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 X 55 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc all 56 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc X 57 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc 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) (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 150For example: 151 152.. code-block:: console 153 154 testpmd> show port info 0 155 156 ********************* Infos for port 0 ********************* 157 158 MAC address: XX:XX:XX:XX:XX:XX 159 Connect to socket: 0 160 memory allocation on the socket: 0 161 Link status: up 162 Link speed: 40000 Mbps 163 Link duplex: full-duplex 164 Promiscuous mode: enabled 165 Allmulticast mode: disabled 166 Maximum number of MAC addresses: 64 167 Maximum number of MAC addresses of hash filtering: 0 168 VLAN offload: 169 strip on 170 filter on 171 qinq(extend) off 172 Redirection table size: 512 173 Supported flow types: 174 ipv4-frag 175 ipv4-tcp 176 ipv4-udp 177 ipv4-sctp 178 ipv4-other 179 ipv6-frag 180 ipv6-tcp 181 ipv6-udp 182 ipv6-sctp 183 ipv6-other 184 l2_payload 185 port 186 vxlan 187 geneve 188 nvgre 189 190show port rss reta 191~~~~~~~~~~~~~~~~~~ 192 193Display the rss redirection table entry indicated by masks on port X:: 194 195 testpmd> show port (port_id) rss reta (size) (mask0, mask1...) 196 197size is used to indicate the hardware supported reta size 198 199show port rss-hash 200~~~~~~~~~~~~~~~~~~ 201 202Display the RSS hash functions and RSS hash key of a port:: 203 204 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] 205 206clear port 207~~~~~~~~~~ 208 209Clear the port statistics for a given port or for all ports:: 210 211 testpmd> clear port (info|stats|xstats|fdir|stat_qmap) (port_id|all) 212 213For example:: 214 215 testpmd> clear port stats all 216 217show (rxq|txq) 218~~~~~~~~~~~~~~ 219 220Display information for a given port's RX/TX queue:: 221 222 testpmd> show (rxq|txq) info (port_id) (queue_id) 223 224show config 225~~~~~~~~~~~ 226 227Displays the configuration of the application. 228The configuration comes from the command-line, the runtime or the application defaults:: 229 230 testpmd> show config (rxtx|cores|fwd|txpkts) 231 232The available information categories are: 233 234* ``rxtx``: RX/TX configuration items. 235 236* ``cores``: List of forwarding cores. 237 238* ``fwd``: Packet forwarding configuration. 239 240* ``txpkts``: Packets to TX configuration. 241 242For example: 243 244.. code-block:: console 245 246 testpmd> show config rxtx 247 248 io packet forwarding - CRC stripping disabled - packets/burst=16 249 nb forwarding cores=2 - nb forwarding ports=1 250 RX queues=1 - RX desc=128 - RX free threshold=0 251 RX threshold registers: pthresh=8 hthresh=8 wthresh=4 252 TX queues=1 - TX desc=512 - TX free threshold=0 253 TX threshold registers: pthresh=36 hthresh=0 wthresh=0 254 TX RS bit threshold=0 - TXQ flags=0x0 255 256set fwd 257~~~~~~~ 258 259Set the packet forwarding mode:: 260 261 testpmd> set fwd (io|mac|macswap|flowgen| \ 262 rxonly|txonly|csum|icmpecho) (""|retry) 263 264``retry`` can be specified for forwarding engines except ``rx_only``. 265 266The available information categories are: 267 268* ``io``: Forwards packets "as-is" in I/O mode. 269 This is the fastest possible forwarding operation as it does not access packets data. 270 This is the default mode. 271 272* ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them. 273 Default application behaviour is to set source Ethernet address to that of the transmitting interface, and destination 274 address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or 275 'eth-peer-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address. 276 277* ``macswap``: MAC swap forwarding mode. 278 Swaps the source and the destination Ethernet addresses of packets before forwarding them. 279 280* ``flowgen``: Multi-flow generation mode. 281 Originates a number of flows (with varying destination IP addresses), and terminate receive traffic. 282 283* ``rxonly``: Receives packets but doesn't transmit them. 284 285* ``txonly``: Generates and transmits packets without receiving any. 286 287* ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet. 288 289* ``icmpecho``: Receives a burst of packets, lookup for IMCP echo requests and, if any, send back ICMP echo replies. 290 291* ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX. Requires ``CONFIG_RTE_LIBRTE_IEEE1588=y``. 292 293Note: TX timestamping is only available in the "Full Featured" TX path. To force ``testpmd`` into this mode set ``--txqflags=0``. 294 295Example:: 296 297 testpmd> set fwd rxonly 298 299 Set rxonly packet forwarding mode 300 301 302read rxd 303~~~~~~~~ 304 305Display an RX descriptor for a port RX queue:: 306 307 testpmd> read rxd (port_id) (queue_id) (rxd_id) 308 309For example:: 310 311 testpmd> read rxd 0 0 4 312 0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180 313 314read txd 315~~~~~~~~ 316 317Display a TX descriptor for a port TX queue:: 318 319 testpmd> read txd (port_id) (queue_id) (txd_id) 320 321For example:: 322 323 testpmd> read txd 0 0 4 324 0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C 325 326 327Configuration Functions 328----------------------- 329 330The testpmd application can be configured from the runtime as well as from the command-line. 331 332This section details the available configuration functions that are available. 333 334.. note:: 335 336 Configuration changes only become active when forwarding is started/restarted. 337 338set default 339~~~~~~~~~~~ 340 341Reset forwarding to the default configuration:: 342 343 testpmd> set default 344 345set verbose 346~~~~~~~~~~~ 347 348Set the debug verbosity level:: 349 350 testpmd> set verbose (level) 351 352Currently the only available levels are 0 (silent except for error) and 1 (fully verbose). 353 354set nbport 355~~~~~~~~~~ 356 357Set the number of ports used by the application: 358 359set nbport (num) 360 361This is equivalent to the ``--nb-ports`` command-line option. 362 363set nbcore 364~~~~~~~~~~ 365 366Set the number of cores used by the application:: 367 368 testpmd> set nbcore (num) 369 370This is equivalent to the ``--nb-cores`` command-line option. 371 372.. note:: 373 374 The number of cores used must not be greater than number of ports used multiplied by the number of queues per port. 375 376set coremask 377~~~~~~~~~~~~ 378 379Set the forwarding cores hexadecimal mask:: 380 381 testpmd> set coremask (mask) 382 383This is equivalent to the ``--coremask`` command-line option. 384 385.. note:: 386 387 The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding. 388 389set portmask 390~~~~~~~~~~~~ 391 392Set the forwarding ports hexadecimal mask:: 393 394 testpmd> set portmask (mask) 395 396This is equivalent to the ``--portmask`` command-line option. 397 398set burst 399~~~~~~~~~ 400 401Set number of packets per burst:: 402 403 testpmd> set burst (num) 404 405This is equivalent to the ``--burst command-line`` option. 406 407When retry is enabled, the transmit delay time and number of retries can also be set:: 408 409 testpmd> set burst tx delay (microseconds) retry (num) 410 411set txpkts 412~~~~~~~~~~ 413 414Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode:: 415 416 testpmd> set txpkts (x[,y]*) 417 418Where x[,y]* represents a CSV list of values, without white space. 419 420set txsplit 421~~~~~~~~~~~ 422 423Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes:: 424 425 testpmd> set txsplit (off|on|rand) 426 427Where: 428 429* ``off`` disable packet copy & split for CSUM mode. 430 431* ``on`` split outgoing packet into multiple segments. Size of each segment 432 and number of segments per packet is determined by ``set txpkts`` command 433 (see above). 434 435* ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments. 436 437set corelist 438~~~~~~~~~~~~ 439 440Set the list of forwarding cores:: 441 442 testpmd> set corelist (x[,y]*) 443 444For example, to change the forwarding cores: 445 446.. code-block:: console 447 448 testpmd> set corelist 3,1 449 testpmd> show config fwd 450 451 io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled 452 Logical Core 3 (socket 0) forwards packets on 1 streams: 453 RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01 454 Logical Core 1 (socket 0) forwards packets on 1 streams: 455 RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00 456 457.. note:: 458 459 The cores are used in the same order as specified on the command line. 460 461set portlist 462~~~~~~~~~~~~ 463 464Set the list of forwarding ports:: 465 466 testpmd> set portlist (x[,y]*) 467 468For example, to change the port forwarding: 469 470.. code-block:: console 471 472 testpmd> set portlist 0,2,1,3 473 testpmd> show config fwd 474 475 io packet forwarding - ports=4 - cores=1 - streams=4 476 Logical Core 3 (socket 0) forwards packets on 4 streams: 477 RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01 478 RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00 479 RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03 480 RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02 481 482set tx loopback 483~~~~~~~~~~~~~~~ 484 485Enable/disable tx loopback:: 486 487 testpmd> set tx loopback (port_id) (on|off) 488 489set drop enable 490~~~~~~~~~~~~~~~ 491 492set drop enable bit for all queues:: 493 494 testpmd> set all queues drop (port_id) (on|off) 495 496set split drop enable (for VF) 497~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 498 499set split drop enable bit for VF from PF:: 500 501 testpmd> set vf split drop (port_id) (vf_id) (on|off) 502 503set mac antispoof (for VF) 504~~~~~~~~~~~~~~~~~~~~~~~~~~ 505 506Set mac antispoof for a VF from the PF:: 507 508 testpmd> set vf mac antispoof (port_id) (vf_id) (on|off) 509 510vlan set strip 511~~~~~~~~~~~~~~ 512 513Set the VLAN strip on a port:: 514 515 testpmd> vlan set strip (on|off) (port_id) 516 517vlan set stripq 518~~~~~~~~~~~~~~~ 519 520Set the VLAN strip for a queue on a port:: 521 522 testpmd> vlan set stripq (on|off) (port_id,queue_id) 523 524vlan set stripq (for VF) 525~~~~~~~~~~~~~~~~~~~~~~~~ 526 527Set VLAN strip for all queues in a pool for a VF from the PF:: 528 529 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off) 530 531vlan set insert (for VF) 532~~~~~~~~~~~~~~~~~~~~~~~~ 533 534Set VLAN insert for a VF from the PF:: 535 536 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id) 537 538vlan set antispoof (for VF) 539~~~~~~~~~~~~~~~~~~~~~~~~~~~ 540 541Set VLAN antispoof for a VF from the PF:: 542 543 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off) 544 545vlan set filter 546~~~~~~~~~~~~~~~ 547 548Set the VLAN filter on a port:: 549 550 testpmd> vlan set filter (on|off) (port_id) 551 552vlan set qinq 553~~~~~~~~~~~~~ 554 555Set the VLAN QinQ (extended queue in queue) on for a port:: 556 557 testpmd> vlan set qinq (on|off) (port_id) 558 559vlan set tpid 560~~~~~~~~~~~~~ 561 562Set the inner or outer VLAN TPID for packet filtering on a port:: 563 564 testpmd> vlan set (inner|outer) tpid (value) (port_id) 565 566.. note:: 567 568 TPID value must be a 16-bit number (value <= 65536). 569 570rx_vlan add 571~~~~~~~~~~~ 572 573Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID:: 574 575 testpmd> rx_vlan add (vlan_id|all) (port_id) 576 577.. note:: 578 579 VLAN filter must be set on that port. VLAN ID < 4096. 580 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries 581 in VFTA table. This is important if enabling all vlan_ids. 582 583rx_vlan rm 584~~~~~~~~~~ 585 586Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID:: 587 588 testpmd> rx_vlan rm (vlan_id|all) (port_id) 589 590rx_vlan add (for VF) 591~~~~~~~~~~~~~~~~~~~~ 592 593Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID:: 594 595 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask) 596 597rx_vlan rm (for VF) 598~~~~~~~~~~~~~~~~~~~ 599 600Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID:: 601 602 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask) 603 604tunnel_filter add 605~~~~~~~~~~~~~~~~~ 606 607Add a tunnel filter on a port:: 608 609 testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \ 610 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\ 611 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id) 612 613The available information categories are: 614 615* ``vxlan``: Set tunnel type as VXLAN. 616 617* ``nvgre``: Set tunnel type as NVGRE. 618 619* ``ipingre``: Set tunnel type as IP-in-GRE. 620 621* ``imac-ivlan``: Set filter type as Inner MAC and VLAN. 622 623* ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID. 624 625* ``imac-tenid``: Set filter type as Inner MAC and tenant ID. 626 627* ``imac``: Set filter type as Inner MAC. 628 629* ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID. 630 631* ``oip``: Set filter type as Outer IP. 632 633* ``iip``: Set filter type as Inner IP. 634 635Example:: 636 637 testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \ 638 192.168.2.2 0 ipingre oip 1 1 639 640 Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP. 641 642tunnel_filter remove 643~~~~~~~~~~~~~~~~~~~~ 644 645Remove a tunnel filter on a port:: 646 647 testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \ 648 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\ 649 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id) 650 651rx_vxlan_port add 652~~~~~~~~~~~~~~~~~ 653 654Add an UDP port for VXLAN packet filter on a port:: 655 656 testpmd> rx_vxlan_port add (udp_port) (port_id) 657 658rx_vxlan_port remove 659~~~~~~~~~~~~~~~~~~~~ 660 661Remove an UDP port for VXLAN packet filter on a port:: 662 663 testpmd> rx_vxlan_port rm (udp_port) (port_id) 664 665tx_vlan set 666~~~~~~~~~~~ 667 668Set hardware insertion of VLAN IDs in packets sent on a port:: 669 670 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer] 671 672For example, set a single VLAN ID (5) insertion on port 0:: 673 674 tx_vlan set 0 5 675 676Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1:: 677 678 tx_vlan set 1 2 3 679 680 681tx_vlan set pvid 682~~~~~~~~~~~~~~~~ 683 684Set port based hardware insertion of VLAN ID in packets sent on a port:: 685 686 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off) 687 688tx_vlan reset 689~~~~~~~~~~~~~ 690 691Disable hardware insertion of a VLAN header in packets sent on a port:: 692 693 testpmd> tx_vlan reset (port_id) 694 695csum set 696~~~~~~~~ 697 698Select hardware or software calculation of the checksum when 699transmitting a packet using the ``csum`` forwarding engine:: 700 701 testpmd> csum set (ip|udp|tcp|sctp|outer-ip) (hw|sw) (port_id) 702 703Where: 704 705* ``ip|udp|tcp|sctp`` always relate to the inner layer. 706 707* ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized 708 as a tunnel packet by the forwarding engine (vxlan, gre and ipip are 709 supported). See also the ``csum parse-tunnel`` command. 710 711.. note:: 712 713 Check the NIC Datasheet for hardware limits. 714 715csum parse-tunnel 716~~~~~~~~~~~~~~~~~ 717 718Define how tunneled packets should be handled by the csum forward 719engine:: 720 721 testpmd> csum parse-tunnel (on|off) (tx_port_id) 722 723If enabled, the csum forward engine will try to recognize supported 724tunnel headers (vxlan, gre, ipip). 725 726If disabled, treat tunnel packets as non-tunneled packets (a inner 727header is handled as a packet payload). 728 729.. note:: 730 731 The port argument is the TX port like in the ``csum set`` command. 732 733Example: 734 735Consider a packet in packet like the following:: 736 737 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in 738 739* If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set`` 740 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the 741 ``outer-ip parameter`` relates to the outer headers (here ``ipv4_out``). 742 743* If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set`` 744 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``. 745 746csum show 747~~~~~~~~~ 748 749Display tx checksum offload configuration:: 750 751 testpmd> csum show (port_id) 752 753tso set 754~~~~~~~ 755 756Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine:: 757 758 testpmd> tso set (segsize) (port_id) 759 760.. note:: 761 762 Check the NIC datasheet for hardware limits. 763 764tso show 765~~~~~~~~ 766 767Display the status of TCP Segmentation Offload:: 768 769 testpmd> tso show (port_id) 770 771mac_addr add 772~~~~~~~~~~~~ 773 774Add an alternative MAC address to a port:: 775 776 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX) 777 778mac_addr remove 779~~~~~~~~~~~~~~~ 780 781Remove a MAC address from a port:: 782 783 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX) 784 785mac_addr add (for VF) 786~~~~~~~~~~~~~~~~~~~~~ 787 788Add an alternative MAC address for a VF to a port:: 789 790 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX) 791 792mac_addr set (for VF) 793~~~~~~~~~~~~~~~~~~~~~ 794 795Set the MAC address for a VF from the PF:: 796 797 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX) 798 799set port-uta 800~~~~~~~~~~~~ 801 802Set the unicast hash filter(s) on/off for a port:: 803 804 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off) 805 806set promisc 807~~~~~~~~~~~ 808 809Set the promiscuous mode on for a port or for all ports. 810In promiscuous mode packets are not dropped if they aren't for the specified MAC address:: 811 812 testpmd> set promisc (port_id|all) (on|off) 813 814set allmulti 815~~~~~~~~~~~~ 816 817Set the allmulti mode for a port or for all ports:: 818 819 testpmd> set allmulti (port_id|all) (on|off) 820 821Same as the ifconfig (8) option. Controls how multicast packets are handled. 822 823set flow_ctrl rx 824~~~~~~~~~~~~~~~~ 825 826Set the link flow control parameter on a port:: 827 828 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \ 829 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \ 830 autoneg (on|off) (port_id) 831 832Where: 833 834* ``high_water`` (integer): High threshold value to trigger XOFF. 835 836* ``low_water`` (integer): Low threshold value to trigger XON. 837 838* ``pause_time`` (integer): Pause quota in the Pause frame. 839 840* ``send_xon`` (0/1): Send XON frame. 841 842* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames. 843 844* ``autoneg``: Change the auto-negotiation para mete. 845 846set pfc_ctrl rx 847~~~~~~~~~~~~~~~ 848 849Set the priority flow control parameter on a port:: 850 851 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \ 852 (pause_time) (priority) (port_id) 853 854Where: 855 856* ``high_water`` (integer): High threshold value. 857 858* ``low_water`` (integer): Low threshold value. 859 860* ``pause_time`` (integer): Pause quota in the Pause frame. 861 862* ``priority`` (0-7): VLAN User Priority. 863 864set stat_qmap 865~~~~~~~~~~~~~ 866 867Set statistics mapping (qmapping 0..15) for RX/TX queue on port:: 868 869 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping) 870 871For example, to set rx queue 2 on port 0 to mapping 5:: 872 873 testpmd>set stat_qmap rx 0 2 5 874 875set port - rx/tx (for VF) 876~~~~~~~~~~~~~~~~~~~~~~~~~ 877 878Set VF receive/transmit from a port:: 879 880 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off) 881 882set port - mac address filter (for VF) 883~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 884 885Add/Remove unicast or multicast MAC addr filter for a VF:: 886 887 testpmd> set port (port_id) vf (vf_id) (mac_addr) \ 888 (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off) 889 890set port - rx mode(for VF) 891~~~~~~~~~~~~~~~~~~~~~~~~~~ 892 893Set the VF receive mode of a port:: 894 895 testpmd> set port (port_id) vf (vf_id) \ 896 rxmode (AUPE|ROPE|BAM|MPE) (on|off) 897 898The available receive modes are: 899 900* ``AUPE``: Accepts untagged VLAN. 901 902* ``ROPE``: Accepts unicast hash. 903 904* ``BAM``: Accepts broadcast packets. 905 906* ``MPE``: Accepts all multicast packets. 907 908set port - tx_rate (for Queue) 909~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 910 911Set TX rate limitation for a queue on a port:: 912 913 testpmd> set port (port_id) queue (queue_id) rate (rate_value) 914 915set port - tx_rate (for VF) 916~~~~~~~~~~~~~~~~~~~~~~~~~~~ 917 918Set TX rate limitation for queues in VF on a port:: 919 920 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask) 921 922set port - mirror rule 923~~~~~~~~~~~~~~~~~~~~~~ 924 925Set pool or vlan type mirror rule for a port:: 926 927 testpmd> set port (port_id) mirror-rule (rule_id) \ 928 (pool-mirror-up|pool-mirror-down|vlan-mirror) \ 929 (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off) 930 931Set link mirror rule for a port:: 932 933 testpmd> set port (port_id) mirror-rule (rule_id) \ 934 (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off) 935 936For example to enable mirror traffic with vlan 0,1 to pool 0:: 937 938 set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on 939 940reset port - mirror rule 941~~~~~~~~~~~~~~~~~~~~~~~~ 942 943Reset a mirror rule for a port:: 944 945 testpmd> reset port (port_id) mirror-rule (rule_id) 946 947set flush_rx 948~~~~~~~~~~~~ 949 950Set the flush on RX streams before forwarding. 951The default is flush ``on``. 952Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams:: 953 954 testpmd> set flush_rx off 955 956set bypass mode 957~~~~~~~~~~~~~~~ 958 959Set the bypass mode for the lowest port on bypass enabled NIC:: 960 961 testpmd> set bypass mode (normal|bypass|isolate) (port_id) 962 963set bypass event 964~~~~~~~~~~~~~~~~ 965 966Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled:: 967 968 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \ 969 mode (normal|bypass|isolate) (port_id) 970 971Where: 972 973* ``timeout``: Enable bypass after watchdog timeout. 974 975* ``os_on``: Enable bypass when OS/board is powered on. 976 977* ``os_off``: Enable bypass when OS/board is powered off. 978 979* ``power_on``: Enable bypass when power supply is turned on. 980 981* ``power_off``: Enable bypass when power supply is turned off. 982 983 984set bypass timeout 985~~~~~~~~~~~~~~~~~~ 986 987Set the bypass watchdog timeout to ``n`` seconds where 0 = instant:: 988 989 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32) 990 991show bypass config 992~~~~~~~~~~~~~~~~~~ 993 994Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC:: 995 996 testpmd> show bypass config (port_id) 997 998set link up 999~~~~~~~~~~~ 1000 1001Set link up for a port:: 1002 1003 testpmd> set link-up port (port id) 1004 1005set link down 1006~~~~~~~~~~~~~ 1007 1008Set link down for a port:: 1009 1010 testpmd> set link-down port (port id) 1011 1012E-tag set 1013~~~~~~~~~ 1014 1015Enable E-tag insertion for a VF on a port:: 1016 1017 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id) 1018 1019Disable E-tag insertion for a VF on a port:: 1020 1021 testpmd> E-tag set insertion off port (port_id) vf (vf_id) 1022 1023Enable/disable E-tag stripping on a port:: 1024 1025 testpmd> E-tag set stripping (on|off) port (port_id) 1026 1027Enable/disable E-tag based forwarding on a port:: 1028 1029 testpmd> E-tag set forwarding (on|off) port (port_id) 1030 1031Add an E-tag forwarding filter on a port:: 1032 1033 testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id) 1034 1035Delete an E-tag forwarding filter on a port:: 1036 testpmd> E-tag set filter del e-tag-id (value) port (port_id) 1037 1038 1039Port Functions 1040-------------- 1041 1042The following sections show functions for configuring ports. 1043 1044.. note:: 1045 1046 Port configuration changes only become active when forwarding is started/restarted. 1047 1048port attach 1049~~~~~~~~~~~ 1050 1051Attach a port specified by pci address or virtual device args:: 1052 1053 testpmd> port attach (identifier) 1054 1055To attach a new pci device, the device should be recognized by kernel first. 1056Then it should be moved under DPDK management. 1057Finally the port can be attached to testpmd. 1058 1059For example, to move a pci device using ixgbe under DPDK management: 1060 1061.. code-block:: console 1062 1063 # Check the status of the available devices. 1064 ./usertools/dpdk-devbind.py --status 1065 1066 Network devices using DPDK-compatible driver 1067 ============================================ 1068 <none> 1069 1070 Network devices using kernel driver 1071 =================================== 1072 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused= 1073 1074 1075 # Bind the device to igb_uio. 1076 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0 1077 1078 1079 # Recheck the status of the devices. 1080 ./usertools/dpdk-devbind.py --status 1081 Network devices using DPDK-compatible driver 1082 ============================================ 1083 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused= 1084 1085To attach a port created by virtual device, above steps are not needed. 1086 1087For example, to attach a port whose pci address is 0000:0a:00.0. 1088 1089.. code-block:: console 1090 1091 testpmd> port attach 0000:0a:00.0 1092 Attaching a new port... 1093 EAL: PCI device 0000:0a:00.0 on NUMA socket -1 1094 EAL: probe driver: 8086:10fb rte_ixgbe_pmd 1095 EAL: PCI memory mapped at 0x7f83bfa00000 1096 EAL: PCI memory mapped at 0x7f83bfa80000 1097 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5 1098 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb 1099 Port 0 is attached. Now total ports is 1 1100 Done 1101 1102For example, to attach a port created by pcap PMD. 1103 1104.. code-block:: console 1105 1106 testpmd> port attach net_pcap0 1107 Attaching a new port... 1108 PMD: Initializing pmd_pcap for net_pcap0 1109 PMD: Creating pcap-backed ethdev on numa socket 0 1110 Port 0 is attached. Now total ports is 1 1111 Done 1112 1113In this case, identifier is ``net_pcap0``. 1114This identifier format is the same as ``--vdev`` format of DPDK applications. 1115 1116For example, to re-attach a bonded port which has been previously detached, 1117the mode and slave parameters must be given. 1118 1119.. code-block:: console 1120 1121 testpmd> port attach net_bond_0,mode=0,slave=1 1122 Attaching a new port... 1123 EAL: Initializing pmd_bond for net_bond_0 1124 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0. 1125 Port 0 is attached. Now total ports is 1 1126 Done 1127 1128 1129port detach 1130~~~~~~~~~~~ 1131 1132Detach a specific port:: 1133 1134 testpmd> port detach (port_id) 1135 1136Before detaching a port, the port should be stopped and closed. 1137 1138For example, to detach a pci device port 0. 1139 1140.. code-block:: console 1141 1142 testpmd> port stop 0 1143 Stopping ports... 1144 Done 1145 testpmd> port close 0 1146 Closing ports... 1147 Done 1148 1149 testpmd> port detach 0 1150 Detaching a port... 1151 EAL: PCI device 0000:0a:00.0 on NUMA socket -1 1152 EAL: remove driver: 8086:10fb rte_ixgbe_pmd 1153 EAL: PCI memory unmapped at 0x7f83bfa00000 1154 EAL: PCI memory unmapped at 0x7f83bfa80000 1155 Done 1156 1157 1158For example, to detach a virtual device port 0. 1159 1160.. code-block:: console 1161 1162 testpmd> port stop 0 1163 Stopping ports... 1164 Done 1165 testpmd> port close 0 1166 Closing ports... 1167 Done 1168 1169 testpmd> port detach 0 1170 Detaching a port... 1171 PMD: Closing pcap ethdev on numa socket 0 1172 Port 'net_pcap0' is detached. Now total ports is 0 1173 Done 1174 1175To remove a pci device completely from the system, first detach the port from testpmd. 1176Then the device should be moved under kernel management. 1177Finally the device can be removed using kernel pci hotplug functionality. 1178 1179For example, to move a pci device under kernel management: 1180 1181.. code-block:: console 1182 1183 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0 1184 1185 ./usertools/dpdk-devbind.py --status 1186 1187 Network devices using DPDK-compatible driver 1188 ============================================ 1189 <none> 1190 1191 Network devices using kernel driver 1192 =================================== 1193 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio 1194 1195To remove a port created by a virtual device, above steps are not needed. 1196 1197port start 1198~~~~~~~~~~ 1199 1200Start all ports or a specific port:: 1201 1202 testpmd> port start (port_id|all) 1203 1204port stop 1205~~~~~~~~~ 1206 1207Stop all ports or a specific port:: 1208 1209 testpmd> port stop (port_id|all) 1210 1211port close 1212~~~~~~~~~~ 1213 1214Close all ports or a specific port:: 1215 1216 testpmd> port close (port_id|all) 1217 1218port start/stop queue 1219~~~~~~~~~~~~~~~~~~~~~ 1220 1221Start/stop a rx/tx queue on a specific port:: 1222 1223 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop) 1224 1225Only take effect when port is started. 1226 1227port config - speed 1228~~~~~~~~~~~~~~~~~~~ 1229 1230Set the speed and duplex mode for all ports or a specific port:: 1231 1232 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \ 1233 duplex (half|full|auto) 1234 1235port config - queues/descriptors 1236~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1237 1238Set number of queues/descriptors for rxq, txq, rxd and txd:: 1239 1240 testpmd> port config all (rxq|txq|rxd|txd) (value) 1241 1242This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options. 1243 1244port config - max-pkt-len 1245~~~~~~~~~~~~~~~~~~~~~~~~~ 1246 1247Set the maximum packet length:: 1248 1249 testpmd> port config all max-pkt-len (value) 1250 1251This is equivalent to the ``--max-pkt-len`` command-line option. 1252 1253port config - CRC Strip 1254~~~~~~~~~~~~~~~~~~~~~~~ 1255 1256Set hardware CRC stripping on or off for all ports:: 1257 1258 testpmd> port config all crc-strip (on|off) 1259 1260CRC stripping is off by default. 1261 1262The ``on`` option is equivalent to the ``--crc-strip`` command-line option. 1263 1264port config - scatter 1265~~~~~~~~~~~~~~~~~~~~~~~ 1266 1267Set RX scatter mode on or off for all ports:: 1268 1269 testpmd> port config all scatter (on|off) 1270 1271RX scatter mode is off by default. 1272 1273The ``on`` option is equivalent to the ``--enable-scatter`` command-line option. 1274 1275port config - TX queue flags 1276~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1277 1278Set a hexadecimal bitmap of TX queue flags for all ports:: 1279 1280 testpmd> port config all txqflags value 1281 1282This command is equivalent to the ``--txqflags`` command-line option. 1283 1284port config - RX Checksum 1285~~~~~~~~~~~~~~~~~~~~~~~~~ 1286 1287Set hardware RX checksum offload to on or off for all ports:: 1288 1289 testpmd> port config all rx-cksum (on|off) 1290 1291Checksum offload is off by default. 1292 1293The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option. 1294 1295port config - VLAN 1296~~~~~~~~~~~~~~~~~~ 1297 1298Set hardware VLAN on or off for all ports:: 1299 1300 testpmd> port config all hw-vlan (on|off) 1301 1302Hardware VLAN is on by default. 1303 1304The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option. 1305 1306port config - VLAN filter 1307~~~~~~~~~~~~~~~~~~~~~~~~~ 1308 1309Set hardware VLAN filter on or off for all ports:: 1310 1311 testpmd> port config all hw-vlan-filter (on|off) 1312 1313Hardware VLAN filter is on by default. 1314 1315The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option. 1316 1317port config - VLAN strip 1318~~~~~~~~~~~~~~~~~~~~~~~~ 1319 1320Set hardware VLAN strip on or off for all ports:: 1321 1322 testpmd> port config all hw-vlan-strip (on|off) 1323 1324Hardware VLAN strip is on by default. 1325 1326The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option. 1327 1328port config - VLAN extend 1329~~~~~~~~~~~~~~~~~~~~~~~~~ 1330 1331Set hardware VLAN extend on or off for all ports:: 1332 1333 testpmd> port config all hw-vlan-extend (on|off) 1334 1335Hardware VLAN extend is off by default. 1336 1337The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option. 1338 1339port config - Drop Packets 1340~~~~~~~~~~~~~~~~~~~~~~~~~~ 1341 1342Set packet drop for packets with no descriptors on or off for all ports:: 1343 1344 testpmd> port config all drop-en (on|off) 1345 1346Packet dropping for packets with no descriptors is off by default. 1347 1348The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option. 1349 1350port config - RSS 1351~~~~~~~~~~~~~~~~~ 1352 1353Set the RSS (Receive Side Scaling) mode on or off:: 1354 1355 testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none) 1356 1357RSS is on by default. 1358 1359The ``none`` option is equivalent to the ``--disable-rss`` command-line option. 1360 1361port config - RSS Reta 1362~~~~~~~~~~~~~~~~~~~~~~ 1363 1364Set the RSS (Receive Side Scaling) redirection table:: 1365 1366 testpmd> port config all rss reta (hash,queue)[,(hash,queue)] 1367 1368port config - DCB 1369~~~~~~~~~~~~~~~~~ 1370 1371Set the DCB mode for an individual port:: 1372 1373 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off) 1374 1375The traffic class should be 4 or 8. 1376 1377port config - Burst 1378~~~~~~~~~~~~~~~~~~~ 1379 1380Set the number of packets per burst:: 1381 1382 testpmd> port config all burst (value) 1383 1384This is equivalent to the ``--burst`` command-line option. 1385 1386port config - Threshold 1387~~~~~~~~~~~~~~~~~~~~~~~ 1388 1389Set thresholds for TX/RX queues:: 1390 1391 testpmd> port config all (threshold) (value) 1392 1393Where the threshold type can be: 1394 1395* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255. 1396 1397* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255. 1398 1399* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255. 1400 1401* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255. 1402 1403* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255. 1404 1405* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255. 1406 1407* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd. 1408 1409* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd. 1410 1411* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd. 1412 1413These threshold options are also available from the command-line. 1414 1415port config - E-tag 1416~~~~~~~~~~~~~~~~~~~ 1417 1418Set the value of ether-type for E-tag:: 1419 1420 testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value) 1421 1422Enable/disable the E-tag support:: 1423 1424 testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable) 1425 1426 1427Link Bonding Functions 1428---------------------- 1429 1430The Link Bonding functions make it possible to dynamically create and 1431manage link bonding devices from within testpmd interactive prompt. 1432 1433create bonded device 1434~~~~~~~~~~~~~~~~~~~~ 1435 1436Create a new bonding device:: 1437 1438 testpmd> create bonded device (mode) (socket) 1439 1440For example, to create a bonded device in mode 1 on socket 0:: 1441 1442 testpmd> create bonded 1 0 1443 created new bonded device (port X) 1444 1445add bonding slave 1446~~~~~~~~~~~~~~~~~ 1447 1448Adds Ethernet device to a Link Bonding device:: 1449 1450 testpmd> add bonding slave (slave id) (port id) 1451 1452For example, to add Ethernet device (port 6) to a Link Bonding device (port 10):: 1453 1454 testpmd> add bonding slave 6 10 1455 1456 1457remove bonding slave 1458~~~~~~~~~~~~~~~~~~~~ 1459 1460Removes an Ethernet slave device from a Link Bonding device:: 1461 1462 testpmd> remove bonding slave (slave id) (port id) 1463 1464For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10):: 1465 1466 testpmd> remove bonding slave 6 10 1467 1468set bonding mode 1469~~~~~~~~~~~~~~~~ 1470 1471Set the Link Bonding mode of a Link Bonding device:: 1472 1473 testpmd> set bonding mode (value) (port id) 1474 1475For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3):: 1476 1477 testpmd> set bonding mode 3 10 1478 1479set bonding primary 1480~~~~~~~~~~~~~~~~~~~ 1481 1482Set an Ethernet slave device as the primary device on a Link Bonding device:: 1483 1484 testpmd> set bonding primary (slave id) (port id) 1485 1486For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10):: 1487 1488 testpmd> set bonding primary 6 10 1489 1490set bonding mac 1491~~~~~~~~~~~~~~~ 1492 1493Set the MAC address of a Link Bonding device:: 1494 1495 testpmd> set bonding mac (port id) (mac) 1496 1497For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01:: 1498 1499 testpmd> set bonding mac 10 00:00:00:00:00:01 1500 1501set bonding xmit_balance_policy 1502~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1503 1504Set the transmission policy for a Link Bonding device when it is in Balance XOR mode:: 1505 1506 testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34) 1507 1508For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports):: 1509 1510 testpmd> set bonding xmit_balance_policy 10 l34 1511 1512 1513set bonding mon_period 1514~~~~~~~~~~~~~~~~~~~~~~ 1515 1516Set the link status monitoring polling period in milliseconds for a bonding device. 1517 1518This adds support for PMD slave devices which do not support link status interrupts. 1519When the mon_period is set to a value greater than 0 then all PMD's which do not support 1520link status ISR will be queried every polling interval to check if their link status has changed:: 1521 1522 testpmd> set bonding mon_period (port_id) (value) 1523 1524For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms:: 1525 1526 testpmd> set bonding mon_period 5 150 1527 1528 1529show bonding config 1530~~~~~~~~~~~~~~~~~~~ 1531 1532Show the current configuration of a Link Bonding device:: 1533 1534 testpmd> show bonding config (port id) 1535 1536For example, 1537to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4) 1538in balance mode with a transmission policy of layer 2+3:: 1539 1540 testpmd> show bonding config 9 1541 Bonding mode: 2 1542 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23 1543 Slaves (3): [1 3 4] 1544 Active Slaves (3): [1 3 4] 1545 Primary: [3] 1546 1547 1548Register Functions 1549------------------ 1550 1551The Register Functions can be used to read from and write to registers on the network card referenced by a port number. 1552This is mainly useful for debugging purposes. 1553Reference should be made to the appropriate datasheet for the network card for details on the register addresses 1554and fields that can be accessed. 1555 1556read reg 1557~~~~~~~~ 1558 1559Display the value of a port register:: 1560 1561 testpmd> read reg (port_id) (address) 1562 1563For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller:: 1564 1565 testpmd> read reg 0 0xEE00 1566 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241) 1567 1568read regfield 1569~~~~~~~~~~~~~ 1570 1571Display a port register bit field:: 1572 1573 testpmd> read regfield (port_id) (address) (bit_x) (bit_y) 1574 1575For example, reading the lowest two bits from the register in the example above:: 1576 1577 testpmd> read regfield 0 0xEE00 0 1 1578 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1) 1579 1580read regbit 1581~~~~~~~~~~~ 1582 1583Display a single port register bit:: 1584 1585 testpmd> read regbit (port_id) (address) (bit_x) 1586 1587For example, reading the lowest bit from the register in the example above:: 1588 1589 testpmd> read regbit 0 0xEE00 0 1590 port 0 PCI register at offset 0xEE00: bit 0=1 1591 1592write reg 1593~~~~~~~~~ 1594 1595Set the value of a port register:: 1596 1597 testpmd> write reg (port_id) (address) (value) 1598 1599For example, to clear a register:: 1600 1601 testpmd> write reg 0 0xEE00 0x0 1602 port 0 PCI register at offset 0xEE00: 0x00000000 (0) 1603 1604write regfield 1605~~~~~~~~~~~~~~ 1606 1607Set bit field of a port register:: 1608 1609 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value) 1610 1611For example, writing to the register cleared in the example above:: 1612 1613 testpmd> write regfield 0 0xEE00 0 1 2 1614 port 0 PCI register at offset 0xEE00: 0x00000002 (2) 1615 1616write regbit 1617~~~~~~~~~~~~ 1618 1619Set single bit value of a port register:: 1620 1621 testpmd> write regbit (port_id) (address) (bit_x) (value) 1622 1623For example, to set the high bit in the register from the example above:: 1624 1625 testpmd> write regbit 0 0xEE00 31 1 1626 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658) 1627 1628 1629Filter Functions 1630---------------- 1631 1632This section details the available filter functions that are available. 1633 1634Note these functions interface the deprecated legacy filtering framework, 1635superseded by *rte_flow*. See `Flow rules management`_. 1636 1637ethertype_filter 1638~~~~~~~~~~~~~~~~~~~~ 1639 1640Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue:: 1641 1642 ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \ 1643 ethertype (ether_type) (drop|fwd) queue (queue_id) 1644 1645The available information parameters are: 1646 1647* ``port_id``: The port which the Ethertype filter assigned on. 1648 1649* ``mac_addr``: Compare destination mac address. 1650 1651* ``mac_ignr``: Ignore destination mac address match. 1652 1653* ``mac_address``: Destination mac address to match. 1654 1655* ``ether_type``: The EtherType value want to match, 1656 for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid. 1657 1658* ``queue_id``: The receive queue associated with this EtherType filter. 1659 It is meaningless when deleting or dropping. 1660 1661Example, to add/remove an ethertype filter rule:: 1662 1663 testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \ 1664 ethertype 0x0806 fwd queue 3 1665 1666 testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \ 1667 ethertype 0x0806 fwd queue 3 1668 16692tuple_filter 1670~~~~~~~~~~~~~~~~~ 1671 1672Add or delete a 2-tuple filter, 1673which identifies packets by specific protocol and destination TCP/UDP port 1674and forwards packets into one of the receive queues:: 1675 1676 2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \ 1677 protocol (protocol_value) mask (mask_value) \ 1678 tcp_flags (tcp_flags_value) priority (prio_value) \ 1679 queue (queue_id) 1680 1681The available information parameters are: 1682 1683* ``port_id``: The port which the 2-tuple filter assigned on. 1684 1685* ``dst_port_value``: Destination port in L4. 1686 1687* ``protocol_value``: IP L4 protocol. 1688 1689* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate. 1690 1691* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP). 1692 1693* ``prio_value``: Priority of this filter. 1694 1695* ``queue_id``: The receive queue associated with this 2-tuple filter. 1696 1697Example, to add/remove an 2tuple filter rule:: 1698 1699 testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \ 1700 tcp_flags 0x02 priority 3 queue 3 1701 1702 testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \ 1703 tcp_flags 0x02 priority 3 queue 3 1704 17055tuple_filter 1706~~~~~~~~~~~~~~~~~ 1707 1708Add or delete a 5-tuple filter, 1709which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port) 1710and routes packets into one of the receive queues:: 1711 1712 5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \ 1713 (src_address) dst_port (dst_port_value) \ 1714 src_port (src_port_value) protocol (protocol_value) \ 1715 mask (mask_value) tcp_flags (tcp_flags_value) \ 1716 priority (prio_value) queue (queue_id) 1717 1718The available information parameters are: 1719 1720* ``port_id``: The port which the 5-tuple filter assigned on. 1721 1722* ``dst_address``: Destination IP address. 1723 1724* ``src_address``: Source IP address. 1725 1726* ``dst_port_value``: TCP/UDP destination port. 1727 1728* ``src_port_value``: TCP/UDP source port. 1729 1730* ``protocol_value``: L4 protocol. 1731 1732* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate 1733 1734* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP). 1735 1736* ``prio_value``: The priority of this filter. 1737 1738* ``queue_id``: The receive queue associated with this 5-tuple filter. 1739 1740Example, to add/remove an 5tuple filter rule:: 1741 1742 testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \ 1743 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \ 1744 flags 0x0 priority 3 queue 3 1745 1746 testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \ 1747 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \ 1748 flags 0x0 priority 3 queue 3 1749 1750syn_filter 1751~~~~~~~~~~ 1752 1753Using the SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue:: 1754 1755 syn_filter (port_id) (add|del) priority (high|low) queue (queue_id) 1756 1757The available information parameters are: 1758 1759* ``port_id``: The port which the SYN filter assigned on. 1760 1761* ``high``: This SYN filter has higher priority than other filters. 1762 1763* ``low``: This SYN filter has lower priority than other filters. 1764 1765* ``queue_id``: The receive queue associated with this SYN filter 1766 1767Example:: 1768 1769 testpmd> syn_filter 0 add priority high queue 3 1770 1771flex_filter 1772~~~~~~~~~~~ 1773 1774With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet 1775and routed into one of the receive queues:: 1776 1777 flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \ 1778 mask (mask_value) priority (prio_value) queue (queue_id) 1779 1780The available information parameters are: 1781 1782* ``port_id``: The port which the Flex filter is assigned on. 1783 1784* ``len_value``: Filter length in bytes, no greater than 128. 1785 1786* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match. 1787 1788* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match. 1789 1790* ``prio_value``: The priority of this filter. 1791 1792* ``queue_id``: The receive queue associated with this Flex filter. 1793 1794Example:: 1795 1796 testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \ 1797 mask 000C priority 3 queue 3 1798 1799 testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \ 1800 mask 000C priority 3 queue 3 1801 1802 1803.. _testpmd_flow_director: 1804 1805flow_director_filter 1806~~~~~~~~~~~~~~~~~~~~ 1807 1808The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues. 1809 1810Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and 1811Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter: 1812 1813* Perfect match filters. 1814 The hardware checks a match between the masked fields of the received packets and the programmed filters. 1815 The masked fields are for IP flow. 1816 1817* Signature filters. 1818 The hardware checks a match between a hash-based signature of the masked fields of the received packet. 1819 1820* Perfect-mac-vlan match filters. 1821 The hardware checks a match between the masked fields of the received packets and the programmed filters. 1822 The masked fields are for MAC VLAN flow. 1823 1824* Perfect-tunnel match filters. 1825 The hardware checks a match between the masked fields of the received packets and the programmed filters. 1826 The masked fields are for tunnel flow. 1827 1828The Flow Director filters can match the different fields for different type of packet: flow type, specific input set 1829per flow type and the flexible payload. 1830 1831The Flow Director can also mask out parts of all of these fields so that filters 1832are only applied to certain fields or parts of the fields. 1833 1834Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information. 1835 1836# Commands to add flow director filters of different flow types:: 1837 1838 flow_director_filter (port_id) mode IP (add|del|update) \ 1839 flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \ 1840 src (src_ip_address) dst (dst_ip_address) \ 1841 tos (tos_value) proto (proto_value) ttl (ttl_value) \ 1842 vlan (vlan_value) flexbytes (flexbytes_value) \ 1843 (drop|fwd) pf|vf(vf_id) queue (queue_id) \ 1844 fd_id (fd_id_value) 1845 1846 flow_director_filter (port_id) mode IP (add|del|update) \ 1847 flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \ 1848 src (src_ip_address) (src_port) \ 1849 dst (dst_ip_address) (dst_port) \ 1850 tos (tos_value) ttl (ttl_value) \ 1851 vlan (vlan_value) flexbytes (flexbytes_value) \ 1852 (drop|fwd) queue pf|vf(vf_id) (queue_id) \ 1853 fd_id (fd_id_value) 1854 1855 flow_director_filter (port_id) mode IP (add|del|update) \ 1856 flow (ipv4-sctp|ipv6-sctp) \ 1857 src (src_ip_address) (src_port) \ 1858 dst (dst_ip_address) (dst_port) \ 1859 tos (tos_value) ttl (ttl_value) \ 1860 tag (verification_tag) vlan (vlan_value) \ 1861 flexbytes (flexbytes_value) (drop|fwd) \ 1862 pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value) 1863 1864 flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \ 1865 ether (ethertype) flexbytes (flexbytes_value) \ 1866 (drop|fwd) pf|vf(vf_id) queue (queue_id) 1867 fd_id (fd_id_value) 1868 1869 flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \ 1870 mac (mac_address) vlan (vlan_value) \ 1871 flexbytes (flexbytes_value) (drop|fwd) \ 1872 queue (queue_id) fd_id (fd_id_value) 1873 1874 flow_director_filter (port_id) mode Tunnel (add|del|update) \ 1875 mac (mac_address) vlan (vlan_value) \ 1876 tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \ 1877 flexbytes (flexbytes_value) (drop|fwd) \ 1878 queue (queue_id) fd_id (fd_id_value) 1879 1880For example, to add an ipv4-udp flow type filter:: 1881 1882 testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \ 1883 dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \ 1884 fwd pf queue 1 fd_id 1 1885 1886For example, add an ipv4-other flow type filter:: 1887 1888 testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \ 1889 dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \ 1890 flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1 1891 1892flush_flow_director 1893~~~~~~~~~~~~~~~~~~~ 1894 1895Flush all flow director filters on a device:: 1896 1897 testpmd> flush_flow_director (port_id) 1898 1899Example, to flush all flow director filter on port 0:: 1900 1901 testpmd> flush_flow_director 0 1902 1903flow_director_mask 1904~~~~~~~~~~~~~~~~~~ 1905 1906Set flow director's input masks:: 1907 1908 flow_director_mask (port_id) mode IP vlan (vlan_value) \ 1909 src_mask (ipv4_src) (ipv6_src) (src_port) \ 1910 dst_mask (ipv4_dst) (ipv6_dst) (dst_port) 1911 1912 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value) 1913 1914 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \ 1915 mac (mac_value) tunnel-type (tunnel_type_value) \ 1916 tunnel-id (tunnel_id_value) 1917 1918Example, to set flow director mask on port 0:: 1919 1920 testpmd> flow_director_mask 0 mode IP vlan 0xefff \ 1921 src_mask 255.255.255.255 \ 1922 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \ 1923 dst_mask 255.255.255.255 \ 1924 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF 1925 1926flow_director_flex_mask 1927~~~~~~~~~~~~~~~~~~~~~~~ 1928 1929set masks of flow director's flexible payload based on certain flow type:: 1930 1931 testpmd> flow_director_flex_mask (port_id) \ 1932 flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 1933 ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \ 1934 l2_payload|all) (mask) 1935 1936Example, to set flow director's flex mask for all flow type on port 0:: 1937 1938 testpmd> flow_director_flex_mask 0 flow all \ 1939 (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0) 1940 1941 1942flow_director_flex_payload 1943~~~~~~~~~~~~~~~~~~~~~~~~~~ 1944 1945Configure flexible payload selection:: 1946 1947 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config) 1948 1949For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload:: 1950 1951 testpmd> flow_director_flex_payload 0 l4 \ 1952 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19) 1953 1954get_sym_hash_ena_per_port 1955~~~~~~~~~~~~~~~~~~~~~~~~~ 1956 1957Get symmetric hash enable configuration per port:: 1958 1959 get_sym_hash_ena_per_port (port_id) 1960 1961For example, to get symmetric hash enable configuration of port 1:: 1962 1963 testpmd> get_sym_hash_ena_per_port 1 1964 1965set_sym_hash_ena_per_port 1966~~~~~~~~~~~~~~~~~~~~~~~~~ 1967 1968Set symmetric hash enable configuration per port to enable or disable:: 1969 1970 set_sym_hash_ena_per_port (port_id) (enable|disable) 1971 1972For example, to set symmetric hash enable configuration of port 1 to enable:: 1973 1974 testpmd> set_sym_hash_ena_per_port 1 enable 1975 1976get_hash_global_config 1977~~~~~~~~~~~~~~~~~~~~~~ 1978 1979Get the global configurations of hash filters:: 1980 1981 get_hash_global_config (port_id) 1982 1983For example, to get the global configurations of hash filters of port 1:: 1984 1985 testpmd> get_hash_global_config 1 1986 1987set_hash_global_config 1988~~~~~~~~~~~~~~~~~~~~~~ 1989 1990Set the global configurations of hash filters:: 1991 1992 set_hash_global_config (port_id) (toeplitz|simple_xor|default) \ 1993 (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \ 1994 ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \ 1995 (enable|disable) 1996 1997For example, to enable simple_xor for flow type of ipv6 on port 2:: 1998 1999 testpmd> set_hash_global_config 2 simple_xor ipv6 enable 2000 2001set_hash_input_set 2002~~~~~~~~~~~~~~~~~~ 2003 2004Set the input set for hash:: 2005 2006 set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2007 ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \ 2008 l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \ 2009 ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \ 2010 tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \ 2011 udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \ 2012 fld-8th|none) (select|add) 2013 2014For example, to add source IP to hash input set for flow type of ipv4-udp on port 0:: 2015 2016 testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add 2017 2018set_fdir_input_set 2019~~~~~~~~~~~~~~~~~~ 2020 2021The Flow Director filters can match the different fields for different type of packet, i.e. specific input set 2022on per flow type and the flexible payload. This command can be used to change input set for each flow type. 2023 2024Set the input set for flow director:: 2025 2026 set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2027 ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \ 2028 l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \ 2029 ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \ 2030 tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \ 2031 sctp-dst-port|sctp-veri-tag|none) (select|add) 2032 2033For example to add source IP to FD input set for flow type of ipv4-udp on port 0:: 2034 2035 testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add 2036 2037global_config 2038~~~~~~~~~~~~~ 2039 2040Set different GRE key length for input set:: 2041 2042 global_config (port_id) gre-key-len (number in bytes) 2043 2044For example to set GRE key length for input set to 4 bytes on port 0:: 2045 2046 testpmd> global_config 0 gre-key-len 4 2047 2048 2049.. _testpmd_rte_flow: 2050 2051Flow rules management 2052--------------------- 2053 2054Control of the generic flow API (*rte_flow*) is fully exposed through the 2055``flow`` command (validation, creation, destruction and queries). 2056 2057Considering *rte_flow* overlaps with all `Filter Functions`_, using both 2058features simultaneously may cause undefined side-effects and is therefore 2059not recommended. 2060 2061``flow`` syntax 2062~~~~~~~~~~~~~~~ 2063 2064Because the ``flow`` command uses dynamic tokens to handle the large number 2065of possible flow rules combinations, its behavior differs slightly from 2066other commands, in particular: 2067 2068- Pressing *?* or the *<tab>* key displays contextual help for the current 2069 token, not that of the entire command. 2070 2071- Optional and repeated parameters are supported (provided they are listed 2072 in the contextual help). 2073 2074The first parameter stands for the operation mode. Possible operations and 2075their general syntax are described below. They are covered in detail in the 2076following sections. 2077 2078- Check whether a flow rule can be created:: 2079 2080 flow validate {port_id} 2081 [group {group_id}] [priority {level}] [ingress] [egress] 2082 pattern {item} [/ {item} [...]] / end 2083 actions {action} [/ {action} [...]] / end 2084 2085- Create a flow rule:: 2086 2087 flow create {port_id} 2088 [group {group_id}] [priority {level}] [ingress] [egress] 2089 pattern {item} [/ {item} [...]] / end 2090 actions {action} [/ {action} [...]] / end 2091 2092- Destroy specific flow rules:: 2093 2094 flow destroy {port_id} rule {rule_id} [...] 2095 2096- Destroy all flow rules:: 2097 2098 flow flush {port_id} 2099 2100- Query an existing flow rule:: 2101 2102 flow query {port_id} {rule_id} {action} 2103 2104- List existing flow rules sorted by priority, filtered by group 2105 identifiers:: 2106 2107 flow list {port_id} [group {group_id}] [...] 2108 2109Validating flow rules 2110~~~~~~~~~~~~~~~~~~~~~ 2111 2112``flow validate`` reports whether a flow rule would be accepted by the 2113underlying device in its current state but stops short of creating it. It is 2114bound to ``rte_flow_validate()``:: 2115 2116 flow validate {port_id} 2117 [group {group_id}] [priority {level}] [ingress] [egress] 2118 pattern {item} [/ {item} [...]] / end 2119 actions {action} [/ {action} [...]] / end 2120 2121If successful, it will show:: 2122 2123 Flow rule validated 2124 2125Otherwise it will show an error message of the form:: 2126 2127 Caught error type [...] ([...]): [...] 2128 2129This command uses the same parameters as ``flow create``, their format is 2130described in `Creating flow rules`_. 2131 2132Check whether redirecting any Ethernet packet received on port 0 to RX queue 2133index 6 is supported:: 2134 2135 testpmd> flow validate 0 ingress pattern eth / end 2136 actions queue index 6 / end 2137 Flow rule validated 2138 testpmd> 2139 2140Port 0 does not support TCPv6 rules:: 2141 2142 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end 2143 actions drop / end 2144 Caught error type 9 (specific pattern item): Invalid argument 2145 testpmd> 2146 2147Creating flow rules 2148~~~~~~~~~~~~~~~~~~~ 2149 2150``flow create`` validates and creates the specified flow rule. It is bound 2151to ``rte_flow_create()``:: 2152 2153 flow create {port_id} 2154 [group {group_id}] [priority {level}] [ingress] [egress] 2155 pattern {item} [/ {item} [...]] / end 2156 actions {action} [/ {action} [...]] / end 2157 2158If successful, it will return a flow rule ID usable with other commands:: 2159 2160 Flow rule #[...] created 2161 2162Otherwise it will show an error message of the form:: 2163 2164 Caught error type [...] ([...]): [...] 2165 2166Parameters describe in the following order: 2167 2168- Attributes (*group*, *priority*, *ingress*, *egress* tokens). 2169- A matching pattern, starting with the *pattern* token and terminated by an 2170 *end* pattern item. 2171- Actions, starting with the *actions* token and terminated by an *end* 2172 action. 2173 2174These translate directly to *rte_flow* objects provided as-is to the 2175underlying functions. 2176 2177The shortest valid definition only comprises mandatory tokens:: 2178 2179 testpmd> flow create 0 pattern end actions end 2180 2181Note that PMDs may refuse rules that essentially do nothing such as this 2182one. 2183 2184**All unspecified object values are automatically initialized to 0.** 2185 2186Attributes 2187^^^^^^^^^^ 2188 2189These tokens affect flow rule attributes (``struct rte_flow_attr``) and are 2190specified before the ``pattern`` token. 2191 2192- ``group {group id}``: priority group. 2193- ``priority {level}``: priority level within group. 2194- ``ingress``: rule applies to ingress traffic. 2195- ``egress``: rule applies to egress traffic. 2196 2197Each instance of an attribute specified several times overrides the previous 2198value as shown below (group 4 is used):: 2199 2200 testpmd> flow create 0 group 42 group 24 group 4 [...] 2201 2202Note that once enabled, ``ingress`` and ``egress`` cannot be disabled. 2203 2204While not specifying a direction is an error, some rules may allow both 2205simultaneously. 2206 2207Most rules affect RX therefore contain the ``ingress`` token:: 2208 2209 testpmd> flow create 0 ingress pattern [...] 2210 2211Matching pattern 2212^^^^^^^^^^^^^^^^ 2213 2214A matching pattern starts after the ``pattern`` token. It is made of pattern 2215items and is terminated by a mandatory ``end`` item. 2216 2217Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum 2218rte_flow_item_type``). 2219 2220The ``/`` token is used as a separator between pattern items as shown 2221below:: 2222 2223 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...] 2224 2225Note that protocol items like these must be stacked from lowest to highest 2226layer to make sense. For instance, the following rule is either invalid or 2227unlikely to match any packet:: 2228 2229 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...] 2230 2231More information on these restrictions can be found in the *rte_flow* 2232documentation. 2233 2234Several items support additional specification structures, for example 2235``ipv4`` allows specifying source and destination addresses as follows:: 2236 2237 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1 2238 dst is 10.2.0.0 / end [...] 2239 2240This rule matches all IPv4 traffic with the specified properties. 2241 2242In this example, ``src`` and ``dst`` are field names of the underlying 2243``struct rte_flow_item_ipv4`` object. All item properties can be specified 2244in a similar fashion. 2245 2246The ``is`` token means that the subsequent value must be matched exactly, 2247and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item`` 2248accordingly. Possible assignment tokens are: 2249 2250- ``is``: match value perfectly (with full bit-mask). 2251- ``spec``: match value according to configured bit-mask. 2252- ``last``: specify upper bound to establish a range. 2253- ``mask``: specify bit-mask with relevant bits set to one. 2254- ``prefix``: generate bit-mask from a prefix length. 2255 2256These yield identical results:: 2257 2258 ipv4 src is 10.1.1.1 2259 2260:: 2261 2262 ipv4 src spec 10.1.1.1 src mask 255.255.255.255 2263 2264:: 2265 2266 ipv4 src spec 10.1.1.1 src prefix 32 2267 2268:: 2269 2270 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value 2271 2272:: 2273 2274 ipv4 src is 10.1.1.1 src last 0 # 0 disables range 2275 2276Inclusive ranges can be defined with ``last``:: 2277 2278 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4 2279 2280Note that ``mask`` affects both ``spec`` and ``last``:: 2281 2282 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0 2283 # matches 10.1.0.0 to 10.2.255.255 2284 2285Properties can be modified multiple times:: 2286 2287 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4 2288 2289:: 2290 2291 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16 2292 2293Pattern items 2294^^^^^^^^^^^^^ 2295 2296This section lists supported pattern items and their attributes, if any. 2297 2298- ``end``: end list of pattern items. 2299 2300- ``void``: no-op pattern item. 2301 2302- ``invert``: perform actions when pattern does not match. 2303 2304- ``any``: match any protocol for the current layer. 2305 2306 - ``num {unsigned}``: number of layers covered. 2307 2308- ``pf``: match packets addressed to the physical function. 2309 2310- ``vf``: match packets addressed to a virtual function ID. 2311 2312 - ``id {unsigned}``: destination VF ID. 2313 2314- ``port``: device-specific physical port index to use. 2315 2316 - ``index {unsigned}``: physical port index. 2317 2318- ``raw``: match an arbitrary byte string. 2319 2320 - ``relative {boolean}``: look for pattern after the previous item. 2321 - ``search {boolean}``: search pattern from offset (see also limit). 2322 - ``offset {integer}``: absolute or relative offset for pattern. 2323 - ``limit {unsigned}``: search area limit for start of pattern. 2324 - ``pattern {string}``: byte string to look for. 2325 2326- ``eth``: match Ethernet header. 2327 2328 - ``dst {MAC-48}``: destination MAC. 2329 - ``src {MAC-48}``: source MAC. 2330 - ``type {unsigned}``: EtherType. 2331 2332- ``vlan``: match 802.1Q/ad VLAN tag. 2333 2334 - ``tpid {unsigned}``: tag protocol identifier. 2335 - ``tci {unsigned}``: tag control information. 2336 - ``pcp {unsigned}``: priority code point. 2337 - ``dei {unsigned}``: drop eligible indicator. 2338 - ``vid {unsigned}``: VLAN identifier. 2339 2340- ``ipv4``: match IPv4 header. 2341 2342 - ``tos {unsigned}``: type of service. 2343 - ``ttl {unsigned}``: time to live. 2344 - ``proto {unsigned}``: next protocol ID. 2345 - ``src {ipv4 address}``: source address. 2346 - ``dst {ipv4 address}``: destination address. 2347 2348- ``ipv6``: match IPv6 header. 2349 2350 - ``tc {unsigned}``: traffic class. 2351 - ``flow {unsigned}``: flow label. 2352 - ``proto {unsigned}``: protocol (next header). 2353 - ``hop {unsigned}``: hop limit. 2354 - ``src {ipv6 address}``: source address. 2355 - ``dst {ipv6 address}``: destination address. 2356 2357- ``icmp``: match ICMP header. 2358 2359 - ``type {unsigned}``: ICMP packet type. 2360 - ``code {unsigned}``: ICMP packet code. 2361 2362- ``udp``: match UDP header. 2363 2364 - ``src {unsigned}``: UDP source port. 2365 - ``dst {unsigned}``: UDP destination port. 2366 2367- ``tcp``: match TCP header. 2368 2369 - ``src {unsigned}``: TCP source port. 2370 - ``dst {unsigned}``: TCP destination port. 2371 2372- ``sctp``: match SCTP header. 2373 2374 - ``src {unsigned}``: SCTP source port. 2375 - ``dst {unsigned}``: SCTP destination port. 2376 - ``tag {unsigned}``: validation tag. 2377 - ``cksum {unsigned}``: checksum. 2378 2379- ``vxlan``: match VXLAN header. 2380 2381 - ``vni {unsigned}``: VXLAN identifier. 2382 2383Actions list 2384^^^^^^^^^^^^ 2385 2386A list of actions starts after the ``actions`` token in the same fashion as 2387`Matching pattern`_; actions are separated by ``/`` tokens and the list is 2388terminated by a mandatory ``end`` action. 2389 2390Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum 2391rte_flow_action_type``). 2392 2393Dropping all incoming UDPv4 packets can be expressed as follows:: 2394 2395 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2396 actions drop / end 2397 2398Several actions have configurable properties which must be specified when 2399there is no valid default value. For example, ``queue`` requires a target 2400queue index. 2401 2402This rule redirects incoming UDPv4 traffic to queue index 6:: 2403 2404 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2405 actions queue index 6 / end 2406 2407While this one could be rejected by PMDs (unspecified queue index):: 2408 2409 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2410 actions queue / end 2411 2412As defined by *rte_flow*, the list is not ordered, all actions of a given 2413rule are performed simultaneously. These are equivalent:: 2414 2415 queue index 6 / void / mark id 42 / end 2416 2417:: 2418 2419 void / mark id 42 / queue index 6 / end 2420 2421All actions in a list should have different types, otherwise only the last 2422action of a given type is taken into account:: 2423 2424 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6 2425 2426:: 2427 2428 drop / drop / drop / end # drop is performed only once 2429 2430:: 2431 2432 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24 2433 2434Considering they are performed simultaneously, opposite and overlapping 2435actions can sometimes be combined when the end result is unambiguous:: 2436 2437 drop / queue index 6 / end # drop has no effect 2438 2439:: 2440 2441 drop / dup index 6 / end # same as above 2442 2443:: 2444 2445 queue index 6 / rss queues 6 7 8 / end # queue has no effect 2446 2447:: 2448 2449 drop / passthru / end # drop has no effect 2450 2451Note that PMDs may still refuse such combinations. 2452 2453Actions 2454^^^^^^^ 2455 2456This section lists supported actions and their attributes, if any. 2457 2458- ``end``: end list of actions. 2459 2460- ``void``: no-op action. 2461 2462- ``passthru``: let subsequent rule process matched packets. 2463 2464- ``mark``: attach 32 bit value to packets. 2465 2466 - ``id {unsigned}``: 32 bit value to return with packets. 2467 2468- ``flag``: flag packets. 2469 2470- ``queue``: assign packets to a given queue index. 2471 2472 - ``index {unsigned}``: queue index to use. 2473 2474- ``drop``: drop packets (note: passthru has priority). 2475 2476- ``count``: enable counters for this rule. 2477 2478- ``dup``: duplicate packets to a given queue index. 2479 2480 - ``index {unsigned}``: queue index to duplicate packets to. 2481 2482- ``rss``: spread packets among several queues. 2483 2484 - ``queues [{unsigned} [...]] end``: queue indices to use. 2485 2486- ``pf``: redirect packets to physical device function. 2487 2488- ``vf``: redirect packets to virtual device function. 2489 2490 - ``original {boolean}``: use original VF ID if possible. 2491 - ``id {unsigned}``: VF ID to redirect packets to. 2492 2493Destroying flow rules 2494~~~~~~~~~~~~~~~~~~~~~ 2495 2496``flow destroy`` destroys one or more rules from their rule ID (as returned 2497by ``flow create``), this command calls ``rte_flow_destroy()`` as many 2498times as necessary:: 2499 2500 flow destroy {port_id} rule {rule_id} [...] 2501 2502If successful, it will show:: 2503 2504 Flow rule #[...] destroyed 2505 2506It does not report anything for rule IDs that do not exist. The usual error 2507message is shown when a rule cannot be destroyed:: 2508 2509 Caught error type [...] ([...]): [...] 2510 2511``flow flush`` destroys all rules on a device and does not take extra 2512arguments. It is bound to ``rte_flow_flush()``:: 2513 2514 flow flush {port_id} 2515 2516Any errors are reported as above. 2517 2518Creating several rules and destroying them:: 2519 2520 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2521 actions queue index 2 / end 2522 Flow rule #0 created 2523 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2524 actions queue index 3 / end 2525 Flow rule #1 created 2526 testpmd> flow destroy 0 rule 0 rule 1 2527 Flow rule #1 destroyed 2528 Flow rule #0 destroyed 2529 testpmd> 2530 2531The same result can be achieved using ``flow flush``:: 2532 2533 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2534 actions queue index 2 / end 2535 Flow rule #0 created 2536 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2537 actions queue index 3 / end 2538 Flow rule #1 created 2539 testpmd> flow flush 0 2540 testpmd> 2541 2542Non-existent rule IDs are ignored:: 2543 2544 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2545 actions queue index 2 / end 2546 Flow rule #0 created 2547 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2548 actions queue index 3 / end 2549 Flow rule #1 created 2550 testpmd> flow destroy 0 rule 42 rule 10 rule 2 2551 testpmd> 2552 testpmd> flow destroy 0 rule 0 2553 Flow rule #0 destroyed 2554 testpmd> 2555 2556Querying flow rules 2557~~~~~~~~~~~~~~~~~~~ 2558 2559``flow query`` queries a specific action of a flow rule having that 2560ability. Such actions collect information that can be reported using this 2561command. It is bound to ``rte_flow_query()``:: 2562 2563 flow query {port_id} {rule_id} {action} 2564 2565If successful, it will display either the retrieved data for known actions 2566or the following message:: 2567 2568 Cannot display result for action type [...] ([...]) 2569 2570Otherwise, it will complain either that the rule does not exist or that some 2571error occurred:: 2572 2573 Flow rule #[...] not found 2574 2575:: 2576 2577 Caught error type [...] ([...]): [...] 2578 2579Currently only the ``count`` action is supported. This action reports the 2580number of packets that hit the flow rule and the total number of bytes. Its 2581output has the following format:: 2582 2583 count: 2584 hits_set: [...] # whether "hits" contains a valid value 2585 bytes_set: [...] # whether "bytes" contains a valid value 2586 hits: [...] # number of packets 2587 bytes: [...] # number of bytes 2588 2589Querying counters for TCPv6 packets redirected to queue 6:: 2590 2591 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end 2592 actions queue index 6 / count / end 2593 Flow rule #4 created 2594 testpmd> flow query 0 4 count 2595 count: 2596 hits_set: 1 2597 bytes_set: 0 2598 hits: 386446 2599 bytes: 0 2600 testpmd> 2601 2602Listing flow rules 2603~~~~~~~~~~~~~~~~~~ 2604 2605``flow list`` lists existing flow rules sorted by priority and optionally 2606filtered by group identifiers:: 2607 2608 flow list {port_id} [group {group_id}] [...] 2609 2610This command only fails with the following message if the device does not 2611exist:: 2612 2613 Invalid port [...] 2614 2615Output consists of a header line followed by a short description of each 2616flow rule, one per line. There is no output at all when no flow rules are 2617configured on the device:: 2618 2619 ID Group Prio Attr Rule 2620 [...] [...] [...] [...] [...] 2621 2622``Attr`` column flags: 2623 2624- ``i`` for ``ingress``. 2625- ``e`` for ``egress``. 2626 2627Creating several flow rules and listing them:: 2628 2629 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2630 actions queue index 6 / end 2631 Flow rule #0 created 2632 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2633 actions queue index 2 / end 2634 Flow rule #1 created 2635 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end 2636 actions rss queues 6 7 8 end / end 2637 Flow rule #2 created 2638 testpmd> flow list 0 2639 ID Group Prio Attr Rule 2640 0 0 0 i- ETH IPV4 => QUEUE 2641 1 0 0 i- ETH IPV6 => QUEUE 2642 2 0 5 i- ETH IPV4 UDP => RSS 2643 testpmd> 2644 2645Rules are sorted by priority (i.e. group ID first, then priority level):: 2646 2647 testpmd> flow list 1 2648 ID Group Prio Attr Rule 2649 0 0 0 i- ETH => COUNT 2650 6 0 500 i- ETH IPV6 TCP => DROP COUNT 2651 5 0 1000 i- ETH IPV6 ICMP => QUEUE 2652 1 24 0 i- ETH IPV4 UDP => QUEUE 2653 4 24 10 i- ETH IPV4 TCP => DROP 2654 3 24 20 i- ETH IPV4 => DROP 2655 2 24 42 i- ETH IPV4 UDP => QUEUE 2656 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE 2657 testpmd> 2658 2659Output can be limited to specific groups:: 2660 2661 testpmd> flow list 1 group 0 group 63 2662 ID Group Prio Attr Rule 2663 0 0 0 i- ETH => COUNT 2664 6 0 500 i- ETH IPV6 TCP => DROP COUNT 2665 5 0 1000 i- ETH IPV6 ICMP => QUEUE 2666 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE 2667 testpmd> 2668