parameters describe in the following order: 2275 2276- Attributes (*group*, *priority*, *ingress*, *egress* tokens). 2277- A matching pattern, starting with the *pattern* token and terminated by an 2278 *end* pattern item. 2279- Actions, starting with the *actions* token and terminated by an *end* 2280 action. 2281 2282These translate directly to *rte_flow* objects provided as-is to the 2283underlying functions. 2284 2285The shortest valid definition only comprises mandatory tokens:: 2286 2287 testpmd> flow create 0 pattern end actions end 2288 2289Note that PMDs may refuse rules that essentially do nothing such as this 2290one. 2291 2292**All unspecified object values are automatically initialized to 0.** 2293 2294Attributes 2295^^^^^^^^^^ 2296 2297These tokens affect flow rule attributes (``struct rte_flow_attr``) and are 2298specified before the ``pattern`` token. 2299 2300- ``group {group id}``: priority group. 2301- ``priority {level}``: priority level within group. 2302- ``ingress``: rule applies to ingress traffic. 2303- ``egress``: rule applies to egress traffic. 2304 2305Each instance of an attribute specified several times overrides the previous 2306value as shown below (group 4 is used):: 2307 2308 testpmd> flow create 0 group 42 group 24 group 4 [...] 2309 2310Note that once enabled, ``ingress`` and ``egress`` cannot be disabled. 2311 2312While not specifying a direction is an error, some rules may allow both 2313simultaneously. 2314 2315Most rules affect RX therefore contain the ``ingress`` token:: 2316 2317 testpmd> flow create 0 ingress pattern [...] 2318 2319Matching pattern 2320^^^^^^^^^^^^^^^^ 2321 2322A matching pattern starts after the ``pattern`` token. It is made of pattern 2323items and is terminated by a mandatory ``end`` item. 2324 2325Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum 2326rte_flow_item_type``). 2327 2328The ``/`` token is used as a separator between pattern items as shown 2329below:: 2330 2331 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...] 2332 2333Note that protocol items like these must be stacked from lowest to highest 2334layer to make sense. For instance, the following rule is either invalid or 2335unlikely to match any packet:: 2336 2337 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...] 2338 2339More information on these restrictions can be found in the *rte_flow* 2340documentation. 2341 2342Several items support additional specification structures, for example 2343``ipv4`` allows specifying source and destination addresses as follows:: 2344 2345 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1 2346 dst is 10.2.0.0 / end [...] 2347 2348This rule matches all IPv4 traffic with the specified properties. 2349 2350In this example, ``src`` and ``dst`` are field names of the underlying 2351``struct rte_flow_item_ipv4`` object. All item properties can be specified 2352in a similar fashion. 2353 2354The ``is`` token means that the subsequent value must be matched exactly, 2355and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item`` 2356accordingly. Possible assignment tokens are: 2357 2358- ``is``: match value perfectly (with full bit-mask). 2359- ``spec``: match value according to configured bit-mask. 2360- ``last``: specify upper bound to establish a range. 2361- ``mask``: specify bit-mask with relevant bits set to one. 2362- ``prefix``: generate bit-mask from a prefix length. 2363 2364These yield identical results:: 2365 2366 ipv4 src is 10.1.1.1 2367 2368:: 2369 2370 ipv4 src spec 10.1.1.1 src mask 255.255.255.255 2371 2372:: 2373 2374 ipv4 src spec 10.1.1.1 src prefix 32 2375 2376:: 2377 2378 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value 2379 2380:: 2381 2382 ipv4 src is 10.1.1.1 src last 0 # 0 disables range 2383 2384Inclusive ranges can be defined with ``last``:: 2385 2386 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4 2387 2388Note that ``mask`` affects both ``spec`` and ``last``:: 2389 2390 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0 2391 # matches 10.1.0.0 to 10.2.255.255 2392 2393Properties can be modified multiple times:: 2394 2395 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4 2396 2397:: 2398 2399 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16 2400 2401Pattern items 2402^^^^^^^^^^^^^ 2403 2404This section lists supported pattern items and their attributes, if any. 2405 2406- ``end``: end list of pattern items. 2407 2408- ``void``: no-op pattern item. 2409 2410- ``invert``: perform actions when pattern does not match. 2411 2412- ``any``: match any protocol for the current layer. 2413 2414 - ``num {unsigned}``: number of layers covered. 2415 2416- ``pf``: match packets addressed to the physical function. 2417 2418- ``vf``: match packets addressed to a virtual function ID. 2419 2420 - ``id {unsigned}``: destination VF ID. 2421 2422- ``port``: device-specific physical port index to use. 2423 2424 - ``index {unsigned}``: physical port index. 2425 2426- ``raw``: match an arbitrary byte string. 2427 2428 - ``relative {boolean}``: look for pattern after the previous item. 2429 - ``search {boolean}``: search pattern from offset (see also limit). 2430 - ``offset {integer}``: absolute or relative offset for pattern. 2431 - ``limit {unsigned}``: search area limit for start of pattern. 2432 - ``pattern {string}``: byte string to look for. 2433 2434- ``eth``: match Ethernet header. 2435 2436 - ``dst {MAC-48}``: destination MAC. 2437 - ``src {MAC-48}``: source MAC. 2438 - ``type {unsigned}``: EtherType. 2439 2440- ``vlan``: match 802.1Q/ad VLAN tag. 2441 2442 - ``tpid {unsigned}``: tag protocol identifier. 2443 - ``tci {unsigned}``: tag control information. 2444 - ``pcp {unsigned}``: priority code point. 2445 - ``dei {unsigned}``: drop eligible indicator. 2446 - ``vid {unsigned}``: VLAN identifier. 2447 2448- ``ipv4``: match IPv4 header. 2449 2450 - ``tos {unsigned}``: type of service. 2451 - ``ttl {unsigned}``: time to live. 2452 - ``proto {unsigned}``: next protocol ID. 2453 - ``src {ipv4 address}``: source address. 2454 - ``dst {ipv4 address}``: destination address. 2455 2456- ``ipv6``: match IPv6 header. 2457 2458 - ``tc {unsigned}``: traffic class. 2459 - ``flow {unsigned}``: flow label. 2460 - ``proto {unsigned}``: protocol (next header). 2461 - ``hop {unsigned}``: hop limit. 2462 - ``src {ipv6 address}``: source address. 2463 - ``dst {ipv6 address}``: destination address. 2464 2465- ``icmp``: match ICMP header. 2466 2467 - ``type {unsigned}``: ICMP packet type. 2468 - ``code {unsigned}``: ICMP packet code. 2469 2470- ``udp``: match UDP header. 2471 2472 - ``src {unsigned}``: UDP source port. 2473 - ``dst {unsigned}``: UDP destination port. 2474 2475- ``tcp``: match TCP header. 2476 2477 - ``src {unsigned}``: TCP source port. 2478 - ``dst {unsigned}``: TCP destination port. 2479 2480- ``sctp``: match SCTP header. 2481 2482 - ``src {unsigned}``: SCTP source port. 2483 - ``dst {unsigned}``: SCTP destination port. 2484 - ``tag {unsigned}``: validation tag. 2485 - ``cksum {unsigned}``: checksum. 2486 2487- ``vxlan``: match VXLAN header. 2488 2489 - ``vni {unsigned}``: VXLAN identifier. 2490 2491- ``mpls``: match MPLS header. 2492 2493 - ``label {unsigned}``: MPLS label. 2494 2495- ``gre``: match GRE header. 2496 2497 - ``protocol {unsigned}``: protocol type. 2498 2499Actions list 2500^^^^^^^^^^^^ 2501 2502A list of actions starts after the ``actions`` token in the same fashion as 2503`Matching pattern`_; actions are separated by ``/`` tokens and the list is 2504terminated by a mandatory ``end`` action. 2505 2506Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum 2507rte_flow_action_type``). 2508 2509Dropping all incoming UDPv4 packets can be expressed as follows:: 2510 2511 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2512 actions drop / end 2513 2514Several actions have configurable properties which must be specified when 2515there is no valid default value. For example, ``queue`` requires a target 2516queue index. 2517 2518This rule redirects incoming UDPv4 traffic to queue index 6:: 2519 2520 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2521 actions queue index 6 / end 2522 2523While this one could be rejected by PMDs (unspecified queue index):: 2524 2525 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2526 actions queue / end 2527 2528As defined by *rte_flow*, the list is not ordered, all actions of a given 2529rule are performed simultaneously. These are equivalent:: 2530 2531 queue index 6 / void / mark id 42 / end 2532 2533:: 2534 2535 void / mark id 42 / queue index 6 / end 2536 2537All actions in a list should have different types, otherwise only the last 2538action of a given type is taken into account:: 2539 2540 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6 2541 2542:: 2543 2544 drop / drop / drop / end # drop is performed only once 2545 2546:: 2547 2548 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24 2549 2550Considering they are performed simultaneously, opposite and overlapping 2551actions can sometimes be combined when the end result is unambiguous:: 2552 2553 drop / queue index 6 / end # drop has no effect 2554 2555:: 2556 2557 drop / dup index 6 / end # same as above 2558 2559:: 2560 2561 queue index 6 / rss queues 6 7 8 / end # queue has no effect 2562 2563:: 2564 2565 drop / passthru / end # drop has no effect 2566 2567Note that PMDs may still refuse such combinations. 2568 2569Actions 2570^^^^^^^ 2571 2572This section lists supported actions and their attributes, if any. 2573 2574- ``end``: end list of actions. 2575 2576- ``void``: no-op action. 2577 2578- ``passthru``: let subsequent rule process matched packets. 2579 2580- ``mark``: attach 32 bit value to packets. 2581 2582 - ``id {unsigned}``: 32 bit value to return with packets. 2583 2584- ``flag``: flag packets. 2585 2586- ``queue``: assign packets to a given queue index. 2587 2588 - ``index {unsigned}``: queue index to use. 2589 2590- ``drop``: drop packets (note: passthru has priority). 2591 2592- ``count``: enable counters for this rule. 2593 2594- ``dup``: duplicate packets to a given queue index. 2595 2596 - ``index {unsigned}``: queue index to duplicate packets to. 2597 2598- ``rss``: spread packets among several queues. 2599 2600 - ``queues [{unsigned} [...]] end``: queue indices to use. 2601 2602- ``pf``: redirect packets to physical device function. 2603 2604- ``vf``: redirect packets to virtual device function. 2605 2606 - ``original {boolean}``: use original VF ID if possible. 2607 - ``id {unsigned}``: VF ID to redirect packets to. 2608 2609Destroying flow rules 2610~~~~~~~~~~~~~~~~~~~~~ 2611 2612``flow destroy`` destroys one or more rules from their rule ID (as returned 2613by ``flow create``), this command calls ``rte_flow_destroy()`` as many 2614times as necessary:: 2615 2616 flow destroy {port_id} rule {rule_id} [...] 2617 2618If successful, it will show:: 2619 2620 Flow rule #[...] destroyed 2621 2622It does not report anything for rule IDs that do not exist. The usual error 2623message is shown when a rule cannot be destroyed:: 2624 2625 Caught error type [...] ([...]): [...] 2626 2627``flow flush`` destroys all rules on a device and does not take extra 2628arguments. It is bound to ``rte_flow_flush()``:: 2629 2630 flow flush {port_id} 2631 2632Any errors are reported as above. 2633 2634Creating several rules and destroying them:: 2635 2636 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2637 actions queue index 2 / end 2638 Flow rule #0 created 2639 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2640 actions queue index 3 / end 2641 Flow rule #1 created 2642 testpmd> flow destroy 0 rule 0 rule 1 2643 Flow rule #1 destroyed 2644 Flow rule #0 destroyed 2645 testpmd> 2646 2647The same result can be achieved using ``flow flush``:: 2648 2649 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2650 actions queue index 2 / end 2651 Flow rule #0 created 2652 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2653 actions queue index 3 / end 2654 Flow rule #1 created 2655 testpmd> flow flush 0 2656 testpmd> 2657 2658Non-existent rule IDs are ignored:: 2659 2660 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2661 actions queue index 2 / end 2662 Flow rule #0 created 2663 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2664 actions queue index 3 / end 2665 Flow rule #1 created 2666 testpmd> flow destroy 0 rule 42 rule 10 rule 2 2667 testpmd> 2668 testpmd> flow destroy 0 rule 0 2669 Flow rule #0 destroyed 2670 testpmd> 2671 2672Querying flow rules 2673~~~~~~~~~~~~~~~~~~~ 2674 2675``flow query`` queries a specific action of a flow rule having that 2676ability. Such actions collect information that can be reported using this 2677command. It is bound to ``rte_flow_query()``:: 2678 2679 flow query {port_id} {rule_id} {action} 2680 2681If successful, it will display either the retrieved data for known actions 2682or the following message:: 2683 2684 Cannot display result for action type [...] ([...]) 2685 2686Otherwise, it will complain either that the rule does not exist or that some 2687error occurred:: 2688 2689 Flow rule #[...] not found 2690 2691:: 2692 2693 Caught error type [...] ([...]): [...] 2694 2695Currently only the ``count`` action is supported. This action reports the 2696number of packets that hit the flow rule and the total number of bytes. Its 2697output has the following format:: 2698 2699 count: 2700 hits_set: [...] # whether "hits" contains a valid value 2701 bytes_set: [...] # whether "bytes" contains a valid value 2702 hits: [...] # number of packets 2703 bytes: [...] # number of bytes 2704 2705Querying counters for TCPv6 packets redirected to queue 6:: 2706 2707 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end 2708 actions queue index 6 / count / end 2709 Flow rule #4 created 2710 testpmd> flow query 0 4 count 2711 count: 2712 hits_set: 1 2713 bytes_set: 0 2714 hits: 386446 2715 bytes: 0 2716 testpmd> 2717 2718Listing flow rules 2719~~~~~~~~~~~~~~~~~~ 2720 2721``flow list`` lists existing flow rules sorted by priority and optionally 2722filtered by group identifiers:: 2723 2724 flow list {port_id} [group {group_id}] [...] 2725 2726This command only fails with the following message if the device does not 2727exist:: 2728 2729 Invalid port [...] 2730 2731Output consists of a header line followed by a short description of each 2732flow rule, one per line. There is no output at all when no flow rules are 2733configured on the device:: 2734 2735 ID Group Prio Attr Rule 2736 [...] [...] [...] [...] [...] 2737 2738``Attr`` column flags: 2739 2740- ``i`` for ``ingress``. 2741- ``e`` for ``egress``. 2742 2743Creating several flow rules and listing them:: 2744 2745 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2746 actions queue index 6 / end 2747 Flow rule #0 created 2748 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2749 actions queue index 2 / end 2750 Flow rule #1 created 2751 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end 2752 actions rss queues 6 7 8 end / end 2753 Flow rule #2 created 2754 testpmd> flow list 0 2755 ID Group Prio Attr Rule 2756 0 0 0 i- ETH IPV4 => QUEUE 2757 1 0 0 i- ETH IPV6 => QUEUE 2758 2 0 5 i- ETH IPV4 UDP => RSS 2759 testpmd> 2760 2761Rules are sorted by priority (i.e. group ID first, then priority level):: 2762 2763 testpmd> flow list 1 2764 ID Group Prio Attr Rule 2765 0 0 0 i- ETH => COUNT 2766 6 0 500 i- ETH IPV6 TCP => DROP COUNT 2767 5 0 1000 i- ETH IPV6 ICMP => QUEUE 2768 1 24 0 i- ETH IPV4 UDP => QUEUE 2769 4 24 10 i- ETH IPV4 TCP => DROP 2770 3 24 20 i- ETH IPV4 => DROP 2771 2 24 42 i- ETH IPV4 UDP => QUEUE 2772 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE 2773 testpmd> 2774 2775Output can be limited to specific groups:: 2776 2777 testpmd> flow list 1 group 0 group 63 2778 ID Group Prio Attr Rule 2779 0 0 0 i- ETH => COUNT 2780 6 0 500 i- ETH IPV6 TCP => DROP COUNT 2781 5 0 1000 i- ETH IPV6 ICMP => QUEUE 2782 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE 2783 testpmd> 2784