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 89Command File Functions 90---------------------- 91 92To facilitate loading large number of commands or to avoid cutting and pasting where not 93practical or possible testpmd supports alternative methods for executing commands. 94 95* If started with the ``--cmdline-file=FILENAME`` command line argument testpmd 96 will execute all CLI commands contained within the file immediately before 97 starting packet forwarding or entering interactive mode. 98 99.. code-block:: console 100 101 ./testpmd -n4 -r2 ... -- -i --cmdline-file=/home/ubuntu/flow-create-commands.txt 102 Interactive-mode selected 103 CLI commands to be read from /home/ubuntu/flow-create-commands.txt 104 Configuring Port 0 (socket 0) 105 Port 0: 7C:FE:90:CB:74:CE 106 Configuring Port 1 (socket 0) 107 Port 1: 7C:FE:90:CB:74:CA 108 Checking link statuses... 109 Port 0 Link Up - speed 10000 Mbps - full-duplex 110 Port 1 Link Up - speed 10000 Mbps - full-duplex 111 Done 112 Flow rule #0 created 113 Flow rule #1 created 114 ... 115 ... 116 Flow rule #498 created 117 Flow rule #499 created 118 Read all CLI commands from /home/ubuntu/flow-create-commands.txt 119 testpmd> 120 121 122* At run-time additional commands can be loaded in bulk by invoking the ``load FILENAME`` 123 command. 124 125.. code-block:: console 126 127 testpmd> load /home/ubuntu/flow-create-commands.txt 128 Flow rule #0 created 129 Flow rule #1 created 130 ... 131 ... 132 Flow rule #498 created 133 Flow rule #499 created 134 Read all CLI commands from /home/ubuntu/flow-create-commands.txt 135 testpmd> 136 137 138In all cases output from any included command will be displayed as standard output. 139Execution will continue until the end of the file is reached regardless of 140whether any errors occur. The end user must examine the output to determine if 141any failures occurred. 142 143 144Control Functions 145----------------- 146 147start 148~~~~~ 149 150Start packet forwarding with current configuration:: 151 152 testpmd> start 153 154start tx_first 155~~~~~~~~~~~~~~ 156 157Start packet forwarding with current configuration after sending specified number of bursts of packets:: 158 159 testpmd> start tx_first (""|burst_num) 160 161The default burst number is 1 when ``burst_num`` not presented. 162 163stop 164~~~~ 165 166Stop packet forwarding, and display accumulated statistics:: 167 168 testpmd> stop 169 170quit 171~~~~ 172 173Quit to prompt:: 174 175 testpmd> quit 176 177 178Display Functions 179----------------- 180 181The functions in the following sections are used to display information about the 182testpmd configuration or the NIC status. 183 184show port 185~~~~~~~~~ 186 187Display information for a given port or all ports:: 188 189 testpmd> show port (info|stats|xstats|fdir|stat_qmap|dcb_tc|cap) (port_id|all) 190 191The available information categories are: 192 193* ``info``: General port information such as MAC address. 194 195* ``stats``: RX/TX statistics. 196 197* ``xstats``: RX/TX extended NIC statistics. 198 199* ``fdir``: Flow Director information and statistics. 200 201* ``stat_qmap``: Queue statistics mapping. 202 203* ``dcb_tc``: DCB information such as TC mapping. 204 205* ``cap``: Supported offload capabilities. 206 207For example: 208 209.. code-block:: console 210 211 testpmd> show port info 0 212 213 ********************* Infos for port 0 ********************* 214 215 MAC address: XX:XX:XX:XX:XX:XX 216 Connect to socket: 0 217 memory allocation on the socket: 0 218 Link status: up 219 Link speed: 40000 Mbps 220 Link duplex: full-duplex 221 Promiscuous mode: enabled 222 Allmulticast mode: disabled 223 Maximum number of MAC addresses: 64 224 Maximum number of MAC addresses of hash filtering: 0 225 VLAN offload: 226 strip on 227 filter on 228 qinq(extend) off 229 Redirection table size: 512 230 Supported flow types: 231 ipv4-frag 232 ipv4-tcp 233 ipv4-udp 234 ipv4-sctp 235 ipv4-other 236 ipv6-frag 237 ipv6-tcp 238 ipv6-udp 239 ipv6-sctp 240 ipv6-other 241 l2_payload 242 port 243 vxlan 244 geneve 245 nvgre 246 247show port rss reta 248~~~~~~~~~~~~~~~~~~ 249 250Display the rss redirection table entry indicated by masks on port X:: 251 252 testpmd> show port (port_id) rss reta (size) (mask0, mask1...) 253 254size is used to indicate the hardware supported reta size 255 256show port rss-hash 257~~~~~~~~~~~~~~~~~~ 258 259Display the RSS hash functions and RSS hash key of a port:: 260 261 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] 262 263clear port 264~~~~~~~~~~ 265 266Clear the port statistics for a given port or for all ports:: 267 268 testpmd> clear port (info|stats|xstats|fdir|stat_qmap) (port_id|all) 269 270For example:: 271 272 testpmd> clear port stats all 273 274show (rxq|txq) 275~~~~~~~~~~~~~~ 276 277Display information for a given port's RX/TX queue:: 278 279 testpmd> show (rxq|txq) info (port_id) (queue_id) 280 281show config 282~~~~~~~~~~~ 283 284Displays the configuration of the application. 285The configuration comes from the command-line, the runtime or the application defaults:: 286 287 testpmd> show config (rxtx|cores|fwd|txpkts) 288 289The available information categories are: 290 291* ``rxtx``: RX/TX configuration items. 292 293* ``cores``: List of forwarding cores. 294 295* ``fwd``: Packet forwarding configuration. 296 297* ``txpkts``: Packets to TX configuration. 298 299For example: 300 301.. code-block:: console 302 303 testpmd> show config rxtx 304 305 io packet forwarding - CRC stripping disabled - packets/burst=16 306 nb forwarding cores=2 - nb forwarding ports=1 307 RX queues=1 - RX desc=128 - RX free threshold=0 308 RX threshold registers: pthresh=8 hthresh=8 wthresh=4 309 TX queues=1 - TX desc=512 - TX free threshold=0 310 TX threshold registers: pthresh=36 hthresh=0 wthresh=0 311 TX RS bit threshold=0 - TXQ flags=0x0 312 313set fwd 314~~~~~~~ 315 316Set the packet forwarding mode:: 317 318 testpmd> set fwd (io|mac|macswap|flowgen| \ 319 rxonly|txonly|csum|icmpecho) (""|retry) 320 321``retry`` can be specified for forwarding engines except ``rx_only``. 322 323The available information categories are: 324 325* ``io``: Forwards packets "as-is" in I/O mode. 326 This is the fastest possible forwarding operation as it does not access packets data. 327 This is the default mode. 328 329* ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them. 330 Default application behaviour is to set source Ethernet address to that of the transmitting interface, and destination 331 address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or 332 'eth-peer-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address. 333 334* ``macswap``: MAC swap forwarding mode. 335 Swaps the source and the destination Ethernet addresses of packets before forwarding them. 336 337* ``flowgen``: Multi-flow generation mode. 338 Originates a number of flows (with varying destination IP addresses), and terminate receive traffic. 339 340* ``rxonly``: Receives packets but doesn't transmit them. 341 342* ``txonly``: Generates and transmits packets without receiving any. 343 344* ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet. 345 346* ``icmpecho``: Receives a burst of packets, lookup for IMCP echo requests and, if any, send back ICMP echo replies. 347 348* ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX. Requires ``CONFIG_RTE_LIBRTE_IEEE1588=y``. 349 350Note: TX timestamping is only available in the "Full Featured" TX path. To force ``testpmd`` into this mode set ``--txqflags=0``. 351 352Example:: 353 354 testpmd> set fwd rxonly 355 356 Set rxonly packet forwarding mode 357 358 359read rxd 360~~~~~~~~ 361 362Display an RX descriptor for a port RX queue:: 363 364 testpmd> read rxd (port_id) (queue_id) (rxd_id) 365 366For example:: 367 368 testpmd> read rxd 0 0 4 369 0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180 370 371read txd 372~~~~~~~~ 373 374Display a TX descriptor for a port TX queue:: 375 376 testpmd> read txd (port_id) (queue_id) (txd_id) 377 378For example:: 379 380 testpmd> read txd 0 0 4 381 0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C 382 383ddp get list 384~~~~~~~~~~~~ 385 386Get loaded dynamic device personalization (DDP) package info list:: 387 388 testpmd> ddp get list (port_id) 389 390show vf stats 391~~~~~~~~~~~~~ 392 393Display VF statistics:: 394 395 testpmd> show vf stats (port_id) (vf_id) 396 397clear vf stats 398~~~~~~~~~~~~~~ 399 400Reset VF statistics:: 401 402 testpmd> clear vf stats (port_id) (vf_id) 403 404Configuration Functions 405----------------------- 406 407The testpmd application can be configured from the runtime as well as from the command-line. 408 409This section details the available configuration functions that are available. 410 411.. note:: 412 413 Configuration changes only become active when forwarding is started/restarted. 414 415set default 416~~~~~~~~~~~ 417 418Reset forwarding to the default configuration:: 419 420 testpmd> set default 421 422set verbose 423~~~~~~~~~~~ 424 425Set the debug verbosity level:: 426 427 testpmd> set verbose (level) 428 429Currently the only available levels are 0 (silent except for error) and 1 (fully verbose). 430 431set nbport 432~~~~~~~~~~ 433 434Set the number of ports used by the application: 435 436set nbport (num) 437 438This is equivalent to the ``--nb-ports`` command-line option. 439 440set nbcore 441~~~~~~~~~~ 442 443Set the number of cores used by the application:: 444 445 testpmd> set nbcore (num) 446 447This is equivalent to the ``--nb-cores`` command-line option. 448 449.. note:: 450 451 The number of cores used must not be greater than number of ports used multiplied by the number of queues per port. 452 453set coremask 454~~~~~~~~~~~~ 455 456Set the forwarding cores hexadecimal mask:: 457 458 testpmd> set coremask (mask) 459 460This is equivalent to the ``--coremask`` command-line option. 461 462.. note:: 463 464 The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding. 465 466set portmask 467~~~~~~~~~~~~ 468 469Set the forwarding ports hexadecimal mask:: 470 471 testpmd> set portmask (mask) 472 473This is equivalent to the ``--portmask`` command-line option. 474 475set burst 476~~~~~~~~~ 477 478Set number of packets per burst:: 479 480 testpmd> set burst (num) 481 482This is equivalent to the ``--burst command-line`` option. 483 484When retry is enabled, the transmit delay time and number of retries can also be set:: 485 486 testpmd> set burst tx delay (microseconds) retry (num) 487 488set txpkts 489~~~~~~~~~~ 490 491Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode:: 492 493 testpmd> set txpkts (x[,y]*) 494 495Where x[,y]* represents a CSV list of values, without white space. 496 497set txsplit 498~~~~~~~~~~~ 499 500Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes:: 501 502 testpmd> set txsplit (off|on|rand) 503 504Where: 505 506* ``off`` disable packet copy & split for CSUM mode. 507 508* ``on`` split outgoing packet into multiple segments. Size of each segment 509 and number of segments per packet is determined by ``set txpkts`` command 510 (see above). 511 512* ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments. 513 514set corelist 515~~~~~~~~~~~~ 516 517Set the list of forwarding cores:: 518 519 testpmd> set corelist (x[,y]*) 520 521For example, to change the forwarding cores: 522 523.. code-block:: console 524 525 testpmd> set corelist 3,1 526 testpmd> show config fwd 527 528 io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled 529 Logical Core 3 (socket 0) forwards packets on 1 streams: 530 RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01 531 Logical Core 1 (socket 0) forwards packets on 1 streams: 532 RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00 533 534.. note:: 535 536 The cores are used in the same order as specified on the command line. 537 538set portlist 539~~~~~~~~~~~~ 540 541Set the list of forwarding ports:: 542 543 testpmd> set portlist (x[,y]*) 544 545For example, to change the port forwarding: 546 547.. code-block:: console 548 549 testpmd> set portlist 0,2,1,3 550 testpmd> show config fwd 551 552 io packet forwarding - ports=4 - cores=1 - streams=4 553 Logical Core 3 (socket 0) forwards packets on 4 streams: 554 RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01 555 RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00 556 RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03 557 RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02 558 559set tx loopback 560~~~~~~~~~~~~~~~ 561 562Enable/disable tx loopback:: 563 564 testpmd> set tx loopback (port_id) (on|off) 565 566set drop enable 567~~~~~~~~~~~~~~~ 568 569set drop enable bit for all queues:: 570 571 testpmd> set all queues drop (port_id) (on|off) 572 573set split drop enable (for VF) 574~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 575 576set split drop enable bit for VF from PF:: 577 578 testpmd> set vf split drop (port_id) (vf_id) (on|off) 579 580set mac antispoof (for VF) 581~~~~~~~~~~~~~~~~~~~~~~~~~~ 582 583Set mac antispoof for a VF from the PF:: 584 585 testpmd> set vf mac antispoof (port_id) (vf_id) (on|off) 586 587set macsec offload 588~~~~~~~~~~~~~~~~~~ 589 590Enable/disable MACsec offload:: 591 592 testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off) 593 testpmd> set macsec offload (port_id) off 594 595set macsec sc 596~~~~~~~~~~~~~ 597 598Configure MACsec secure connection (SC):: 599 600 testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi) 601 602.. note:: 603 604 The pi argument is ignored for tx. 605 Check the NIC Datasheet for hardware limits. 606 607set macsec sa 608~~~~~~~~~~~~~ 609 610Configure MACsec secure association (SA):: 611 612 testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key) 613 614.. note:: 615 616 The IDX value must be 0 or 1. 617 Check the NIC Datasheet for hardware limits. 618 619set broadcast mode (for VF) 620~~~~~~~~~~~~~~~~~~~~~~~~~~~ 621 622Set broadcast mode for a VF from the PF:: 623 624 testpmd> set vf broadcast (port_id) (vf_id) (on|off) 625 626vlan set strip 627~~~~~~~~~~~~~~ 628 629Set the VLAN strip on a port:: 630 631 testpmd> vlan set strip (on|off) (port_id) 632 633vlan set stripq 634~~~~~~~~~~~~~~~ 635 636Set the VLAN strip for a queue on a port:: 637 638 testpmd> vlan set stripq (on|off) (port_id,queue_id) 639 640vlan set stripq (for VF) 641~~~~~~~~~~~~~~~~~~~~~~~~ 642 643Set VLAN strip for all queues in a pool for a VF from the PF:: 644 645 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off) 646 647vlan set insert (for VF) 648~~~~~~~~~~~~~~~~~~~~~~~~ 649 650Set VLAN insert for a VF from the PF:: 651 652 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id) 653 654vlan set tag (for VF) 655~~~~~~~~~~~~~~~~~~~~~ 656 657Set VLAN tag for a VF from the PF:: 658 659 testpmd> set vf vlan tag (port_id) (vf_id) (on|off) 660 661vlan set antispoof (for VF) 662~~~~~~~~~~~~~~~~~~~~~~~~~~~ 663 664Set VLAN antispoof for a VF from the PF:: 665 666 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off) 667 668vlan set filter 669~~~~~~~~~~~~~~~ 670 671Set the VLAN filter on a port:: 672 673 testpmd> vlan set filter (on|off) (port_id) 674 675vlan set qinq 676~~~~~~~~~~~~~ 677 678Set the VLAN QinQ (extended queue in queue) on for a port:: 679 680 testpmd> vlan set qinq (on|off) (port_id) 681 682vlan set tpid 683~~~~~~~~~~~~~ 684 685Set the inner or outer VLAN TPID for packet filtering on a port:: 686 687 testpmd> vlan set (inner|outer) tpid (value) (port_id) 688 689.. note:: 690 691 TPID value must be a 16-bit number (value <= 65536). 692 693rx_vlan add 694~~~~~~~~~~~ 695 696Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID:: 697 698 testpmd> rx_vlan add (vlan_id|all) (port_id) 699 700.. note:: 701 702 VLAN filter must be set on that port. VLAN ID < 4096. 703 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries 704 in VFTA table. This is important if enabling all vlan_ids. 705 706rx_vlan rm 707~~~~~~~~~~ 708 709Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID:: 710 711 testpmd> rx_vlan rm (vlan_id|all) (port_id) 712 713rx_vlan add (for VF) 714~~~~~~~~~~~~~~~~~~~~ 715 716Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID:: 717 718 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask) 719 720rx_vlan rm (for VF) 721~~~~~~~~~~~~~~~~~~~ 722 723Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID:: 724 725 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask) 726 727tunnel_filter add 728~~~~~~~~~~~~~~~~~ 729 730Add a tunnel filter on a port:: 731 732 testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \ 733 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\ 734 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id) 735 736The available information categories are: 737 738* ``vxlan``: Set tunnel type as VXLAN. 739 740* ``nvgre``: Set tunnel type as NVGRE. 741 742* ``ipingre``: Set tunnel type as IP-in-GRE. 743 744* ``imac-ivlan``: Set filter type as Inner MAC and VLAN. 745 746* ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID. 747 748* ``imac-tenid``: Set filter type as Inner MAC and tenant ID. 749 750* ``imac``: Set filter type as Inner MAC. 751 752* ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID. 753 754* ``oip``: Set filter type as Outer IP. 755 756* ``iip``: Set filter type as Inner IP. 757 758Example:: 759 760 testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \ 761 192.168.2.2 0 ipingre oip 1 1 762 763 Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP. 764 765tunnel_filter remove 766~~~~~~~~~~~~~~~~~~~~ 767 768Remove a tunnel filter on a port:: 769 770 testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \ 771 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\ 772 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id) 773 774rx_vxlan_port add 775~~~~~~~~~~~~~~~~~ 776 777Add an UDP port for VXLAN packet filter on a port:: 778 779 testpmd> rx_vxlan_port add (udp_port) (port_id) 780 781rx_vxlan_port remove 782~~~~~~~~~~~~~~~~~~~~ 783 784Remove an UDP port for VXLAN packet filter on a port:: 785 786 testpmd> rx_vxlan_port rm (udp_port) (port_id) 787 788tx_vlan set 789~~~~~~~~~~~ 790 791Set hardware insertion of VLAN IDs in packets sent on a port:: 792 793 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer] 794 795For example, set a single VLAN ID (5) insertion on port 0:: 796 797 tx_vlan set 0 5 798 799Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1:: 800 801 tx_vlan set 1 2 3 802 803 804tx_vlan set pvid 805~~~~~~~~~~~~~~~~ 806 807Set port based hardware insertion of VLAN ID in packets sent on a port:: 808 809 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off) 810 811tx_vlan reset 812~~~~~~~~~~~~~ 813 814Disable hardware insertion of a VLAN header in packets sent on a port:: 815 816 testpmd> tx_vlan reset (port_id) 817 818csum set 819~~~~~~~~ 820 821Select hardware or software calculation of the checksum when 822transmitting a packet using the ``csum`` forwarding engine:: 823 824 testpmd> csum set (ip|udp|tcp|sctp|outer-ip) (hw|sw) (port_id) 825 826Where: 827 828* ``ip|udp|tcp|sctp`` always relate to the inner layer. 829 830* ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized 831 as a tunnel packet by the forwarding engine (vxlan, gre and ipip are 832 supported). See also the ``csum parse-tunnel`` command. 833 834.. note:: 835 836 Check the NIC Datasheet for hardware limits. 837 838csum parse-tunnel 839~~~~~~~~~~~~~~~~~ 840 841Define how tunneled packets should be handled by the csum forward 842engine:: 843 844 testpmd> csum parse-tunnel (on|off) (tx_port_id) 845 846If enabled, the csum forward engine will try to recognize supported 847tunnel headers (vxlan, gre, ipip). 848 849If disabled, treat tunnel packets as non-tunneled packets (a inner 850header is handled as a packet payload). 851 852.. note:: 853 854 The port argument is the TX port like in the ``csum set`` command. 855 856Example: 857 858Consider a packet in packet like the following:: 859 860 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in 861 862* If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set`` 863 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the 864 ``outer-ip parameter`` relates to the outer headers (here ``ipv4_out``). 865 866* If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set`` 867 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``. 868 869csum show 870~~~~~~~~~ 871 872Display tx checksum offload configuration:: 873 874 testpmd> csum show (port_id) 875 876tso set 877~~~~~~~ 878 879Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine:: 880 881 testpmd> tso set (segsize) (port_id) 882 883.. note:: 884 885 Check the NIC datasheet for hardware limits. 886 887tso show 888~~~~~~~~ 889 890Display the status of TCP Segmentation Offload:: 891 892 testpmd> tso show (port_id) 893 894mac_addr add 895~~~~~~~~~~~~ 896 897Add an alternative MAC address to a port:: 898 899 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX) 900 901mac_addr remove 902~~~~~~~~~~~~~~~ 903 904Remove a MAC address from a port:: 905 906 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX) 907 908mac_addr add (for VF) 909~~~~~~~~~~~~~~~~~~~~~ 910 911Add an alternative MAC address for a VF to a port:: 912 913 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX) 914 915mac_addr set 916~~~~~~~~~~~~ 917 918Set the default MAC address for a port:: 919 920 testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX) 921 922mac_addr set (for VF) 923~~~~~~~~~~~~~~~~~~~~~ 924 925Set the MAC address for a VF from the PF:: 926 927 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX) 928 929set port-uta 930~~~~~~~~~~~~ 931 932Set the unicast hash filter(s) on/off for a port:: 933 934 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off) 935 936set promisc 937~~~~~~~~~~~ 938 939Set the promiscuous mode on for a port or for all ports. 940In promiscuous mode packets are not dropped if they aren't for the specified MAC address:: 941 942 testpmd> set promisc (port_id|all) (on|off) 943 944set allmulti 945~~~~~~~~~~~~ 946 947Set the allmulti mode for a port or for all ports:: 948 949 testpmd> set allmulti (port_id|all) (on|off) 950 951Same as the ifconfig (8) option. Controls how multicast packets are handled. 952 953set promisc (for VF) 954~~~~~~~~~~~~~~~~~~~~ 955 956Set the unicast promiscuous mode for a VF from PF. 957It's supported by Intel i40e NICs now. 958In promiscuous mode packets are not dropped if they aren't for the specified MAC address:: 959 960 testpmd> set vf promisc (port_id) (vf_id) (on|off) 961 962set allmulticast (for VF) 963~~~~~~~~~~~~~~~~~~~~~~~~~ 964 965Set the multicast promiscuous mode for a VF from PF. 966It's supported by Intel i40e NICs now. 967In promiscuous mode packets are not dropped if they aren't for the specified MAC address:: 968 969 testpmd> set vf allmulti (port_id) (vf_id) (on|off) 970 971set tx max bandwidth (for VF) 972~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 973 974Set TX max absolute bandwidth (Mbps) for a VF from PF:: 975 976 testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth) 977 978set tc tx min bandwidth (for VF) 979~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 980 981Set all TCs' TX min relative bandwidth (%) for a VF from PF:: 982 983 testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...) 984 985set tc tx max bandwidth (for VF) 986~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 987 988Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF:: 989 990 testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth) 991 992set tc strict link priority mode 993~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 994 995Set some TCs' strict link priority mode on a physical port:: 996 997 testpmd> set tx strict-link-priority (port_id) (tc_bitmap) 998 999set tc tx min bandwidth 1000~~~~~~~~~~~~~~~~~~~~~~~ 1001 1002Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs:: 1003 1004 testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...) 1005 1006set flow_ctrl rx 1007~~~~~~~~~~~~~~~~ 1008 1009Set the link flow control parameter on a port:: 1010 1011 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \ 1012 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \ 1013 autoneg (on|off) (port_id) 1014 1015Where: 1016 1017* ``high_water`` (integer): High threshold value to trigger XOFF. 1018 1019* ``low_water`` (integer): Low threshold value to trigger XON. 1020 1021* ``pause_time`` (integer): Pause quota in the Pause frame. 1022 1023* ``send_xon`` (0/1): Send XON frame. 1024 1025* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames. 1026 1027* ``autoneg``: Change the auto-negotiation parameter. 1028 1029set pfc_ctrl rx 1030~~~~~~~~~~~~~~~ 1031 1032Set the priority flow control parameter on a port:: 1033 1034 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \ 1035 (pause_time) (priority) (port_id) 1036 1037Where: 1038 1039* ``high_water`` (integer): High threshold value. 1040 1041* ``low_water`` (integer): Low threshold value. 1042 1043* ``pause_time`` (integer): Pause quota in the Pause frame. 1044 1045* ``priority`` (0-7): VLAN User Priority. 1046 1047set stat_qmap 1048~~~~~~~~~~~~~ 1049 1050Set statistics mapping (qmapping 0..15) for RX/TX queue on port:: 1051 1052 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping) 1053 1054For example, to set rx queue 2 on port 0 to mapping 5:: 1055 1056 testpmd>set stat_qmap rx 0 2 5 1057 1058set port - rx/tx (for VF) 1059~~~~~~~~~~~~~~~~~~~~~~~~~ 1060 1061Set VF receive/transmit from a port:: 1062 1063 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off) 1064 1065set port - mac address filter (for VF) 1066~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1067 1068Add/Remove unicast or multicast MAC addr filter for a VF:: 1069 1070 testpmd> set port (port_id) vf (vf_id) (mac_addr) \ 1071 (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off) 1072 1073set port - rx mode(for VF) 1074~~~~~~~~~~~~~~~~~~~~~~~~~~ 1075 1076Set the VF receive mode of a port:: 1077 1078 testpmd> set port (port_id) vf (vf_id) \ 1079 rxmode (AUPE|ROPE|BAM|MPE) (on|off) 1080 1081The available receive modes are: 1082 1083* ``AUPE``: Accepts untagged VLAN. 1084 1085* ``ROPE``: Accepts unicast hash. 1086 1087* ``BAM``: Accepts broadcast packets. 1088 1089* ``MPE``: Accepts all multicast packets. 1090 1091set port - tx_rate (for Queue) 1092~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1093 1094Set TX rate limitation for a queue on a port:: 1095 1096 testpmd> set port (port_id) queue (queue_id) rate (rate_value) 1097 1098set port - tx_rate (for VF) 1099~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1100 1101Set TX rate limitation for queues in VF on a port:: 1102 1103 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask) 1104 1105set port - mirror rule 1106~~~~~~~~~~~~~~~~~~~~~~ 1107 1108Set pool or vlan type mirror rule for a port:: 1109 1110 testpmd> set port (port_id) mirror-rule (rule_id) \ 1111 (pool-mirror-up|pool-mirror-down|vlan-mirror) \ 1112 (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off) 1113 1114Set link mirror rule for a port:: 1115 1116 testpmd> set port (port_id) mirror-rule (rule_id) \ 1117 (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off) 1118 1119For example to enable mirror traffic with vlan 0,1 to pool 0:: 1120 1121 set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on 1122 1123reset port - mirror rule 1124~~~~~~~~~~~~~~~~~~~~~~~~ 1125 1126Reset a mirror rule for a port:: 1127 1128 testpmd> reset port (port_id) mirror-rule (rule_id) 1129 1130set flush_rx 1131~~~~~~~~~~~~ 1132 1133Set the flush on RX streams before forwarding. 1134The default is flush ``on``. 1135Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams:: 1136 1137 testpmd> set flush_rx off 1138 1139set bypass mode 1140~~~~~~~~~~~~~~~ 1141 1142Set the bypass mode for the lowest port on bypass enabled NIC:: 1143 1144 testpmd> set bypass mode (normal|bypass|isolate) (port_id) 1145 1146set bypass event 1147~~~~~~~~~~~~~~~~ 1148 1149Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled:: 1150 1151 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \ 1152 mode (normal|bypass|isolate) (port_id) 1153 1154Where: 1155 1156* ``timeout``: Enable bypass after watchdog timeout. 1157 1158* ``os_on``: Enable bypass when OS/board is powered on. 1159 1160* ``os_off``: Enable bypass when OS/board is powered off. 1161 1162* ``power_on``: Enable bypass when power supply is turned on. 1163 1164* ``power_off``: Enable bypass when power supply is turned off. 1165 1166 1167set bypass timeout 1168~~~~~~~~~~~~~~~~~~ 1169 1170Set the bypass watchdog timeout to ``n`` seconds where 0 = instant:: 1171 1172 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32) 1173 1174show bypass config 1175~~~~~~~~~~~~~~~~~~ 1176 1177Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC:: 1178 1179 testpmd> show bypass config (port_id) 1180 1181set link up 1182~~~~~~~~~~~ 1183 1184Set link up for a port:: 1185 1186 testpmd> set link-up port (port id) 1187 1188set link down 1189~~~~~~~~~~~~~ 1190 1191Set link down for a port:: 1192 1193 testpmd> set link-down port (port id) 1194 1195E-tag set 1196~~~~~~~~~ 1197 1198Enable E-tag insertion for a VF on a port:: 1199 1200 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id) 1201 1202Disable E-tag insertion for a VF on a port:: 1203 1204 testpmd> E-tag set insertion off port (port_id) vf (vf_id) 1205 1206Enable/disable E-tag stripping on a port:: 1207 1208 testpmd> E-tag set stripping (on|off) port (port_id) 1209 1210Enable/disable E-tag based forwarding on a port:: 1211 1212 testpmd> E-tag set forwarding (on|off) port (port_id) 1213 1214Add an E-tag forwarding filter on a port:: 1215 1216 testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id) 1217 1218Delete an E-tag forwarding filter on a port:: 1219 testpmd> E-tag set filter del e-tag-id (value) port (port_id) 1220 1221ddp add 1222~~~~~~~ 1223 1224Load a dynamic device personalization (DDP) package:: 1225 1226 testpmd> ddp add (port_id) (package_path) 1227 1228ptype mapping 1229~~~~~~~~~~~~~ 1230 1231List all items from the ptype mapping table:: 1232 1233 testpmd> ptype mapping get (port_id) (valid_only) 1234 1235Where: 1236 1237* ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0). 1238 1239Replace a specific or a group of software defined ptype with a new one:: 1240 1241 testpmd> ptype mapping replace (port_id) (target) (mask) (pkt_type) 1242 1243where: 1244 1245* ``target``: A specific software ptype or a mask to represent a group of software ptypes. 1246 1247* ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1). 1248 1249* ``pkt_type``: The new software ptype to replace the old ones. 1250 1251Update hardware defined ptype to software defined packet type mapping table:: 1252 1253 testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype) 1254 1255where: 1256 1257* ``hw_ptype``: hardware ptype as the index of the ptype mapping table. 1258 1259* ``sw_ptype``: software ptype as the value of the ptype mapping table. 1260 1261Reset ptype mapping table:: 1262 1263 testpmd> ptype mapping reset (port_id) 1264 1265Port Functions 1266-------------- 1267 1268The following sections show functions for configuring ports. 1269 1270.. note:: 1271 1272 Port configuration changes only become active when forwarding is started/restarted. 1273 1274port attach 1275~~~~~~~~~~~ 1276 1277Attach a port specified by pci address or virtual device args:: 1278 1279 testpmd> port attach (identifier) 1280 1281To attach a new pci device, the device should be recognized by kernel first. 1282Then it should be moved under DPDK management. 1283Finally the port can be attached to testpmd. 1284 1285For example, to move a pci device using ixgbe under DPDK management: 1286 1287.. code-block:: console 1288 1289 # Check the status of the available devices. 1290 ./usertools/dpdk-devbind.py --status 1291 1292 Network devices using DPDK-compatible driver 1293 ============================================ 1294 <none> 1295 1296 Network devices using kernel driver 1297 =================================== 1298 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused= 1299 1300 1301 # Bind the device to igb_uio. 1302 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0 1303 1304 1305 # Recheck the status of the devices. 1306 ./usertools/dpdk-devbind.py --status 1307 Network devices using DPDK-compatible driver 1308 ============================================ 1309 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused= 1310 1311To attach a port created by virtual device, above steps are not needed. 1312 1313For example, to attach a port whose pci address is 0000:0a:00.0. 1314 1315.. code-block:: console 1316 1317 testpmd> port attach 0000:0a:00.0 1318 Attaching a new port... 1319 EAL: PCI device 0000:0a:00.0 on NUMA socket -1 1320 EAL: probe driver: 8086:10fb rte_ixgbe_pmd 1321 EAL: PCI memory mapped at 0x7f83bfa00000 1322 EAL: PCI memory mapped at 0x7f83bfa80000 1323 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5 1324 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb 1325 Port 0 is attached. Now total ports is 1 1326 Done 1327 1328For example, to attach a port created by pcap PMD. 1329 1330.. code-block:: console 1331 1332 testpmd> port attach net_pcap0 1333 Attaching a new port... 1334 PMD: Initializing pmd_pcap for net_pcap0 1335 PMD: Creating pcap-backed ethdev on numa socket 0 1336 Port 0 is attached. Now total ports is 1 1337 Done 1338 1339In this case, identifier is ``net_pcap0``. 1340This identifier format is the same as ``--vdev`` format of DPDK applications. 1341 1342For example, to re-attach a bonded port which has been previously detached, 1343the mode and slave parameters must be given. 1344 1345.. code-block:: console 1346 1347 testpmd> port attach net_bond_0,mode=0,slave=1 1348 Attaching a new port... 1349 EAL: Initializing pmd_bond for net_bond_0 1350 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0. 1351 Port 0 is attached. Now total ports is 1 1352 Done 1353 1354 1355port detach 1356~~~~~~~~~~~ 1357 1358Detach a specific port:: 1359 1360 testpmd> port detach (port_id) 1361 1362Before detaching a port, the port should be stopped and closed. 1363 1364For example, to detach a pci device port 0. 1365 1366.. code-block:: console 1367 1368 testpmd> port stop 0 1369 Stopping ports... 1370 Done 1371 testpmd> port close 0 1372 Closing ports... 1373 Done 1374 1375 testpmd> port detach 0 1376 Detaching a port... 1377 EAL: PCI device 0000:0a:00.0 on NUMA socket -1 1378 EAL: remove driver: 8086:10fb rte_ixgbe_pmd 1379 EAL: PCI memory unmapped at 0x7f83bfa00000 1380 EAL: PCI memory unmapped at 0x7f83bfa80000 1381 Done 1382 1383 1384For example, to detach a virtual device port 0. 1385 1386.. code-block:: console 1387 1388 testpmd> port stop 0 1389 Stopping ports... 1390 Done 1391 testpmd> port close 0 1392 Closing ports... 1393 Done 1394 1395 testpmd> port detach 0 1396 Detaching a port... 1397 PMD: Closing pcap ethdev on numa socket 0 1398 Port 'net_pcap0' is detached. Now total ports is 0 1399 Done 1400 1401To remove a pci device completely from the system, first detach the port from testpmd. 1402Then the device should be moved under kernel management. 1403Finally the device can be removed using kernel pci hotplug functionality. 1404 1405For example, to move a pci device under kernel management: 1406 1407.. code-block:: console 1408 1409 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0 1410 1411 ./usertools/dpdk-devbind.py --status 1412 1413 Network devices using DPDK-compatible driver 1414 ============================================ 1415 <none> 1416 1417 Network devices using kernel driver 1418 =================================== 1419 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio 1420 1421To remove a port created by a virtual device, above steps are not needed. 1422 1423port start 1424~~~~~~~~~~ 1425 1426Start all ports or a specific port:: 1427 1428 testpmd> port start (port_id|all) 1429 1430port stop 1431~~~~~~~~~ 1432 1433Stop all ports or a specific port:: 1434 1435 testpmd> port stop (port_id|all) 1436 1437port close 1438~~~~~~~~~~ 1439 1440Close all ports or a specific port:: 1441 1442 testpmd> port close (port_id|all) 1443 1444port start/stop queue 1445~~~~~~~~~~~~~~~~~~~~~ 1446 1447Start/stop a rx/tx queue on a specific port:: 1448 1449 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop) 1450 1451Only take effect when port is started. 1452 1453port config - speed 1454~~~~~~~~~~~~~~~~~~~ 1455 1456Set the speed and duplex mode for all ports or a specific port:: 1457 1458 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \ 1459 duplex (half|full|auto) 1460 1461port config - queues/descriptors 1462~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1463 1464Set number of queues/descriptors for rxq, txq, rxd and txd:: 1465 1466 testpmd> port config all (rxq|txq|rxd|txd) (value) 1467 1468This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options. 1469 1470port config - max-pkt-len 1471~~~~~~~~~~~~~~~~~~~~~~~~~ 1472 1473Set the maximum packet length:: 1474 1475 testpmd> port config all max-pkt-len (value) 1476 1477This is equivalent to the ``--max-pkt-len`` command-line option. 1478 1479port config - CRC Strip 1480~~~~~~~~~~~~~~~~~~~~~~~ 1481 1482Set hardware CRC stripping on or off for all ports:: 1483 1484 testpmd> port config all crc-strip (on|off) 1485 1486CRC stripping is on by default. 1487 1488The ``off`` option is equivalent to the ``--disable-crc-strip`` command-line option. 1489 1490port config - scatter 1491~~~~~~~~~~~~~~~~~~~~~~~ 1492 1493Set RX scatter mode on or off for all ports:: 1494 1495 testpmd> port config all scatter (on|off) 1496 1497RX scatter mode is off by default. 1498 1499The ``on`` option is equivalent to the ``--enable-scatter`` command-line option. 1500 1501port config - TX queue flags 1502~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1503 1504Set a hexadecimal bitmap of TX queue flags for all ports:: 1505 1506 testpmd> port config all txqflags value 1507 1508This command is equivalent to the ``--txqflags`` command-line option. 1509 1510port config - RX Checksum 1511~~~~~~~~~~~~~~~~~~~~~~~~~ 1512 1513Set hardware RX checksum offload to on or off for all ports:: 1514 1515 testpmd> port config all rx-cksum (on|off) 1516 1517Checksum offload is off by default. 1518 1519The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option. 1520 1521port config - VLAN 1522~~~~~~~~~~~~~~~~~~ 1523 1524Set hardware VLAN on or off for all ports:: 1525 1526 testpmd> port config all hw-vlan (on|off) 1527 1528Hardware VLAN is on by default. 1529 1530The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option. 1531 1532port config - VLAN filter 1533~~~~~~~~~~~~~~~~~~~~~~~~~ 1534 1535Set hardware VLAN filter on or off for all ports:: 1536 1537 testpmd> port config all hw-vlan-filter (on|off) 1538 1539Hardware VLAN filter is on by default. 1540 1541The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option. 1542 1543port config - VLAN strip 1544~~~~~~~~~~~~~~~~~~~~~~~~ 1545 1546Set hardware VLAN strip on or off for all ports:: 1547 1548 testpmd> port config all hw-vlan-strip (on|off) 1549 1550Hardware VLAN strip is on by default. 1551 1552The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option. 1553 1554port config - VLAN extend 1555~~~~~~~~~~~~~~~~~~~~~~~~~ 1556 1557Set hardware VLAN extend on or off for all ports:: 1558 1559 testpmd> port config all hw-vlan-extend (on|off) 1560 1561Hardware VLAN extend is off by default. 1562 1563The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option. 1564 1565port config - Drop Packets 1566~~~~~~~~~~~~~~~~~~~~~~~~~~ 1567 1568Set packet drop for packets with no descriptors on or off for all ports:: 1569 1570 testpmd> port config all drop-en (on|off) 1571 1572Packet dropping for packets with no descriptors is off by default. 1573 1574The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option. 1575 1576port config - RSS 1577~~~~~~~~~~~~~~~~~ 1578 1579Set the RSS (Receive Side Scaling) mode on or off:: 1580 1581 testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none) 1582 1583RSS is on by default. 1584 1585The ``none`` option is equivalent to the ``--disable-rss`` command-line option. 1586 1587port config - RSS Reta 1588~~~~~~~~~~~~~~~~~~~~~~ 1589 1590Set the RSS (Receive Side Scaling) redirection table:: 1591 1592 testpmd> port config all rss reta (hash,queue)[,(hash,queue)] 1593 1594port config - DCB 1595~~~~~~~~~~~~~~~~~ 1596 1597Set the DCB mode for an individual port:: 1598 1599 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off) 1600 1601The traffic class should be 4 or 8. 1602 1603port config - Burst 1604~~~~~~~~~~~~~~~~~~~ 1605 1606Set the number of packets per burst:: 1607 1608 testpmd> port config all burst (value) 1609 1610This is equivalent to the ``--burst`` command-line option. 1611 1612port config - Threshold 1613~~~~~~~~~~~~~~~~~~~~~~~ 1614 1615Set thresholds for TX/RX queues:: 1616 1617 testpmd> port config all (threshold) (value) 1618 1619Where the threshold type can be: 1620 1621* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255. 1622 1623* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255. 1624 1625* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255. 1626 1627* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255. 1628 1629* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255. 1630 1631* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255. 1632 1633* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd. 1634 1635* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd. 1636 1637* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd. 1638 1639These threshold options are also available from the command-line. 1640 1641port config - E-tag 1642~~~~~~~~~~~~~~~~~~~ 1643 1644Set the value of ether-type for E-tag:: 1645 1646 testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value) 1647 1648Enable/disable the E-tag support:: 1649 1650 testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable) 1651 1652 1653Link Bonding Functions 1654---------------------- 1655 1656The Link Bonding functions make it possible to dynamically create and 1657manage link bonding devices from within testpmd interactive prompt. 1658 1659create bonded device 1660~~~~~~~~~~~~~~~~~~~~ 1661 1662Create a new bonding device:: 1663 1664 testpmd> create bonded device (mode) (socket) 1665 1666For example, to create a bonded device in mode 1 on socket 0:: 1667 1668 testpmd> create bonded 1 0 1669 created new bonded device (port X) 1670 1671add bonding slave 1672~~~~~~~~~~~~~~~~~ 1673 1674Adds Ethernet device to a Link Bonding device:: 1675 1676 testpmd> add bonding slave (slave id) (port id) 1677 1678For example, to add Ethernet device (port 6) to a Link Bonding device (port 10):: 1679 1680 testpmd> add bonding slave 6 10 1681 1682 1683remove bonding slave 1684~~~~~~~~~~~~~~~~~~~~ 1685 1686Removes an Ethernet slave device from a Link Bonding device:: 1687 1688 testpmd> remove bonding slave (slave id) (port id) 1689 1690For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10):: 1691 1692 testpmd> remove bonding slave 6 10 1693 1694set bonding mode 1695~~~~~~~~~~~~~~~~ 1696 1697Set the Link Bonding mode of a Link Bonding device:: 1698 1699 testpmd> set bonding mode (value) (port id) 1700 1701For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3):: 1702 1703 testpmd> set bonding mode 3 10 1704 1705set bonding primary 1706~~~~~~~~~~~~~~~~~~~ 1707 1708Set an Ethernet slave device as the primary device on a Link Bonding device:: 1709 1710 testpmd> set bonding primary (slave id) (port id) 1711 1712For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10):: 1713 1714 testpmd> set bonding primary 6 10 1715 1716set bonding mac 1717~~~~~~~~~~~~~~~ 1718 1719Set the MAC address of a Link Bonding device:: 1720 1721 testpmd> set bonding mac (port id) (mac) 1722 1723For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01:: 1724 1725 testpmd> set bonding mac 10 00:00:00:00:00:01 1726 1727set bonding xmit_balance_policy 1728~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1729 1730Set the transmission policy for a Link Bonding device when it is in Balance XOR mode:: 1731 1732 testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34) 1733 1734For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports):: 1735 1736 testpmd> set bonding xmit_balance_policy 10 l34 1737 1738 1739set bonding mon_period 1740~~~~~~~~~~~~~~~~~~~~~~ 1741 1742Set the link status monitoring polling period in milliseconds for a bonding device. 1743 1744This adds support for PMD slave devices which do not support link status interrupts. 1745When the mon_period is set to a value greater than 0 then all PMD's which do not support 1746link status ISR will be queried every polling interval to check if their link status has changed:: 1747 1748 testpmd> set bonding mon_period (port_id) (value) 1749 1750For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms:: 1751 1752 testpmd> set bonding mon_period 5 150 1753 1754 1755show bonding config 1756~~~~~~~~~~~~~~~~~~~ 1757 1758Show the current configuration of a Link Bonding device:: 1759 1760 testpmd> show bonding config (port id) 1761 1762For example, 1763to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4) 1764in balance mode with a transmission policy of layer 2+3:: 1765 1766 testpmd> show bonding config 9 1767 Bonding mode: 2 1768 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23 1769 Slaves (3): [1 3 4] 1770 Active Slaves (3): [1 3 4] 1771 Primary: [3] 1772 1773 1774Register Functions 1775------------------ 1776 1777The Register Functions can be used to read from and write to registers on the network card referenced by a port number. 1778This is mainly useful for debugging purposes. 1779Reference should be made to the appropriate datasheet for the network card for details on the register addresses 1780and fields that can be accessed. 1781 1782read reg 1783~~~~~~~~ 1784 1785Display the value of a port register:: 1786 1787 testpmd> read reg (port_id) (address) 1788 1789For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller:: 1790 1791 testpmd> read reg 0 0xEE00 1792 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241) 1793 1794read regfield 1795~~~~~~~~~~~~~ 1796 1797Display a port register bit field:: 1798 1799 testpmd> read regfield (port_id) (address) (bit_x) (bit_y) 1800 1801For example, reading the lowest two bits from the register in the example above:: 1802 1803 testpmd> read regfield 0 0xEE00 0 1 1804 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1) 1805 1806read regbit 1807~~~~~~~~~~~ 1808 1809Display a single port register bit:: 1810 1811 testpmd> read regbit (port_id) (address) (bit_x) 1812 1813For example, reading the lowest bit from the register in the example above:: 1814 1815 testpmd> read regbit 0 0xEE00 0 1816 port 0 PCI register at offset 0xEE00: bit 0=1 1817 1818write reg 1819~~~~~~~~~ 1820 1821Set the value of a port register:: 1822 1823 testpmd> write reg (port_id) (address) (value) 1824 1825For example, to clear a register:: 1826 1827 testpmd> write reg 0 0xEE00 0x0 1828 port 0 PCI register at offset 0xEE00: 0x00000000 (0) 1829 1830write regfield 1831~~~~~~~~~~~~~~ 1832 1833Set bit field of a port register:: 1834 1835 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value) 1836 1837For example, writing to the register cleared in the example above:: 1838 1839 testpmd> write regfield 0 0xEE00 0 1 2 1840 port 0 PCI register at offset 0xEE00: 0x00000002 (2) 1841 1842write regbit 1843~~~~~~~~~~~~ 1844 1845Set single bit value of a port register:: 1846 1847 testpmd> write regbit (port_id) (address) (bit_x) (value) 1848 1849For example, to set the high bit in the register from the example above:: 1850 1851 testpmd> write regbit 0 0xEE00 31 1 1852 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658) 1853 1854 1855Filter Functions 1856---------------- 1857 1858This section details the available filter functions that are available. 1859 1860Note these functions interface the deprecated legacy filtering framework, 1861superseded by *rte_flow*. See `Flow rules management`_. 1862 1863ethertype_filter 1864~~~~~~~~~~~~~~~~~~~~ 1865 1866Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue:: 1867 1868 ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \ 1869 ethertype (ether_type) (drop|fwd) queue (queue_id) 1870 1871The available information parameters are: 1872 1873* ``port_id``: The port which the Ethertype filter assigned on. 1874 1875* ``mac_addr``: Compare destination mac address. 1876 1877* ``mac_ignr``: Ignore destination mac address match. 1878 1879* ``mac_address``: Destination mac address to match. 1880 1881* ``ether_type``: The EtherType value want to match, 1882 for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid. 1883 1884* ``queue_id``: The receive queue associated with this EtherType filter. 1885 It is meaningless when deleting or dropping. 1886 1887Example, to add/remove an ethertype filter rule:: 1888 1889 testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \ 1890 ethertype 0x0806 fwd queue 3 1891 1892 testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \ 1893 ethertype 0x0806 fwd queue 3 1894 18952tuple_filter 1896~~~~~~~~~~~~~~~~~ 1897 1898Add or delete a 2-tuple filter, 1899which identifies packets by specific protocol and destination TCP/UDP port 1900and forwards packets into one of the receive queues:: 1901 1902 2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \ 1903 protocol (protocol_value) mask (mask_value) \ 1904 tcp_flags (tcp_flags_value) priority (prio_value) \ 1905 queue (queue_id) 1906 1907The available information parameters are: 1908 1909* ``port_id``: The port which the 2-tuple filter assigned on. 1910 1911* ``dst_port_value``: Destination port in L4. 1912 1913* ``protocol_value``: IP L4 protocol. 1914 1915* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate. 1916 1917* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP). 1918 1919* ``prio_value``: Priority of this filter. 1920 1921* ``queue_id``: The receive queue associated with this 2-tuple filter. 1922 1923Example, to add/remove an 2tuple filter rule:: 1924 1925 testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \ 1926 tcp_flags 0x02 priority 3 queue 3 1927 1928 testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \ 1929 tcp_flags 0x02 priority 3 queue 3 1930 19315tuple_filter 1932~~~~~~~~~~~~~~~~~ 1933 1934Add or delete a 5-tuple filter, 1935which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port) 1936and routes packets into one of the receive queues:: 1937 1938 5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \ 1939 (src_address) dst_port (dst_port_value) \ 1940 src_port (src_port_value) protocol (protocol_value) \ 1941 mask (mask_value) tcp_flags (tcp_flags_value) \ 1942 priority (prio_value) queue (queue_id) 1943 1944The available information parameters are: 1945 1946* ``port_id``: The port which the 5-tuple filter assigned on. 1947 1948* ``dst_address``: Destination IP address. 1949 1950* ``src_address``: Source IP address. 1951 1952* ``dst_port_value``: TCP/UDP destination port. 1953 1954* ``src_port_value``: TCP/UDP source port. 1955 1956* ``protocol_value``: L4 protocol. 1957 1958* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate 1959 1960* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP). 1961 1962* ``prio_value``: The priority of this filter. 1963 1964* ``queue_id``: The receive queue associated with this 5-tuple filter. 1965 1966Example, to add/remove an 5tuple filter rule:: 1967 1968 testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \ 1969 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \ 1970 flags 0x0 priority 3 queue 3 1971 1972 testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \ 1973 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \ 1974 flags 0x0 priority 3 queue 3 1975 1976syn_filter 1977~~~~~~~~~~ 1978 1979Using the SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue:: 1980 1981 syn_filter (port_id) (add|del) priority (high|low) queue (queue_id) 1982 1983The available information parameters are: 1984 1985* ``port_id``: The port which the SYN filter assigned on. 1986 1987* ``high``: This SYN filter has higher priority than other filters. 1988 1989* ``low``: This SYN filter has lower priority than other filters. 1990 1991* ``queue_id``: The receive queue associated with this SYN filter 1992 1993Example:: 1994 1995 testpmd> syn_filter 0 add priority high queue 3 1996 1997flex_filter 1998~~~~~~~~~~~ 1999 2000With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet 2001and routed into one of the receive queues:: 2002 2003 flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \ 2004 mask (mask_value) priority (prio_value) queue (queue_id) 2005 2006The available information parameters are: 2007 2008* ``port_id``: The port which the Flex filter is assigned on. 2009 2010* ``len_value``: Filter length in bytes, no greater than 128. 2011 2012* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match. 2013 2014* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match. 2015 2016* ``prio_value``: The priority of this filter. 2017 2018* ``queue_id``: The receive queue associated with this Flex filter. 2019 2020Example:: 2021 2022 testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \ 2023 mask 000C priority 3 queue 3 2024 2025 testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \ 2026 mask 000C priority 3 queue 3 2027 2028 2029.. _testpmd_flow_director: 2030 2031flow_director_filter 2032~~~~~~~~~~~~~~~~~~~~ 2033 2034The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues. 2035 2036Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and 2037Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter: 2038 2039* Perfect match filters. 2040 The hardware checks a match between the masked fields of the received packets and the programmed filters. 2041 The masked fields are for IP flow. 2042 2043* Signature filters. 2044 The hardware checks a match between a hash-based signature of the masked fields of the received packet. 2045 2046* Perfect-mac-vlan match filters. 2047 The hardware checks a match between the masked fields of the received packets and the programmed filters. 2048 The masked fields are for MAC VLAN flow. 2049 2050* Perfect-tunnel match filters. 2051 The hardware checks a match between the masked fields of the received packets and the programmed filters. 2052 The masked fields are for tunnel flow. 2053 2054The Flow Director filters can match the different fields for different type of packet: flow type, specific input set 2055per flow type and the flexible payload. 2056 2057The Flow Director can also mask out parts of all of these fields so that filters 2058are only applied to certain fields or parts of the fields. 2059 2060Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information. 2061 2062# Commands to add flow director filters of different flow types:: 2063 2064 flow_director_filter (port_id) mode IP (add|del|update) \ 2065 flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \ 2066 src (src_ip_address) dst (dst_ip_address) \ 2067 tos (tos_value) proto (proto_value) ttl (ttl_value) \ 2068 vlan (vlan_value) flexbytes (flexbytes_value) \ 2069 (drop|fwd) pf|vf(vf_id) queue (queue_id) \ 2070 fd_id (fd_id_value) 2071 2072 flow_director_filter (port_id) mode IP (add|del|update) \ 2073 flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \ 2074 src (src_ip_address) (src_port) \ 2075 dst (dst_ip_address) (dst_port) \ 2076 tos (tos_value) ttl (ttl_value) \ 2077 vlan (vlan_value) flexbytes (flexbytes_value) \ 2078 (drop|fwd) queue pf|vf(vf_id) (queue_id) \ 2079 fd_id (fd_id_value) 2080 2081 flow_director_filter (port_id) mode IP (add|del|update) \ 2082 flow (ipv4-sctp|ipv6-sctp) \ 2083 src (src_ip_address) (src_port) \ 2084 dst (dst_ip_address) (dst_port) \ 2085 tos (tos_value) ttl (ttl_value) \ 2086 tag (verification_tag) vlan (vlan_value) \ 2087 flexbytes (flexbytes_value) (drop|fwd) \ 2088 pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value) 2089 2090 flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \ 2091 ether (ethertype) flexbytes (flexbytes_value) \ 2092 (drop|fwd) pf|vf(vf_id) queue (queue_id) 2093 fd_id (fd_id_value) 2094 2095 flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \ 2096 mac (mac_address) vlan (vlan_value) \ 2097 flexbytes (flexbytes_value) (drop|fwd) \ 2098 queue (queue_id) fd_id (fd_id_value) 2099 2100 flow_director_filter (port_id) mode Tunnel (add|del|update) \ 2101 mac (mac_address) vlan (vlan_value) \ 2102 tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \ 2103 flexbytes (flexbytes_value) (drop|fwd) \ 2104 queue (queue_id) fd_id (fd_id_value) 2105 2106For example, to add an ipv4-udp flow type filter:: 2107 2108 testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \ 2109 dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \ 2110 fwd pf queue 1 fd_id 1 2111 2112For example, add an ipv4-other flow type filter:: 2113 2114 testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \ 2115 dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \ 2116 flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1 2117 2118flush_flow_director 2119~~~~~~~~~~~~~~~~~~~ 2120 2121Flush all flow director filters on a device:: 2122 2123 testpmd> flush_flow_director (port_id) 2124 2125Example, to flush all flow director filter on port 0:: 2126 2127 testpmd> flush_flow_director 0 2128 2129flow_director_mask 2130~~~~~~~~~~~~~~~~~~ 2131 2132Set flow director's input masks:: 2133 2134 flow_director_mask (port_id) mode IP vlan (vlan_value) \ 2135 src_mask (ipv4_src) (ipv6_src) (src_port) \ 2136 dst_mask (ipv4_dst) (ipv6_dst) (dst_port) 2137 2138 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value) 2139 2140 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \ 2141 mac (mac_value) tunnel-type (tunnel_type_value) \ 2142 tunnel-id (tunnel_id_value) 2143 2144Example, to set flow director mask on port 0:: 2145 2146 testpmd> flow_director_mask 0 mode IP vlan 0xefff \ 2147 src_mask 255.255.255.255 \ 2148 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \ 2149 dst_mask 255.255.255.255 \ 2150 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF 2151 2152flow_director_flex_mask 2153~~~~~~~~~~~~~~~~~~~~~~~ 2154 2155set masks of flow director's flexible payload based on certain flow type:: 2156 2157 testpmd> flow_director_flex_mask (port_id) \ 2158 flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2159 ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \ 2160 l2_payload|all) (mask) 2161 2162Example, to set flow director's flex mask for all flow type on port 0:: 2163 2164 testpmd> flow_director_flex_mask 0 flow all \ 2165 (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0) 2166 2167 2168flow_director_flex_payload 2169~~~~~~~~~~~~~~~~~~~~~~~~~~ 2170 2171Configure flexible payload selection:: 2172 2173 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config) 2174 2175For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload:: 2176 2177 testpmd> flow_director_flex_payload 0 l4 \ 2178 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19) 2179 2180get_sym_hash_ena_per_port 2181~~~~~~~~~~~~~~~~~~~~~~~~~ 2182 2183Get symmetric hash enable configuration per port:: 2184 2185 get_sym_hash_ena_per_port (port_id) 2186 2187For example, to get symmetric hash enable configuration of port 1:: 2188 2189 testpmd> get_sym_hash_ena_per_port 1 2190 2191set_sym_hash_ena_per_port 2192~~~~~~~~~~~~~~~~~~~~~~~~~ 2193 2194Set symmetric hash enable configuration per port to enable or disable:: 2195 2196 set_sym_hash_ena_per_port (port_id) (enable|disable) 2197 2198For example, to set symmetric hash enable configuration of port 1 to enable:: 2199 2200 testpmd> set_sym_hash_ena_per_port 1 enable 2201 2202get_hash_global_config 2203~~~~~~~~~~~~~~~~~~~~~~ 2204 2205Get the global configurations of hash filters:: 2206 2207 get_hash_global_config (port_id) 2208 2209For example, to get the global configurations of hash filters of port 1:: 2210 2211 testpmd> get_hash_global_config 1 2212 2213set_hash_global_config 2214~~~~~~~~~~~~~~~~~~~~~~ 2215 2216Set the global configurations of hash filters:: 2217 2218 set_hash_global_config (port_id) (toeplitz|simple_xor|default) \ 2219 (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \ 2220 ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \ 2221 (enable|disable) 2222 2223For example, to enable simple_xor for flow type of ipv6 on port 2:: 2224 2225 testpmd> set_hash_global_config 2 simple_xor ipv6 enable 2226 2227set_hash_input_set 2228~~~~~~~~~~~~~~~~~~ 2229 2230Set the input set for hash:: 2231 2232 set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2233 ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \ 2234 l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \ 2235 ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \ 2236 tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \ 2237 udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \ 2238 fld-8th|none) (select|add) 2239 2240For example, to add source IP to hash input set for flow type of ipv4-udp on port 0:: 2241 2242 testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add 2243 2244set_fdir_input_set 2245~~~~~~~~~~~~~~~~~~ 2246 2247The Flow Director filters can match the different fields for different type of packet, i.e. specific input set 2248on per flow type and the flexible payload. This command can be used to change input set for each flow type. 2249 2250Set the input set for flow director:: 2251 2252 set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2253 ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \ 2254 l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \ 2255 ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \ 2256 tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \ 2257 sctp-dst-port|sctp-veri-tag|none) (select|add) 2258 2259For example to add source IP to FD input set for flow type of ipv4-udp on port 0:: 2260 2261 testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add 2262 2263global_config 2264~~~~~~~~~~~~~ 2265 2266Set different GRE key length for input set:: 2267 2268 global_config (port_id) gre-key-len (number in bytes) 2269 2270For example to set GRE key length for input set to 4 bytes on port 0:: 2271 2272 testpmd> global_config 0 gre-key-len 4 2273 2274 2275.. _testpmd_rte_flow: 2276 2277Flow rules management 2278--------------------- 2279 2280Control of the generic flow API (*rte_flow*) is fully exposed through the 2281``flow`` command (validation, creation, destruction, queries and operation 2282modes). 2283 2284Considering *rte_flow* overlaps with all `Filter Functions`_, using both 2285features simultaneously may cause undefined side-effects and is therefore 2286not recommended. 2287 2288``flow`` syntax 2289~~~~~~~~~~~~~~~ 2290 2291Because the ``flow`` command uses dynamic tokens to handle the large number 2292of possible flow rules combinations, its behavior differs slightly from 2293other commands, in particular: 2294 2295- Pressing *?* or the *<tab>* key displays contextual help for the current 2296 token, not that of the entire command. 2297 2298- Optional and repeated parameters are supported (provided they are listed 2299 in the contextual help). 2300 2301The first parameter stands for the operation mode. Possible operations and 2302their general syntax are described below. They are covered in detail in the 2303following sections. 2304 2305- Check whether a flow rule can be created:: 2306 2307 flow validate {port_id} 2308 [group {group_id}] [priority {level}] [ingress] [egress] 2309 pattern {item} [/ {item} [...]] / end 2310 actions {action} [/ {action} [...]] / end 2311 2312- Create a flow rule:: 2313 2314 flow create {port_id} 2315 [group {group_id}] [priority {level}] [ingress] [egress] 2316 pattern {item} [/ {item} [...]] / end 2317 actions {action} [/ {action} [...]] / end 2318 2319- Destroy specific flow rules:: 2320 2321 flow destroy {port_id} rule {rule_id} [...] 2322 2323- Destroy all flow rules:: 2324 2325 flow flush {port_id} 2326 2327- Query an existing flow rule:: 2328 2329 flow query {port_id} {rule_id} {action} 2330 2331- List existing flow rules sorted by priority, filtered by group 2332 identifiers:: 2333 2334 flow list {port_id} [group {group_id}] [...] 2335 2336- Restrict ingress traffic to the defined flow rules:: 2337 2338 flow isolate {port_id} {boolean} 2339 2340Validating flow rules 2341~~~~~~~~~~~~~~~~~~~~~ 2342 2343``flow validate`` reports whether a flow rule would be accepted by the 2344underlying device in its current state but stops short of creating it. It is 2345bound to ``rte_flow_validate()``:: 2346 2347 flow validate {port_id} 2348 [group {group_id}] [priority {level}] [ingress] [egress] 2349 pattern {item} [/ {item} [...]] / end 2350 actions {action} [/ {action} [...]] / end 2351 2352If successful, it will show:: 2353 2354 Flow rule validated 2355 2356Otherwise it will show an error message of the form:: 2357 2358 Caught error type [...] ([...]): [...] 2359 2360This command uses the same parameters as ``flow create``, their format is 2361described in `Creating flow rules`_. 2362 2363Check whether redirecting any Ethernet packet received on port 0 to RX queue 2364index 6 is supported:: 2365 2366 testpmd> flow validate 0 ingress pattern eth / end 2367 actions queue index 6 / end 2368 Flow rule validated 2369 testpmd> 2370 2371Port 0 does not support TCPv6 rules:: 2372 2373 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end 2374 actions drop / end 2375 Caught error type 9 (specific pattern item): Invalid argument 2376 testpmd> 2377 2378Creating flow rules 2379~~~~~~~~~~~~~~~~~~~ 2380 2381``flow create`` validates and creates the specified flow rule. It is bound 2382to ``rte_flow_create()``:: 2383 2384 flow create {port_id} 2385 [group {group_id}] [priority {level}] [ingress] [egress] 2386 pattern {item} [/ {item} [...]] / end 2387 actions {action} [/ {action} [...]] / end 2388 2389If successful, it will return a flow rule ID usable with other commands:: 2390 2391 Flow rule #[...] created 2392 2393Otherwise it will show an error message of the form:: 2394 2395 Caught error type [...] ([...]): [...] 2396 2397Parameters describe in the following order: 2398 2399- Attributes (*group*, *priority*, *ingress*, *egress* tokens). 2400- A matching pattern, starting with the *pattern* token and terminated by an 2401 *end* pattern item. 2402- Actions, starting with the *actions* token and terminated by an *end* 2403 action. 2404 2405These translate directly to *rte_flow* objects provided as-is to the 2406underlying functions. 2407 2408The shortest valid definition only comprises mandatory tokens:: 2409 2410 testpmd> flow create 0 pattern end actions end 2411 2412Note that PMDs may refuse rules that essentially do nothing such as this 2413one. 2414 2415**All unspecified object values are automatically initialized to 0.** 2416 2417Attributes 2418^^^^^^^^^^ 2419 2420These tokens affect flow rule attributes (``struct rte_flow_attr``) and are 2421specified before the ``pattern`` token. 2422 2423- ``group {group id}``: priority group. 2424- ``priority {level}``: priority level within group. 2425- ``ingress``: rule applies to ingress traffic. 2426- ``egress``: rule applies to egress traffic. 2427 2428Each instance of an attribute specified several times overrides the previous 2429value as shown below (group 4 is used):: 2430 2431 testpmd> flow create 0 group 42 group 24 group 4 [...] 2432 2433Note that once enabled, ``ingress`` and ``egress`` cannot be disabled. 2434 2435While not specifying a direction is an error, some rules may allow both 2436simultaneously. 2437 2438Most rules affect RX therefore contain the ``ingress`` token:: 2439 2440 testpmd> flow create 0 ingress pattern [...] 2441 2442Matching pattern 2443^^^^^^^^^^^^^^^^ 2444 2445A matching pattern starts after the ``pattern`` token. It is made of pattern 2446items and is terminated by a mandatory ``end`` item. 2447 2448Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum 2449rte_flow_item_type``). 2450 2451The ``/`` token is used as a separator between pattern items as shown 2452below:: 2453 2454 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...] 2455 2456Note that protocol items like these must be stacked from lowest to highest 2457layer to make sense. For instance, the following rule is either invalid or 2458unlikely to match any packet:: 2459 2460 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...] 2461 2462More information on these restrictions can be found in the *rte_flow* 2463documentation. 2464 2465Several items support additional specification structures, for example 2466``ipv4`` allows specifying source and destination addresses as follows:: 2467 2468 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1 2469 dst is 10.2.0.0 / end [...] 2470 2471This rule matches all IPv4 traffic with the specified properties. 2472 2473In this example, ``src`` and ``dst`` are field names of the underlying 2474``struct rte_flow_item_ipv4`` object. All item properties can be specified 2475in a similar fashion. 2476 2477The ``is`` token means that the subsequent value must be matched exactly, 2478and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item`` 2479accordingly. Possible assignment tokens are: 2480 2481- ``is``: match value perfectly (with full bit-mask). 2482- ``spec``: match value according to configured bit-mask. 2483- ``last``: specify upper bound to establish a range. 2484- ``mask``: specify bit-mask with relevant bits set to one. 2485- ``prefix``: generate bit-mask from a prefix length. 2486 2487These yield identical results:: 2488 2489 ipv4 src is 10.1.1.1 2490 2491:: 2492 2493 ipv4 src spec 10.1.1.1 src mask 255.255.255.255 2494 2495:: 2496 2497 ipv4 src spec 10.1.1.1 src prefix 32 2498 2499:: 2500 2501 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value 2502 2503:: 2504 2505 ipv4 src is 10.1.1.1 src last 0 # 0 disables range 2506 2507Inclusive ranges can be defined with ``last``:: 2508 2509 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4 2510 2511Note that ``mask`` affects both ``spec`` and ``last``:: 2512 2513 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0 2514 # matches 10.1.0.0 to 10.2.255.255 2515 2516Properties can be modified multiple times:: 2517 2518 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4 2519 2520:: 2521 2522 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16 2523 2524Pattern items 2525^^^^^^^^^^^^^ 2526 2527This section lists supported pattern items and their attributes, if any. 2528 2529- ``end``: end list of pattern items. 2530 2531- ``void``: no-op pattern item. 2532 2533- ``invert``: perform actions when pattern does not match. 2534 2535- ``any``: match any protocol for the current layer. 2536 2537 - ``num {unsigned}``: number of layers covered. 2538 2539- ``pf``: match packets addressed to the physical function. 2540 2541- ``vf``: match packets addressed to a virtual function ID. 2542 2543 - ``id {unsigned}``: destination VF ID. 2544 2545- ``port``: device-specific physical port index to use. 2546 2547 - ``index {unsigned}``: physical port index. 2548 2549- ``raw``: match an arbitrary byte string. 2550 2551 - ``relative {boolean}``: look for pattern after the previous item. 2552 - ``search {boolean}``: search pattern from offset (see also limit). 2553 - ``offset {integer}``: absolute or relative offset for pattern. 2554 - ``limit {unsigned}``: search area limit for start of pattern. 2555 - ``pattern {string}``: byte string to look for. 2556 2557- ``eth``: match Ethernet header. 2558 2559 - ``dst {MAC-48}``: destination MAC. 2560 - ``src {MAC-48}``: source MAC. 2561 - ``type {unsigned}``: EtherType. 2562 2563- ``vlan``: match 802.1Q/ad VLAN tag. 2564 2565 - ``tpid {unsigned}``: tag protocol identifier. 2566 - ``tci {unsigned}``: tag control information. 2567 - ``pcp {unsigned}``: priority code point. 2568 - ``dei {unsigned}``: drop eligible indicator. 2569 - ``vid {unsigned}``: VLAN identifier. 2570 2571- ``ipv4``: match IPv4 header. 2572 2573 - ``tos {unsigned}``: type of service. 2574 - ``ttl {unsigned}``: time to live. 2575 - ``proto {unsigned}``: next protocol ID. 2576 - ``src {ipv4 address}``: source address. 2577 - ``dst {ipv4 address}``: destination address. 2578 2579- ``ipv6``: match IPv6 header. 2580 2581 - ``tc {unsigned}``: traffic class. 2582 - ``flow {unsigned}``: flow label. 2583 - ``proto {unsigned}``: protocol (next header). 2584 - ``hop {unsigned}``: hop limit. 2585 - ``src {ipv6 address}``: source address. 2586 - ``dst {ipv6 address}``: destination address. 2587 2588- ``icmp``: match ICMP header. 2589 2590 - ``type {unsigned}``: ICMP packet type. 2591 - ``code {unsigned}``: ICMP packet code. 2592 2593- ``udp``: match UDP header. 2594 2595 - ``src {unsigned}``: UDP source port. 2596 - ``dst {unsigned}``: UDP destination port. 2597 2598- ``tcp``: match TCP header. 2599 2600 - ``src {unsigned}``: TCP source port. 2601 - ``dst {unsigned}``: TCP destination port. 2602 2603- ``sctp``: match SCTP header. 2604 2605 - ``src {unsigned}``: SCTP source port. 2606 - ``dst {unsigned}``: SCTP destination port. 2607 - ``tag {unsigned}``: validation tag. 2608 - ``cksum {unsigned}``: checksum. 2609 2610- ``vxlan``: match VXLAN header. 2611 2612 - ``vni {unsigned}``: VXLAN identifier. 2613 2614- ``e_tag``: match IEEE 802.1BR E-Tag header. 2615 2616 - ``grp_ecid_b {unsigned}``: GRP and E-CID base. 2617 2618- ``nvgre``: match NVGRE header. 2619 2620 - ``tni {unsigned}``: virtual subnet ID. 2621 2622- ``mpls``: match MPLS header. 2623 2624 - ``label {unsigned}``: MPLS label. 2625 2626- ``gre``: match GRE header. 2627 2628 - ``protocol {unsigned}``: protocol type. 2629 2630- ``fuzzy``: fuzzy pattern match, expect faster than default. 2631 2632 - ``thresh {unsigned}``: accuracy threshold. 2633 2634Actions list 2635^^^^^^^^^^^^ 2636 2637A list of actions starts after the ``actions`` token in the same fashion as 2638`Matching pattern`_; actions are separated by ``/`` tokens and the list is 2639terminated by a mandatory ``end`` action. 2640 2641Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum 2642rte_flow_action_type``). 2643 2644Dropping all incoming UDPv4 packets can be expressed as follows:: 2645 2646 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2647 actions drop / end 2648 2649Several actions have configurable properties which must be specified when 2650there is no valid default value. For example, ``queue`` requires a target 2651queue index. 2652 2653This rule redirects incoming UDPv4 traffic to queue index 6:: 2654 2655 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2656 actions queue index 6 / end 2657 2658While this one could be rejected by PMDs (unspecified queue index):: 2659 2660 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2661 actions queue / end 2662 2663As defined by *rte_flow*, the list is not ordered, all actions of a given 2664rule are performed simultaneously. These are equivalent:: 2665 2666 queue index 6 / void / mark id 42 / end 2667 2668:: 2669 2670 void / mark id 42 / queue index 6 / end 2671 2672All actions in a list should have different types, otherwise only the last 2673action of a given type is taken into account:: 2674 2675 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6 2676 2677:: 2678 2679 drop / drop / drop / end # drop is performed only once 2680 2681:: 2682 2683 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24 2684 2685Considering they are performed simultaneously, opposite and overlapping 2686actions can sometimes be combined when the end result is unambiguous:: 2687 2688 drop / queue index 6 / end # drop has no effect 2689 2690:: 2691 2692 drop / dup index 6 / end # same as above 2693 2694:: 2695 2696 queue index 6 / rss queues 6 7 8 / end # queue has no effect 2697 2698:: 2699 2700 drop / passthru / end # drop has no effect 2701 2702Note that PMDs may still refuse such combinations. 2703 2704Actions 2705^^^^^^^ 2706 2707This section lists supported actions and their attributes, if any. 2708 2709- ``end``: end list of actions. 2710 2711- ``void``: no-op action. 2712 2713- ``passthru``: let subsequent rule process matched packets. 2714 2715- ``mark``: attach 32 bit value to packets. 2716 2717 - ``id {unsigned}``: 32 bit value to return with packets. 2718 2719- ``flag``: flag packets. 2720 2721- ``queue``: assign packets to a given queue index. 2722 2723 - ``index {unsigned}``: queue index to use. 2724 2725- ``drop``: drop packets (note: passthru has priority). 2726 2727- ``count``: enable counters for this rule. 2728 2729- ``dup``: duplicate packets to a given queue index. 2730 2731 - ``index {unsigned}``: queue index to duplicate packets to. 2732 2733- ``rss``: spread packets among several queues. 2734 2735 - ``queues [{unsigned} [...]] end``: queue indices to use. 2736 2737- ``pf``: redirect packets to physical device function. 2738 2739- ``vf``: redirect packets to virtual device function. 2740 2741 - ``original {boolean}``: use original VF ID if possible. 2742 - ``id {unsigned}``: VF ID to redirect packets to. 2743 2744Destroying flow rules 2745~~~~~~~~~~~~~~~~~~~~~ 2746 2747``flow destroy`` destroys one or more rules from their rule ID (as returned 2748by ``flow create``), this command calls ``rte_flow_destroy()`` as many 2749times as necessary:: 2750 2751 flow destroy {port_id} rule {rule_id} [...] 2752 2753If successful, it will show:: 2754 2755 Flow rule #[...] destroyed 2756 2757It does not report anything for rule IDs that do not exist. The usual error 2758message is shown when a rule cannot be destroyed:: 2759 2760 Caught error type [...] ([...]): [...] 2761 2762``flow flush`` destroys all rules on a device and does not take extra 2763arguments. It is bound to ``rte_flow_flush()``:: 2764 2765 flow flush {port_id} 2766 2767Any errors are reported as above. 2768 2769Creating several rules and destroying them:: 2770 2771 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2772 actions queue index 2 / end 2773 Flow rule #0 created 2774 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2775 actions queue index 3 / end 2776 Flow rule #1 created 2777 testpmd> flow destroy 0 rule 0 rule 1 2778 Flow rule #1 destroyed 2779 Flow rule #0 destroyed 2780 testpmd> 2781 2782The same result can be achieved using ``flow flush``:: 2783 2784 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2785 actions queue index 2 / end 2786 Flow rule #0 created 2787 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2788 actions queue index 3 / end 2789 Flow rule #1 created 2790 testpmd> flow flush 0 2791 testpmd> 2792 2793Non-existent rule IDs are ignored:: 2794 2795 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2796 actions queue index 2 / end 2797 Flow rule #0 created 2798 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2799 actions queue index 3 / end 2800 Flow rule #1 created 2801 testpmd> flow destroy 0 rule 42 rule 10 rule 2 2802 testpmd> 2803 testpmd> flow destroy 0 rule 0 2804 Flow rule #0 destroyed 2805 testpmd> 2806 2807Querying flow rules 2808~~~~~~~~~~~~~~~~~~~ 2809 2810``flow query`` queries a specific action of a flow rule having that 2811ability. Such actions collect information that can be reported using this 2812command. It is bound to ``rte_flow_query()``:: 2813 2814 flow query {port_id} {rule_id} {action} 2815 2816If successful, it will display either the retrieved data for known actions 2817or the following message:: 2818 2819 Cannot display result for action type [...] ([...]) 2820 2821Otherwise, it will complain either that the rule does not exist or that some 2822error occurred:: 2823 2824 Flow rule #[...] not found 2825 2826:: 2827 2828 Caught error type [...] ([...]): [...] 2829 2830Currently only the ``count`` action is supported. This action reports the 2831number of packets that hit the flow rule and the total number of bytes. Its 2832output has the following format:: 2833 2834 count: 2835 hits_set: [...] # whether "hits" contains a valid value 2836 bytes_set: [...] # whether "bytes" contains a valid value 2837 hits: [...] # number of packets 2838 bytes: [...] # number of bytes 2839 2840Querying counters for TCPv6 packets redirected to queue 6:: 2841 2842 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end 2843 actions queue index 6 / count / end 2844 Flow rule #4 created 2845 testpmd> flow query 0 4 count 2846 count: 2847 hits_set: 1 2848 bytes_set: 0 2849 hits: 386446 2850 bytes: 0 2851 testpmd> 2852 2853Listing flow rules 2854~~~~~~~~~~~~~~~~~~ 2855 2856``flow list`` lists existing flow rules sorted by priority and optionally 2857filtered by group identifiers:: 2858 2859 flow list {port_id} [group {group_id}] [...] 2860 2861This command only fails with the following message if the device does not 2862exist:: 2863 2864 Invalid port [...] 2865 2866Output consists of a header line followed by a short description of each 2867flow rule, one per line. There is no output at all when no flow rules are 2868configured on the device:: 2869 2870 ID Group Prio Attr Rule 2871 [...] [...] [...] [...] [...] 2872 2873``Attr`` column flags: 2874 2875- ``i`` for ``ingress``. 2876- ``e`` for ``egress``. 2877 2878Creating several flow rules and listing them:: 2879 2880 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2881 actions queue index 6 / end 2882 Flow rule #0 created 2883 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2884 actions queue index 2 / end 2885 Flow rule #1 created 2886 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end 2887 actions rss queues 6 7 8 end / end 2888 Flow rule #2 created 2889 testpmd> flow list 0 2890 ID Group Prio Attr Rule 2891 0 0 0 i- ETH IPV4 => QUEUE 2892 1 0 0 i- ETH IPV6 => QUEUE 2893 2 0 5 i- ETH IPV4 UDP => RSS 2894 testpmd> 2895 2896Rules are sorted by priority (i.e. group ID first, then priority level):: 2897 2898 testpmd> flow list 1 2899 ID Group Prio Attr Rule 2900 0 0 0 i- ETH => COUNT 2901 6 0 500 i- ETH IPV6 TCP => DROP COUNT 2902 5 0 1000 i- ETH IPV6 ICMP => QUEUE 2903 1 24 0 i- ETH IPV4 UDP => QUEUE 2904 4 24 10 i- ETH IPV4 TCP => DROP 2905 3 24 20 i- ETH IPV4 => DROP 2906 2 24 42 i- ETH IPV4 UDP => QUEUE 2907 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE 2908 testpmd> 2909 2910Output can be limited to specific groups:: 2911 2912 testpmd> flow list 1 group 0 group 63 2913 ID Group Prio Attr Rule 2914 0 0 0 i- ETH => COUNT 2915 6 0 500 i- ETH IPV6 TCP => DROP COUNT 2916 5 0 1000 i- ETH IPV6 ICMP => QUEUE 2917 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE 2918 testpmd> 2919 2920Toggling isolated mode 2921~~~~~~~~~~~~~~~~~~~~~~ 2922 2923``flow isolate`` can be used to tell the underlying PMD that ingress traffic 2924must only be injected from the defined flow rules; that no default traffic 2925is expected outside those rules and the driver is free to assign more 2926resources to handle them. It is bound to ``rte_flow_isolate()``:: 2927 2928 flow isolate {port_id} {boolean} 2929 2930If successful, enabling or disabling isolated mode shows either:: 2931 2932 Ingress traffic on port [...] 2933 is now restricted to the defined flow rules 2934 2935Or:: 2936 2937 Ingress traffic on port [...] 2938 is not restricted anymore to the defined flow rules 2939 2940Otherwise, in case of error:: 2941 2942 Caught error type [...] ([...]): [...] 2943 2944Mainly due to its side effects, PMDs supporting this mode may not have the 2945ability to toggle it more than once without reinitializing affected ports 2946first (e.g. by exiting testpmd). 2947 2948Enabling isolated mode:: 2949 2950 testpmd> flow isolate 0 true 2951 Ingress traffic on port 0 is now restricted to the defined flow rules 2952 testpmd> 2953 2954Disabling isolated mode:: 2955 2956 testpmd> flow isolate 0 false 2957 Ingress traffic on port 0 is not restricted anymore to the defined flow rules 2958 testpmd> 2959 2960Sample QinQ flow rules 2961~~~~~~~~~~~~~~~~~~~~~~ 2962 2963Before creating QinQ rule(s) the following commands should be issued to enable QinQ:: 2964 2965 testpmd> port stop 0 2966 testpmd> vlan set qinq on 0 2967 2968The above command sets the inner and outer TPID's to 0x8100. 2969 2970To change the TPID's the following commands should be used:: 2971 2972 testpmd> vlan set outer tpid 0xa100 0 2973 testpmd> vlan set inner tpid 0x9100 0 2974 testpmd> port start 0 2975 2976Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM. 2977 2978:: 2979 2980 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 / 2981 vlan tci is 456 / end actions vf id 1 / queue index 0 / end 2982 Flow rule #0 validated 2983 2984 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 / 2985 vlan tci is 456 / end actions vf id 123 / queue index 0 / end 2986 Flow rule #0 created 2987 2988 testpmd> flow list 0 2989 ID Group Prio Attr Rule 2990 0 0 0 i- ETH VLAN VLAN=>VF QUEUE 2991 2992Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host. 2993 2994:: 2995 2996 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 / 2997 vlan tci is 654 / end actions pf / queue index 0 / end 2998 Flow rule #1 validated 2999 3000 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 / 3001 vlan tci is 654 / end actions pf / queue index 1 / end 3002 Flow rule #1 created 3003 3004 testpmd> flow list 0 3005 ID Group Prio Attr Rule 3006 0 0 0 i- ETH VLAN VLAN=>VF QUEUE 3007 1 0 0 i- ETH VLAN VLAN=>PF QUEUE 3008