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 390ddp get info 391~~~~~~~~~~~~ 392 393Display information about dynamic device personalization (DDP) profile:: 394 395 testpmd> ddp get info (profile_patch) 396 397show vf stats 398~~~~~~~~~~~~~ 399 400Display VF statistics:: 401 402 testpmd> show vf stats (port_id) (vf_id) 403 404clear vf stats 405~~~~~~~~~~~~~~ 406 407Reset VF statistics:: 408 409 testpmd> clear vf stats (port_id) (vf_id) 410 411Configuration Functions 412----------------------- 413 414The testpmd application can be configured from the runtime as well as from the command-line. 415 416This section details the available configuration functions that are available. 417 418.. note:: 419 420 Configuration changes only become active when forwarding is started/restarted. 421 422set default 423~~~~~~~~~~~ 424 425Reset forwarding to the default configuration:: 426 427 testpmd> set default 428 429set verbose 430~~~~~~~~~~~ 431 432Set the debug verbosity level:: 433 434 testpmd> set verbose (level) 435 436Currently the only available levels are 0 (silent except for error) and 1 (fully verbose). 437 438set nbport 439~~~~~~~~~~ 440 441Set the number of ports used by the application: 442 443set nbport (num) 444 445This is equivalent to the ``--nb-ports`` command-line option. 446 447set nbcore 448~~~~~~~~~~ 449 450Set the number of cores used by the application:: 451 452 testpmd> set nbcore (num) 453 454This is equivalent to the ``--nb-cores`` command-line option. 455 456.. note:: 457 458 The number of cores used must not be greater than number of ports used multiplied by the number of queues per port. 459 460set coremask 461~~~~~~~~~~~~ 462 463Set the forwarding cores hexadecimal mask:: 464 465 testpmd> set coremask (mask) 466 467This is equivalent to the ``--coremask`` command-line option. 468 469.. note:: 470 471 The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding. 472 473set portmask 474~~~~~~~~~~~~ 475 476Set the forwarding ports hexadecimal mask:: 477 478 testpmd> set portmask (mask) 479 480This is equivalent to the ``--portmask`` command-line option. 481 482set burst 483~~~~~~~~~ 484 485Set number of packets per burst:: 486 487 testpmd> set burst (num) 488 489This is equivalent to the ``--burst command-line`` option. 490 491When retry is enabled, the transmit delay time and number of retries can also be set:: 492 493 testpmd> set burst tx delay (microseconds) retry (num) 494 495set txpkts 496~~~~~~~~~~ 497 498Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode:: 499 500 testpmd> set txpkts (x[,y]*) 501 502Where x[,y]* represents a CSV list of values, without white space. 503 504set txsplit 505~~~~~~~~~~~ 506 507Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes:: 508 509 testpmd> set txsplit (off|on|rand) 510 511Where: 512 513* ``off`` disable packet copy & split for CSUM mode. 514 515* ``on`` split outgoing packet into multiple segments. Size of each segment 516 and number of segments per packet is determined by ``set txpkts`` command 517 (see above). 518 519* ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments. 520 521set corelist 522~~~~~~~~~~~~ 523 524Set the list of forwarding cores:: 525 526 testpmd> set corelist (x[,y]*) 527 528For example, to change the forwarding cores: 529 530.. code-block:: console 531 532 testpmd> set corelist 3,1 533 testpmd> show config fwd 534 535 io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled 536 Logical Core 3 (socket 0) forwards packets on 1 streams: 537 RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01 538 Logical Core 1 (socket 0) forwards packets on 1 streams: 539 RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00 540 541.. note:: 542 543 The cores are used in the same order as specified on the command line. 544 545set portlist 546~~~~~~~~~~~~ 547 548Set the list of forwarding ports:: 549 550 testpmd> set portlist (x[,y]*) 551 552For example, to change the port forwarding: 553 554.. code-block:: console 555 556 testpmd> set portlist 0,2,1,3 557 testpmd> show config fwd 558 559 io packet forwarding - ports=4 - cores=1 - streams=4 560 Logical Core 3 (socket 0) forwards packets on 4 streams: 561 RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01 562 RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00 563 RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03 564 RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02 565 566set tx loopback 567~~~~~~~~~~~~~~~ 568 569Enable/disable tx loopback:: 570 571 testpmd> set tx loopback (port_id) (on|off) 572 573set drop enable 574~~~~~~~~~~~~~~~ 575 576set drop enable bit for all queues:: 577 578 testpmd> set all queues drop (port_id) (on|off) 579 580set split drop enable (for VF) 581~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 582 583set split drop enable bit for VF from PF:: 584 585 testpmd> set vf split drop (port_id) (vf_id) (on|off) 586 587set mac antispoof (for VF) 588~~~~~~~~~~~~~~~~~~~~~~~~~~ 589 590Set mac antispoof for a VF from the PF:: 591 592 testpmd> set vf mac antispoof (port_id) (vf_id) (on|off) 593 594set macsec offload 595~~~~~~~~~~~~~~~~~~ 596 597Enable/disable MACsec offload:: 598 599 testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off) 600 testpmd> set macsec offload (port_id) off 601 602set macsec sc 603~~~~~~~~~~~~~ 604 605Configure MACsec secure connection (SC):: 606 607 testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi) 608 609.. note:: 610 611 The pi argument is ignored for tx. 612 Check the NIC Datasheet for hardware limits. 613 614set macsec sa 615~~~~~~~~~~~~~ 616 617Configure MACsec secure association (SA):: 618 619 testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key) 620 621.. note:: 622 623 The IDX value must be 0 or 1. 624 Check the NIC Datasheet for hardware limits. 625 626set broadcast mode (for VF) 627~~~~~~~~~~~~~~~~~~~~~~~~~~~ 628 629Set broadcast mode for a VF from the PF:: 630 631 testpmd> set vf broadcast (port_id) (vf_id) (on|off) 632 633vlan set strip 634~~~~~~~~~~~~~~ 635 636Set the VLAN strip on a port:: 637 638 testpmd> vlan set strip (on|off) (port_id) 639 640vlan set stripq 641~~~~~~~~~~~~~~~ 642 643Set the VLAN strip for a queue on a port:: 644 645 testpmd> vlan set stripq (on|off) (port_id,queue_id) 646 647vlan set stripq (for VF) 648~~~~~~~~~~~~~~~~~~~~~~~~ 649 650Set VLAN strip for all queues in a pool for a VF from the PF:: 651 652 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off) 653 654vlan set insert (for VF) 655~~~~~~~~~~~~~~~~~~~~~~~~ 656 657Set VLAN insert for a VF from the PF:: 658 659 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id) 660 661vlan set tag (for VF) 662~~~~~~~~~~~~~~~~~~~~~ 663 664Set VLAN tag for a VF from the PF:: 665 666 testpmd> set vf vlan tag (port_id) (vf_id) (on|off) 667 668vlan set antispoof (for VF) 669~~~~~~~~~~~~~~~~~~~~~~~~~~~ 670 671Set VLAN antispoof for a VF from the PF:: 672 673 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off) 674 675vlan set filter 676~~~~~~~~~~~~~~~ 677 678Set the VLAN filter on a port:: 679 680 testpmd> vlan set filter (on|off) (port_id) 681 682vlan set qinq 683~~~~~~~~~~~~~ 684 685Set the VLAN QinQ (extended queue in queue) on for a port:: 686 687 testpmd> vlan set qinq (on|off) (port_id) 688 689vlan set tpid 690~~~~~~~~~~~~~ 691 692Set the inner or outer VLAN TPID for packet filtering on a port:: 693 694 testpmd> vlan set (inner|outer) tpid (value) (port_id) 695 696.. note:: 697 698 TPID value must be a 16-bit number (value <= 65536). 699 700rx_vlan add 701~~~~~~~~~~~ 702 703Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID:: 704 705 testpmd> rx_vlan add (vlan_id|all) (port_id) 706 707.. note:: 708 709 VLAN filter must be set on that port. VLAN ID < 4096. 710 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries 711 in VFTA table. This is important if enabling all vlan_ids. 712 713rx_vlan rm 714~~~~~~~~~~ 715 716Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID:: 717 718 testpmd> rx_vlan rm (vlan_id|all) (port_id) 719 720rx_vlan add (for VF) 721~~~~~~~~~~~~~~~~~~~~ 722 723Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID:: 724 725 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask) 726 727rx_vlan rm (for VF) 728~~~~~~~~~~~~~~~~~~~ 729 730Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID:: 731 732 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask) 733 734tunnel_filter add 735~~~~~~~~~~~~~~~~~ 736 737Add a tunnel filter on a port:: 738 739 testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \ 740 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\ 741 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id) 742 743The available information categories are: 744 745* ``vxlan``: Set tunnel type as VXLAN. 746 747* ``nvgre``: Set tunnel type as NVGRE. 748 749* ``ipingre``: Set tunnel type as IP-in-GRE. 750 751* ``imac-ivlan``: Set filter type as Inner MAC and VLAN. 752 753* ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID. 754 755* ``imac-tenid``: Set filter type as Inner MAC and tenant ID. 756 757* ``imac``: Set filter type as Inner MAC. 758 759* ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID. 760 761* ``oip``: Set filter type as Outer IP. 762 763* ``iip``: Set filter type as Inner IP. 764 765Example:: 766 767 testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \ 768 192.168.2.2 0 ipingre oip 1 1 769 770 Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP. 771 772tunnel_filter remove 773~~~~~~~~~~~~~~~~~~~~ 774 775Remove a tunnel filter on a port:: 776 777 testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \ 778 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\ 779 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id) 780 781rx_vxlan_port add 782~~~~~~~~~~~~~~~~~ 783 784Add an UDP port for VXLAN packet filter on a port:: 785 786 testpmd> rx_vxlan_port add (udp_port) (port_id) 787 788rx_vxlan_port remove 789~~~~~~~~~~~~~~~~~~~~ 790 791Remove an UDP port for VXLAN packet filter on a port:: 792 793 testpmd> rx_vxlan_port rm (udp_port) (port_id) 794 795tx_vlan set 796~~~~~~~~~~~ 797 798Set hardware insertion of VLAN IDs in packets sent on a port:: 799 800 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer] 801 802For example, set a single VLAN ID (5) insertion on port 0:: 803 804 tx_vlan set 0 5 805 806Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1:: 807 808 tx_vlan set 1 2 3 809 810 811tx_vlan set pvid 812~~~~~~~~~~~~~~~~ 813 814Set port based hardware insertion of VLAN ID in packets sent on a port:: 815 816 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off) 817 818tx_vlan reset 819~~~~~~~~~~~~~ 820 821Disable hardware insertion of a VLAN header in packets sent on a port:: 822 823 testpmd> tx_vlan reset (port_id) 824 825csum set 826~~~~~~~~ 827 828Select hardware or software calculation of the checksum when 829transmitting a packet using the ``csum`` forwarding engine:: 830 831 testpmd> csum set (ip|udp|tcp|sctp|outer-ip) (hw|sw) (port_id) 832 833Where: 834 835* ``ip|udp|tcp|sctp`` always relate to the inner layer. 836 837* ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized 838 as a tunnel packet by the forwarding engine (vxlan, gre and ipip are 839 supported). See also the ``csum parse-tunnel`` command. 840 841.. note:: 842 843 Check the NIC Datasheet for hardware limits. 844 845csum parse-tunnel 846~~~~~~~~~~~~~~~~~ 847 848Define how tunneled packets should be handled by the csum forward 849engine:: 850 851 testpmd> csum parse-tunnel (on|off) (tx_port_id) 852 853If enabled, the csum forward engine will try to recognize supported 854tunnel headers (vxlan, gre, ipip). 855 856If disabled, treat tunnel packets as non-tunneled packets (a inner 857header is handled as a packet payload). 858 859.. note:: 860 861 The port argument is the TX port like in the ``csum set`` command. 862 863Example: 864 865Consider a packet in packet like the following:: 866 867 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in 868 869* If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set`` 870 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the 871 ``outer-ip parameter`` relates to the outer headers (here ``ipv4_out``). 872 873* If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set`` 874 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``. 875 876csum show 877~~~~~~~~~ 878 879Display tx checksum offload configuration:: 880 881 testpmd> csum show (port_id) 882 883tso set 884~~~~~~~ 885 886Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine:: 887 888 testpmd> tso set (segsize) (port_id) 889 890.. note:: 891 892 Check the NIC datasheet for hardware limits. 893 894tso show 895~~~~~~~~ 896 897Display the status of TCP Segmentation Offload:: 898 899 testpmd> tso show (port_id) 900 901mac_addr add 902~~~~~~~~~~~~ 903 904Add an alternative MAC address to a port:: 905 906 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX) 907 908mac_addr remove 909~~~~~~~~~~~~~~~ 910 911Remove a MAC address from a port:: 912 913 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX) 914 915mac_addr add (for VF) 916~~~~~~~~~~~~~~~~~~~~~ 917 918Add an alternative MAC address for a VF to a port:: 919 920 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX) 921 922mac_addr set 923~~~~~~~~~~~~ 924 925Set the default MAC address for a port:: 926 927 testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX) 928 929mac_addr set (for VF) 930~~~~~~~~~~~~~~~~~~~~~ 931 932Set the MAC address for a VF from the PF:: 933 934 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX) 935 936set port-uta 937~~~~~~~~~~~~ 938 939Set the unicast hash filter(s) on/off for a port:: 940 941 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off) 942 943set promisc 944~~~~~~~~~~~ 945 946Set the promiscuous mode on for a port or for all ports. 947In promiscuous mode packets are not dropped if they aren't for the specified MAC address:: 948 949 testpmd> set promisc (port_id|all) (on|off) 950 951set allmulti 952~~~~~~~~~~~~ 953 954Set the allmulti mode for a port or for all ports:: 955 956 testpmd> set allmulti (port_id|all) (on|off) 957 958Same as the ifconfig (8) option. Controls how multicast packets are handled. 959 960set promisc (for VF) 961~~~~~~~~~~~~~~~~~~~~ 962 963Set the unicast promiscuous mode for a VF from PF. 964It's supported by Intel i40e NICs now. 965In promiscuous mode packets are not dropped if they aren't for the specified MAC address:: 966 967 testpmd> set vf promisc (port_id) (vf_id) (on|off) 968 969set allmulticast (for VF) 970~~~~~~~~~~~~~~~~~~~~~~~~~ 971 972Set the multicast promiscuous mode for a VF from PF. 973It's supported by Intel i40e NICs now. 974In promiscuous mode packets are not dropped if they aren't for the specified MAC address:: 975 976 testpmd> set vf allmulti (port_id) (vf_id) (on|off) 977 978set tx max bandwidth (for VF) 979~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 980 981Set TX max absolute bandwidth (Mbps) for a VF from PF:: 982 983 testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth) 984 985set tc tx min bandwidth (for VF) 986~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 987 988Set all TCs' TX min relative bandwidth (%) for a VF from PF:: 989 990 testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...) 991 992set tc tx max bandwidth (for VF) 993~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 994 995Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF:: 996 997 testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth) 998 999set tc strict link priority mode 1000~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1001 1002Set some TCs' strict link priority mode on a physical port:: 1003 1004 testpmd> set tx strict-link-priority (port_id) (tc_bitmap) 1005 1006set tc tx min bandwidth 1007~~~~~~~~~~~~~~~~~~~~~~~ 1008 1009Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs:: 1010 1011 testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...) 1012 1013set flow_ctrl rx 1014~~~~~~~~~~~~~~~~ 1015 1016Set the link flow control parameter on a port:: 1017 1018 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \ 1019 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \ 1020 autoneg (on|off) (port_id) 1021 1022Where: 1023 1024* ``high_water`` (integer): High threshold value to trigger XOFF. 1025 1026* ``low_water`` (integer): Low threshold value to trigger XON. 1027 1028* ``pause_time`` (integer): Pause quota in the Pause frame. 1029 1030* ``send_xon`` (0/1): Send XON frame. 1031 1032* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames. 1033 1034* ``autoneg``: Change the auto-negotiation parameter. 1035 1036set pfc_ctrl rx 1037~~~~~~~~~~~~~~~ 1038 1039Set the priority flow control parameter on a port:: 1040 1041 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \ 1042 (pause_time) (priority) (port_id) 1043 1044Where: 1045 1046* ``high_water`` (integer): High threshold value. 1047 1048* ``low_water`` (integer): Low threshold value. 1049 1050* ``pause_time`` (integer): Pause quota in the Pause frame. 1051 1052* ``priority`` (0-7): VLAN User Priority. 1053 1054set stat_qmap 1055~~~~~~~~~~~~~ 1056 1057Set statistics mapping (qmapping 0..15) for RX/TX queue on port:: 1058 1059 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping) 1060 1061For example, to set rx queue 2 on port 0 to mapping 5:: 1062 1063 testpmd>set stat_qmap rx 0 2 5 1064 1065set port - rx/tx (for VF) 1066~~~~~~~~~~~~~~~~~~~~~~~~~ 1067 1068Set VF receive/transmit from a port:: 1069 1070 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off) 1071 1072set port - mac address filter (for VF) 1073~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1074 1075Add/Remove unicast or multicast MAC addr filter for a VF:: 1076 1077 testpmd> set port (port_id) vf (vf_id) (mac_addr) \ 1078 (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off) 1079 1080set port - rx mode(for VF) 1081~~~~~~~~~~~~~~~~~~~~~~~~~~ 1082 1083Set the VF receive mode of a port:: 1084 1085 testpmd> set port (port_id) vf (vf_id) \ 1086 rxmode (AUPE|ROPE|BAM|MPE) (on|off) 1087 1088The available receive modes are: 1089 1090* ``AUPE``: Accepts untagged VLAN. 1091 1092* ``ROPE``: Accepts unicast hash. 1093 1094* ``BAM``: Accepts broadcast packets. 1095 1096* ``MPE``: Accepts all multicast packets. 1097 1098set port - tx_rate (for Queue) 1099~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1100 1101Set TX rate limitation for a queue on a port:: 1102 1103 testpmd> set port (port_id) queue (queue_id) rate (rate_value) 1104 1105set port - tx_rate (for VF) 1106~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1107 1108Set TX rate limitation for queues in VF on a port:: 1109 1110 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask) 1111 1112set port - mirror rule 1113~~~~~~~~~~~~~~~~~~~~~~ 1114 1115Set pool or vlan type mirror rule for a port:: 1116 1117 testpmd> set port (port_id) mirror-rule (rule_id) \ 1118 (pool-mirror-up|pool-mirror-down|vlan-mirror) \ 1119 (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off) 1120 1121Set link mirror rule for a port:: 1122 1123 testpmd> set port (port_id) mirror-rule (rule_id) \ 1124 (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off) 1125 1126For example to enable mirror traffic with vlan 0,1 to pool 0:: 1127 1128 set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on 1129 1130reset port - mirror rule 1131~~~~~~~~~~~~~~~~~~~~~~~~ 1132 1133Reset a mirror rule for a port:: 1134 1135 testpmd> reset port (port_id) mirror-rule (rule_id) 1136 1137set flush_rx 1138~~~~~~~~~~~~ 1139 1140Set the flush on RX streams before forwarding. 1141The default is flush ``on``. 1142Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams:: 1143 1144 testpmd> set flush_rx off 1145 1146set bypass mode 1147~~~~~~~~~~~~~~~ 1148 1149Set the bypass mode for the lowest port on bypass enabled NIC:: 1150 1151 testpmd> set bypass mode (normal|bypass|isolate) (port_id) 1152 1153set bypass event 1154~~~~~~~~~~~~~~~~ 1155 1156Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled:: 1157 1158 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \ 1159 mode (normal|bypass|isolate) (port_id) 1160 1161Where: 1162 1163* ``timeout``: Enable bypass after watchdog timeout. 1164 1165* ``os_on``: Enable bypass when OS/board is powered on. 1166 1167* ``os_off``: Enable bypass when OS/board is powered off. 1168 1169* ``power_on``: Enable bypass when power supply is turned on. 1170 1171* ``power_off``: Enable bypass when power supply is turned off. 1172 1173 1174set bypass timeout 1175~~~~~~~~~~~~~~~~~~ 1176 1177Set the bypass watchdog timeout to ``n`` seconds where 0 = instant:: 1178 1179 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32) 1180 1181show bypass config 1182~~~~~~~~~~~~~~~~~~ 1183 1184Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC:: 1185 1186 testpmd> show bypass config (port_id) 1187 1188set link up 1189~~~~~~~~~~~ 1190 1191Set link up for a port:: 1192 1193 testpmd> set link-up port (port id) 1194 1195set link down 1196~~~~~~~~~~~~~ 1197 1198Set link down for a port:: 1199 1200 testpmd> set link-down port (port id) 1201 1202E-tag set 1203~~~~~~~~~ 1204 1205Enable E-tag insertion for a VF on a port:: 1206 1207 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id) 1208 1209Disable E-tag insertion for a VF on a port:: 1210 1211 testpmd> E-tag set insertion off port (port_id) vf (vf_id) 1212 1213Enable/disable E-tag stripping on a port:: 1214 1215 testpmd> E-tag set stripping (on|off) port (port_id) 1216 1217Enable/disable E-tag based forwarding on a port:: 1218 1219 testpmd> E-tag set forwarding (on|off) port (port_id) 1220 1221Add an E-tag forwarding filter on a port:: 1222 1223 testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id) 1224 1225Delete an E-tag forwarding filter on a port:: 1226 testpmd> E-tag set filter del e-tag-id (value) port (port_id) 1227 1228ddp add 1229~~~~~~~ 1230 1231Load a dynamic device personalization (DDP) package:: 1232 1233 testpmd> ddp add (port_id) (package_path[,output_path]) 1234 1235ddp del 1236~~~~~~~ 1237 1238Delete a dynamic device personalization package:: 1239 1240 testpmd> ddp del (port_id) (package_path) 1241 1242ptype mapping 1243~~~~~~~~~~~~~ 1244 1245List all items from the ptype mapping table:: 1246 1247 testpmd> ptype mapping get (port_id) (valid_only) 1248 1249Where: 1250 1251* ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0). 1252 1253Replace a specific or a group of software defined ptype with a new one:: 1254 1255 testpmd> ptype mapping replace (port_id) (target) (mask) (pkt_type) 1256 1257where: 1258 1259* ``target``: A specific software ptype or a mask to represent a group of software ptypes. 1260 1261* ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1). 1262 1263* ``pkt_type``: The new software ptype to replace the old ones. 1264 1265Update hardware defined ptype to software defined packet type mapping table:: 1266 1267 testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype) 1268 1269where: 1270 1271* ``hw_ptype``: hardware ptype as the index of the ptype mapping table. 1272 1273* ``sw_ptype``: software ptype as the value of the ptype mapping table. 1274 1275Reset ptype mapping table:: 1276 1277 testpmd> ptype mapping reset (port_id) 1278 1279Port Functions 1280-------------- 1281 1282The following sections show functions for configuring ports. 1283 1284.. note:: 1285 1286 Port configuration changes only become active when forwarding is started/restarted. 1287 1288port attach 1289~~~~~~~~~~~ 1290 1291Attach a port specified by pci address or virtual device args:: 1292 1293 testpmd> port attach (identifier) 1294 1295To attach a new pci device, the device should be recognized by kernel first. 1296Then it should be moved under DPDK management. 1297Finally the port can be attached to testpmd. 1298 1299For example, to move a pci device using ixgbe under DPDK management: 1300 1301.. code-block:: console 1302 1303 # Check the status of the available devices. 1304 ./usertools/dpdk-devbind.py --status 1305 1306 Network devices using DPDK-compatible driver 1307 ============================================ 1308 <none> 1309 1310 Network devices using kernel driver 1311 =================================== 1312 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused= 1313 1314 1315 # Bind the device to igb_uio. 1316 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0 1317 1318 1319 # Recheck the status of the devices. 1320 ./usertools/dpdk-devbind.py --status 1321 Network devices using DPDK-compatible driver 1322 ============================================ 1323 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused= 1324 1325To attach a port created by virtual device, above steps are not needed. 1326 1327For example, to attach a port whose pci address is 0000:0a:00.0. 1328 1329.. code-block:: console 1330 1331 testpmd> port attach 0000:0a:00.0 1332 Attaching a new port... 1333 EAL: PCI device 0000:0a:00.0 on NUMA socket -1 1334 EAL: probe driver: 8086:10fb rte_ixgbe_pmd 1335 EAL: PCI memory mapped at 0x7f83bfa00000 1336 EAL: PCI memory mapped at 0x7f83bfa80000 1337 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5 1338 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb 1339 Port 0 is attached. Now total ports is 1 1340 Done 1341 1342For example, to attach a port created by pcap PMD. 1343 1344.. code-block:: console 1345 1346 testpmd> port attach net_pcap0 1347 Attaching a new port... 1348 PMD: Initializing pmd_pcap for net_pcap0 1349 PMD: Creating pcap-backed ethdev on numa socket 0 1350 Port 0 is attached. Now total ports is 1 1351 Done 1352 1353In this case, identifier is ``net_pcap0``. 1354This identifier format is the same as ``--vdev`` format of DPDK applications. 1355 1356For example, to re-attach a bonded port which has been previously detached, 1357the mode and slave parameters must be given. 1358 1359.. code-block:: console 1360 1361 testpmd> port attach net_bond_0,mode=0,slave=1 1362 Attaching a new port... 1363 EAL: Initializing pmd_bond for net_bond_0 1364 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0. 1365 Port 0 is attached. Now total ports is 1 1366 Done 1367 1368 1369port detach 1370~~~~~~~~~~~ 1371 1372Detach a specific port:: 1373 1374 testpmd> port detach (port_id) 1375 1376Before detaching a port, the port should be stopped and closed. 1377 1378For example, to detach a pci device port 0. 1379 1380.. code-block:: console 1381 1382 testpmd> port stop 0 1383 Stopping ports... 1384 Done 1385 testpmd> port close 0 1386 Closing ports... 1387 Done 1388 1389 testpmd> port detach 0 1390 Detaching a port... 1391 EAL: PCI device 0000:0a:00.0 on NUMA socket -1 1392 EAL: remove driver: 8086:10fb rte_ixgbe_pmd 1393 EAL: PCI memory unmapped at 0x7f83bfa00000 1394 EAL: PCI memory unmapped at 0x7f83bfa80000 1395 Done 1396 1397 1398For example, to detach a virtual device port 0. 1399 1400.. code-block:: console 1401 1402 testpmd> port stop 0 1403 Stopping ports... 1404 Done 1405 testpmd> port close 0 1406 Closing ports... 1407 Done 1408 1409 testpmd> port detach 0 1410 Detaching a port... 1411 PMD: Closing pcap ethdev on numa socket 0 1412 Port 'net_pcap0' is detached. Now total ports is 0 1413 Done 1414 1415To remove a pci device completely from the system, first detach the port from testpmd. 1416Then the device should be moved under kernel management. 1417Finally the device can be removed using kernel pci hotplug functionality. 1418 1419For example, to move a pci device under kernel management: 1420 1421.. code-block:: console 1422 1423 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0 1424 1425 ./usertools/dpdk-devbind.py --status 1426 1427 Network devices using DPDK-compatible driver 1428 ============================================ 1429 <none> 1430 1431 Network devices using kernel driver 1432 =================================== 1433 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio 1434 1435To remove a port created by a virtual device, above steps are not needed. 1436 1437port start 1438~~~~~~~~~~ 1439 1440Start all ports or a specific port:: 1441 1442 testpmd> port start (port_id|all) 1443 1444port stop 1445~~~~~~~~~ 1446 1447Stop all ports or a specific port:: 1448 1449 testpmd> port stop (port_id|all) 1450 1451port close 1452~~~~~~~~~~ 1453 1454Close all ports or a specific port:: 1455 1456 testpmd> port close (port_id|all) 1457 1458port start/stop queue 1459~~~~~~~~~~~~~~~~~~~~~ 1460 1461Start/stop a rx/tx queue on a specific port:: 1462 1463 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop) 1464 1465Only take effect when port is started. 1466 1467port config - speed 1468~~~~~~~~~~~~~~~~~~~ 1469 1470Set the speed and duplex mode for all ports or a specific port:: 1471 1472 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \ 1473 duplex (half|full|auto) 1474 1475port config - queues/descriptors 1476~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1477 1478Set number of queues/descriptors for rxq, txq, rxd and txd:: 1479 1480 testpmd> port config all (rxq|txq|rxd|txd) (value) 1481 1482This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options. 1483 1484port config - max-pkt-len 1485~~~~~~~~~~~~~~~~~~~~~~~~~ 1486 1487Set the maximum packet length:: 1488 1489 testpmd> port config all max-pkt-len (value) 1490 1491This is equivalent to the ``--max-pkt-len`` command-line option. 1492 1493port config - CRC Strip 1494~~~~~~~~~~~~~~~~~~~~~~~ 1495 1496Set hardware CRC stripping on or off for all ports:: 1497 1498 testpmd> port config all crc-strip (on|off) 1499 1500CRC stripping is on by default. 1501 1502The ``off`` option is equivalent to the ``--disable-crc-strip`` command-line option. 1503 1504port config - scatter 1505~~~~~~~~~~~~~~~~~~~~~~~ 1506 1507Set RX scatter mode on or off for all ports:: 1508 1509 testpmd> port config all scatter (on|off) 1510 1511RX scatter mode is off by default. 1512 1513The ``on`` option is equivalent to the ``--enable-scatter`` command-line option. 1514 1515port config - TX queue flags 1516~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1517 1518Set a hexadecimal bitmap of TX queue flags for all ports:: 1519 1520 testpmd> port config all txqflags value 1521 1522This command is equivalent to the ``--txqflags`` command-line option. 1523 1524port config - RX Checksum 1525~~~~~~~~~~~~~~~~~~~~~~~~~ 1526 1527Set hardware RX checksum offload to on or off for all ports:: 1528 1529 testpmd> port config all rx-cksum (on|off) 1530 1531Checksum offload is off by default. 1532 1533The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option. 1534 1535port config - VLAN 1536~~~~~~~~~~~~~~~~~~ 1537 1538Set hardware VLAN on or off for all ports:: 1539 1540 testpmd> port config all hw-vlan (on|off) 1541 1542Hardware VLAN is on by default. 1543 1544The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option. 1545 1546port config - VLAN filter 1547~~~~~~~~~~~~~~~~~~~~~~~~~ 1548 1549Set hardware VLAN filter on or off for all ports:: 1550 1551 testpmd> port config all hw-vlan-filter (on|off) 1552 1553Hardware VLAN filter is on by default. 1554 1555The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option. 1556 1557port config - VLAN strip 1558~~~~~~~~~~~~~~~~~~~~~~~~ 1559 1560Set hardware VLAN strip on or off for all ports:: 1561 1562 testpmd> port config all hw-vlan-strip (on|off) 1563 1564Hardware VLAN strip is on by default. 1565 1566The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option. 1567 1568port config - VLAN extend 1569~~~~~~~~~~~~~~~~~~~~~~~~~ 1570 1571Set hardware VLAN extend on or off for all ports:: 1572 1573 testpmd> port config all hw-vlan-extend (on|off) 1574 1575Hardware VLAN extend is off by default. 1576 1577The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option. 1578 1579port config - Drop Packets 1580~~~~~~~~~~~~~~~~~~~~~~~~~~ 1581 1582Set packet drop for packets with no descriptors on or off for all ports:: 1583 1584 testpmd> port config all drop-en (on|off) 1585 1586Packet dropping for packets with no descriptors is off by default. 1587 1588The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option. 1589 1590port config - RSS 1591~~~~~~~~~~~~~~~~~ 1592 1593Set the RSS (Receive Side Scaling) mode on or off:: 1594 1595 testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none) 1596 1597RSS is on by default. 1598 1599The ``none`` option is equivalent to the ``--disable-rss`` command-line option. 1600 1601port config - RSS Reta 1602~~~~~~~~~~~~~~~~~~~~~~ 1603 1604Set the RSS (Receive Side Scaling) redirection table:: 1605 1606 testpmd> port config all rss reta (hash,queue)[,(hash,queue)] 1607 1608port config - DCB 1609~~~~~~~~~~~~~~~~~ 1610 1611Set the DCB mode for an individual port:: 1612 1613 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off) 1614 1615The traffic class should be 4 or 8. 1616 1617port config - Burst 1618~~~~~~~~~~~~~~~~~~~ 1619 1620Set the number of packets per burst:: 1621 1622 testpmd> port config all burst (value) 1623 1624This is equivalent to the ``--burst`` command-line option. 1625 1626port config - Threshold 1627~~~~~~~~~~~~~~~~~~~~~~~ 1628 1629Set thresholds for TX/RX queues:: 1630 1631 testpmd> port config all (threshold) (value) 1632 1633Where the threshold type can be: 1634 1635* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255. 1636 1637* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255. 1638 1639* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255. 1640 1641* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255. 1642 1643* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255. 1644 1645* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255. 1646 1647* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd. 1648 1649* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd. 1650 1651* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd. 1652 1653These threshold options are also available from the command-line. 1654 1655port config - E-tag 1656~~~~~~~~~~~~~~~~~~~ 1657 1658Set the value of ether-type for E-tag:: 1659 1660 testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value) 1661 1662Enable/disable the E-tag support:: 1663 1664 testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable) 1665 1666 1667Link Bonding Functions 1668---------------------- 1669 1670The Link Bonding functions make it possible to dynamically create and 1671manage link bonding devices from within testpmd interactive prompt. 1672 1673create bonded device 1674~~~~~~~~~~~~~~~~~~~~ 1675 1676Create a new bonding device:: 1677 1678 testpmd> create bonded device (mode) (socket) 1679 1680For example, to create a bonded device in mode 1 on socket 0:: 1681 1682 testpmd> create bonded 1 0 1683 created new bonded device (port X) 1684 1685add bonding slave 1686~~~~~~~~~~~~~~~~~ 1687 1688Adds Ethernet device to a Link Bonding device:: 1689 1690 testpmd> add bonding slave (slave id) (port id) 1691 1692For example, to add Ethernet device (port 6) to a Link Bonding device (port 10):: 1693 1694 testpmd> add bonding slave 6 10 1695 1696 1697remove bonding slave 1698~~~~~~~~~~~~~~~~~~~~ 1699 1700Removes an Ethernet slave device from a Link Bonding device:: 1701 1702 testpmd> remove bonding slave (slave id) (port id) 1703 1704For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10):: 1705 1706 testpmd> remove bonding slave 6 10 1707 1708set bonding mode 1709~~~~~~~~~~~~~~~~ 1710 1711Set the Link Bonding mode of a Link Bonding device:: 1712 1713 testpmd> set bonding mode (value) (port id) 1714 1715For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3):: 1716 1717 testpmd> set bonding mode 3 10 1718 1719set bonding primary 1720~~~~~~~~~~~~~~~~~~~ 1721 1722Set an Ethernet slave device as the primary device on a Link Bonding device:: 1723 1724 testpmd> set bonding primary (slave id) (port id) 1725 1726For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10):: 1727 1728 testpmd> set bonding primary 6 10 1729 1730set bonding mac 1731~~~~~~~~~~~~~~~ 1732 1733Set the MAC address of a Link Bonding device:: 1734 1735 testpmd> set bonding mac (port id) (mac) 1736 1737For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01:: 1738 1739 testpmd> set bonding mac 10 00:00:00:00:00:01 1740 1741set bonding xmit_balance_policy 1742~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1743 1744Set the transmission policy for a Link Bonding device when it is in Balance XOR mode:: 1745 1746 testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34) 1747 1748For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports):: 1749 1750 testpmd> set bonding xmit_balance_policy 10 l34 1751 1752 1753set bonding mon_period 1754~~~~~~~~~~~~~~~~~~~~~~ 1755 1756Set the link status monitoring polling period in milliseconds for a bonding device. 1757 1758This adds support for PMD slave devices which do not support link status interrupts. 1759When the mon_period is set to a value greater than 0 then all PMD's which do not support 1760link status ISR will be queried every polling interval to check if their link status has changed:: 1761 1762 testpmd> set bonding mon_period (port_id) (value) 1763 1764For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms:: 1765 1766 testpmd> set bonding mon_period 5 150 1767 1768 1769set bonding lacp dedicated_queue 1770~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1771 1772Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic 1773when in mode 4 (link-aggregration-802.3ad) 1774 1775 testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable) 1776 1777 1778show bonding config 1779~~~~~~~~~~~~~~~~~~~ 1780 1781Show the current configuration of a Link Bonding device:: 1782 1783 testpmd> show bonding config (port id) 1784 1785For example, 1786to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4) 1787in balance mode with a transmission policy of layer 2+3:: 1788 1789 testpmd> show bonding config 9 1790 Bonding mode: 2 1791 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23 1792 Slaves (3): [1 3 4] 1793 Active Slaves (3): [1 3 4] 1794 Primary: [3] 1795 1796 1797Register Functions 1798------------------ 1799 1800The Register Functions can be used to read from and write to registers on the network card referenced by a port number. 1801This is mainly useful for debugging purposes. 1802Reference should be made to the appropriate datasheet for the network card for details on the register addresses 1803and fields that can be accessed. 1804 1805read reg 1806~~~~~~~~ 1807 1808Display the value of a port register:: 1809 1810 testpmd> read reg (port_id) (address) 1811 1812For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller:: 1813 1814 testpmd> read reg 0 0xEE00 1815 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241) 1816 1817read regfield 1818~~~~~~~~~~~~~ 1819 1820Display a port register bit field:: 1821 1822 testpmd> read regfield (port_id) (address) (bit_x) (bit_y) 1823 1824For example, reading the lowest two bits from the register in the example above:: 1825 1826 testpmd> read regfield 0 0xEE00 0 1 1827 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1) 1828 1829read regbit 1830~~~~~~~~~~~ 1831 1832Display a single port register bit:: 1833 1834 testpmd> read regbit (port_id) (address) (bit_x) 1835 1836For example, reading the lowest bit from the register in the example above:: 1837 1838 testpmd> read regbit 0 0xEE00 0 1839 port 0 PCI register at offset 0xEE00: bit 0=1 1840 1841write reg 1842~~~~~~~~~ 1843 1844Set the value of a port register:: 1845 1846 testpmd> write reg (port_id) (address) (value) 1847 1848For example, to clear a register:: 1849 1850 testpmd> write reg 0 0xEE00 0x0 1851 port 0 PCI register at offset 0xEE00: 0x00000000 (0) 1852 1853write regfield 1854~~~~~~~~~~~~~~ 1855 1856Set bit field of a port register:: 1857 1858 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value) 1859 1860For example, writing to the register cleared in the example above:: 1861 1862 testpmd> write regfield 0 0xEE00 0 1 2 1863 port 0 PCI register at offset 0xEE00: 0x00000002 (2) 1864 1865write regbit 1866~~~~~~~~~~~~ 1867 1868Set single bit value of a port register:: 1869 1870 testpmd> write regbit (port_id) (address) (bit_x) (value) 1871 1872For example, to set the high bit in the register from the example above:: 1873 1874 testpmd> write regbit 0 0xEE00 31 1 1875 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658) 1876 1877 1878Filter Functions 1879---------------- 1880 1881This section details the available filter functions that are available. 1882 1883Note these functions interface the deprecated legacy filtering framework, 1884superseded by *rte_flow*. See `Flow rules management`_. 1885 1886ethertype_filter 1887~~~~~~~~~~~~~~~~~~~~ 1888 1889Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue:: 1890 1891 ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \ 1892 ethertype (ether_type) (drop|fwd) queue (queue_id) 1893 1894The available information parameters are: 1895 1896* ``port_id``: The port which the Ethertype filter assigned on. 1897 1898* ``mac_addr``: Compare destination mac address. 1899 1900* ``mac_ignr``: Ignore destination mac address match. 1901 1902* ``mac_address``: Destination mac address to match. 1903 1904* ``ether_type``: The EtherType value want to match, 1905 for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid. 1906 1907* ``queue_id``: The receive queue associated with this EtherType filter. 1908 It is meaningless when deleting or dropping. 1909 1910Example, to add/remove an ethertype filter rule:: 1911 1912 testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \ 1913 ethertype 0x0806 fwd queue 3 1914 1915 testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \ 1916 ethertype 0x0806 fwd queue 3 1917 19182tuple_filter 1919~~~~~~~~~~~~~~~~~ 1920 1921Add or delete a 2-tuple filter, 1922which identifies packets by specific protocol and destination TCP/UDP port 1923and forwards packets into one of the receive queues:: 1924 1925 2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \ 1926 protocol (protocol_value) mask (mask_value) \ 1927 tcp_flags (tcp_flags_value) priority (prio_value) \ 1928 queue (queue_id) 1929 1930The available information parameters are: 1931 1932* ``port_id``: The port which the 2-tuple filter assigned on. 1933 1934* ``dst_port_value``: Destination port in L4. 1935 1936* ``protocol_value``: IP L4 protocol. 1937 1938* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate. 1939 1940* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP). 1941 1942* ``prio_value``: Priority of this filter. 1943 1944* ``queue_id``: The receive queue associated with this 2-tuple filter. 1945 1946Example, to add/remove an 2tuple filter rule:: 1947 1948 testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \ 1949 tcp_flags 0x02 priority 3 queue 3 1950 1951 testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \ 1952 tcp_flags 0x02 priority 3 queue 3 1953 19545tuple_filter 1955~~~~~~~~~~~~~~~~~ 1956 1957Add or delete a 5-tuple filter, 1958which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port) 1959and routes packets into one of the receive queues:: 1960 1961 5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \ 1962 (src_address) dst_port (dst_port_value) \ 1963 src_port (src_port_value) protocol (protocol_value) \ 1964 mask (mask_value) tcp_flags (tcp_flags_value) \ 1965 priority (prio_value) queue (queue_id) 1966 1967The available information parameters are: 1968 1969* ``port_id``: The port which the 5-tuple filter assigned on. 1970 1971* ``dst_address``: Destination IP address. 1972 1973* ``src_address``: Source IP address. 1974 1975* ``dst_port_value``: TCP/UDP destination port. 1976 1977* ``src_port_value``: TCP/UDP source port. 1978 1979* ``protocol_value``: L4 protocol. 1980 1981* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate 1982 1983* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP). 1984 1985* ``prio_value``: The priority of this filter. 1986 1987* ``queue_id``: The receive queue associated with this 5-tuple filter. 1988 1989Example, to add/remove an 5tuple filter rule:: 1990 1991 testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \ 1992 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \ 1993 flags 0x0 priority 3 queue 3 1994 1995 testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \ 1996 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \ 1997 flags 0x0 priority 3 queue 3 1998 1999syn_filter 2000~~~~~~~~~~ 2001 2002Using the SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue:: 2003 2004 syn_filter (port_id) (add|del) priority (high|low) queue (queue_id) 2005 2006The available information parameters are: 2007 2008* ``port_id``: The port which the SYN filter assigned on. 2009 2010* ``high``: This SYN filter has higher priority than other filters. 2011 2012* ``low``: This SYN filter has lower priority than other filters. 2013 2014* ``queue_id``: The receive queue associated with this SYN filter 2015 2016Example:: 2017 2018 testpmd> syn_filter 0 add priority high queue 3 2019 2020flex_filter 2021~~~~~~~~~~~ 2022 2023With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet 2024and routed into one of the receive queues:: 2025 2026 flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \ 2027 mask (mask_value) priority (prio_value) queue (queue_id) 2028 2029The available information parameters are: 2030 2031* ``port_id``: The port which the Flex filter is assigned on. 2032 2033* ``len_value``: Filter length in bytes, no greater than 128. 2034 2035* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match. 2036 2037* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match. 2038 2039* ``prio_value``: The priority of this filter. 2040 2041* ``queue_id``: The receive queue associated with this Flex filter. 2042 2043Example:: 2044 2045 testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \ 2046 mask 000C priority 3 queue 3 2047 2048 testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \ 2049 mask 000C priority 3 queue 3 2050 2051 2052.. _testpmd_flow_director: 2053 2054flow_director_filter 2055~~~~~~~~~~~~~~~~~~~~ 2056 2057The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues. 2058 2059Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and 2060Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter: 2061 2062* Perfect match filters. 2063 The hardware checks a match between the masked fields of the received packets and the programmed filters. 2064 The masked fields are for IP flow. 2065 2066* Signature filters. 2067 The hardware checks a match between a hash-based signature of the masked fields of the received packet. 2068 2069* Perfect-mac-vlan match filters. 2070 The hardware checks a match between the masked fields of the received packets and the programmed filters. 2071 The masked fields are for MAC VLAN flow. 2072 2073* Perfect-tunnel match filters. 2074 The hardware checks a match between the masked fields of the received packets and the programmed filters. 2075 The masked fields are for tunnel flow. 2076 2077The Flow Director filters can match the different fields for different type of packet: flow type, specific input set 2078per flow type and the flexible payload. 2079 2080The Flow Director can also mask out parts of all of these fields so that filters 2081are only applied to certain fields or parts of the fields. 2082 2083Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information. 2084 2085# Commands to add flow director filters of different flow types:: 2086 2087 flow_director_filter (port_id) mode IP (add|del|update) \ 2088 flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \ 2089 src (src_ip_address) dst (dst_ip_address) \ 2090 tos (tos_value) proto (proto_value) ttl (ttl_value) \ 2091 vlan (vlan_value) 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 IP (add|del|update) \ 2096 flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \ 2097 src (src_ip_address) (src_port) \ 2098 dst (dst_ip_address) (dst_port) \ 2099 tos (tos_value) ttl (ttl_value) \ 2100 vlan (vlan_value) flexbytes (flexbytes_value) \ 2101 (drop|fwd) queue pf|vf(vf_id) (queue_id) \ 2102 fd_id (fd_id_value) 2103 2104 flow_director_filter (port_id) mode IP (add|del|update) \ 2105 flow (ipv4-sctp|ipv6-sctp) \ 2106 src (src_ip_address) (src_port) \ 2107 dst (dst_ip_address) (dst_port) \ 2108 tos (tos_value) ttl (ttl_value) \ 2109 tag (verification_tag) vlan (vlan_value) \ 2110 flexbytes (flexbytes_value) (drop|fwd) \ 2111 pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value) 2112 2113 flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \ 2114 ether (ethertype) flexbytes (flexbytes_value) \ 2115 (drop|fwd) pf|vf(vf_id) queue (queue_id) 2116 fd_id (fd_id_value) 2117 2118 flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \ 2119 mac (mac_address) vlan (vlan_value) \ 2120 flexbytes (flexbytes_value) (drop|fwd) \ 2121 queue (queue_id) fd_id (fd_id_value) 2122 2123 flow_director_filter (port_id) mode Tunnel (add|del|update) \ 2124 mac (mac_address) vlan (vlan_value) \ 2125 tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \ 2126 flexbytes (flexbytes_value) (drop|fwd) \ 2127 queue (queue_id) fd_id (fd_id_value) 2128 2129For example, to add an ipv4-udp flow type filter:: 2130 2131 testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \ 2132 dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \ 2133 fwd pf queue 1 fd_id 1 2134 2135For example, add an ipv4-other flow type filter:: 2136 2137 testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \ 2138 dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \ 2139 flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1 2140 2141flush_flow_director 2142~~~~~~~~~~~~~~~~~~~ 2143 2144Flush all flow director filters on a device:: 2145 2146 testpmd> flush_flow_director (port_id) 2147 2148Example, to flush all flow director filter on port 0:: 2149 2150 testpmd> flush_flow_director 0 2151 2152flow_director_mask 2153~~~~~~~~~~~~~~~~~~ 2154 2155Set flow director's input masks:: 2156 2157 flow_director_mask (port_id) mode IP vlan (vlan_value) \ 2158 src_mask (ipv4_src) (ipv6_src) (src_port) \ 2159 dst_mask (ipv4_dst) (ipv6_dst) (dst_port) 2160 2161 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value) 2162 2163 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \ 2164 mac (mac_value) tunnel-type (tunnel_type_value) \ 2165 tunnel-id (tunnel_id_value) 2166 2167Example, to set flow director mask on port 0:: 2168 2169 testpmd> flow_director_mask 0 mode IP vlan 0xefff \ 2170 src_mask 255.255.255.255 \ 2171 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \ 2172 dst_mask 255.255.255.255 \ 2173 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF 2174 2175flow_director_flex_mask 2176~~~~~~~~~~~~~~~~~~~~~~~ 2177 2178set masks of flow director's flexible payload based on certain flow type:: 2179 2180 testpmd> flow_director_flex_mask (port_id) \ 2181 flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2182 ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \ 2183 l2_payload|all) (mask) 2184 2185Example, to set flow director's flex mask for all flow type on port 0:: 2186 2187 testpmd> flow_director_flex_mask 0 flow all \ 2188 (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0) 2189 2190 2191flow_director_flex_payload 2192~~~~~~~~~~~~~~~~~~~~~~~~~~ 2193 2194Configure flexible payload selection:: 2195 2196 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config) 2197 2198For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload:: 2199 2200 testpmd> flow_director_flex_payload 0 l4 \ 2201 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19) 2202 2203get_sym_hash_ena_per_port 2204~~~~~~~~~~~~~~~~~~~~~~~~~ 2205 2206Get symmetric hash enable configuration per port:: 2207 2208 get_sym_hash_ena_per_port (port_id) 2209 2210For example, to get symmetric hash enable configuration of port 1:: 2211 2212 testpmd> get_sym_hash_ena_per_port 1 2213 2214set_sym_hash_ena_per_port 2215~~~~~~~~~~~~~~~~~~~~~~~~~ 2216 2217Set symmetric hash enable configuration per port to enable or disable:: 2218 2219 set_sym_hash_ena_per_port (port_id) (enable|disable) 2220 2221For example, to set symmetric hash enable configuration of port 1 to enable:: 2222 2223 testpmd> set_sym_hash_ena_per_port 1 enable 2224 2225get_hash_global_config 2226~~~~~~~~~~~~~~~~~~~~~~ 2227 2228Get the global configurations of hash filters:: 2229 2230 get_hash_global_config (port_id) 2231 2232For example, to get the global configurations of hash filters of port 1:: 2233 2234 testpmd> get_hash_global_config 1 2235 2236set_hash_global_config 2237~~~~~~~~~~~~~~~~~~~~~~ 2238 2239Set the global configurations of hash filters:: 2240 2241 set_hash_global_config (port_id) (toeplitz|simple_xor|default) \ 2242 (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \ 2243 ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \ 2244 (enable|disable) 2245 2246For example, to enable simple_xor for flow type of ipv6 on port 2:: 2247 2248 testpmd> set_hash_global_config 2 simple_xor ipv6 enable 2249 2250set_hash_input_set 2251~~~~~~~~~~~~~~~~~~ 2252 2253Set the input set for hash:: 2254 2255 set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2256 ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \ 2257 l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \ 2258 ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \ 2259 tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \ 2260 udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \ 2261 fld-8th|none) (select|add) 2262 2263For example, to add source IP to hash input set for flow type of ipv4-udp on port 0:: 2264 2265 testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add 2266 2267set_fdir_input_set 2268~~~~~~~~~~~~~~~~~~ 2269 2270The Flow Director filters can match the different fields for different type of packet, i.e. specific input set 2271on per flow type and the flexible payload. This command can be used to change input set for each flow type. 2272 2273Set the input set for flow director:: 2274 2275 set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2276 ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \ 2277 l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \ 2278 ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \ 2279 tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \ 2280 sctp-dst-port|sctp-veri-tag|none) (select|add) 2281 2282For example to add source IP to FD input set for flow type of ipv4-udp on port 0:: 2283 2284 testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add 2285 2286global_config 2287~~~~~~~~~~~~~ 2288 2289Set different GRE key length for input set:: 2290 2291 global_config (port_id) gre-key-len (number in bytes) 2292 2293For example to set GRE key length for input set to 4 bytes on port 0:: 2294 2295 testpmd> global_config 0 gre-key-len 4 2296 2297 2298.. _testpmd_rte_flow: 2299 2300Flow rules management 2301--------------------- 2302 2303Control of the generic flow API (*rte_flow*) is fully exposed through the 2304``flow`` command (validation, creation, destruction, queries and operation 2305modes). 2306 2307Considering *rte_flow* overlaps with all `Filter Functions`_, using both 2308features simultaneously may cause undefined side-effects and is therefore 2309not recommended. 2310 2311``flow`` syntax 2312~~~~~~~~~~~~~~~ 2313 2314Because the ``flow`` command uses dynamic tokens to handle the large number 2315of possible flow rules combinations, its behavior differs slightly from 2316other commands, in particular: 2317 2318- Pressing *?* or the *<tab>* key displays contextual help for the current 2319 token, not that of the entire command. 2320 2321- Optional and repeated parameters are supported (provided they are listed 2322 in the contextual help). 2323 2324The first parameter stands for the operation mode. Possible operations and 2325their general syntax are described below. They are covered in detail in the 2326following sections. 2327 2328- Check whether a flow rule can be created:: 2329 2330 flow validate {port_id} 2331 [group {group_id}] [priority {level}] [ingress] [egress] 2332 pattern {item} [/ {item} [...]] / end 2333 actions {action} [/ {action} [...]] / end 2334 2335- Create a flow rule:: 2336 2337 flow create {port_id} 2338 [group {group_id}] [priority {level}] [ingress] [egress] 2339 pattern {item} [/ {item} [...]] / end 2340 actions {action} [/ {action} [...]] / end 2341 2342- Destroy specific flow rules:: 2343 2344 flow destroy {port_id} rule {rule_id} [...] 2345 2346- Destroy all flow rules:: 2347 2348 flow flush {port_id} 2349 2350- Query an existing flow rule:: 2351 2352 flow query {port_id} {rule_id} {action} 2353 2354- List existing flow rules sorted by priority, filtered by group 2355 identifiers:: 2356 2357 flow list {port_id} [group {group_id}] [...] 2358 2359- Restrict ingress traffic to the defined flow rules:: 2360 2361 flow isolate {port_id} {boolean} 2362 2363Validating flow rules 2364~~~~~~~~~~~~~~~~~~~~~ 2365 2366``flow validate`` reports whether a flow rule would be accepted by the 2367underlying device in its current state but stops short of creating it. It is 2368bound to ``rte_flow_validate()``:: 2369 2370 flow validate {port_id} 2371 [group {group_id}] [priority {level}] [ingress] [egress] 2372 pattern {item} [/ {item} [...]] / end 2373 actions {action} [/ {action} [...]] / end 2374 2375If successful, it will show:: 2376 2377 Flow rule validated 2378 2379Otherwise it will show an error message of the form:: 2380 2381 Caught error type [...] ([...]): [...] 2382 2383This command uses the same parameters as ``flow create``, their format is 2384described in `Creating flow rules`_. 2385 2386Check whether redirecting any Ethernet packet received on port 0 to RX queue 2387index 6 is supported:: 2388 2389 testpmd> flow validate 0 ingress pattern eth / end 2390 actions queue index 6 / end 2391 Flow rule validated 2392 testpmd> 2393 2394Port 0 does not support TCPv6 rules:: 2395 2396 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end 2397 actions drop / end 2398 Caught error type 9 (specific pattern item): Invalid argument 2399 testpmd> 2400 2401Creating flow rules 2402~~~~~~~~~~~~~~~~~~~ 2403 2404``flow create`` validates and creates the specified flow rule. It is bound 2405to ``rte_flow_create()``:: 2406 2407 flow create {port_id} 2408 [group {group_id}] [priority {level}] [ingress] [egress] 2409 pattern {item} [/ {item} [...]] / end 2410 actions {action} [/ {action} [...]] / end 2411 2412If successful, it will return a flow rule ID usable with other commands:: 2413 2414 Flow rule #[...] created 2415 2416Otherwise it will show an error message of the form:: 2417 2418 Caught error type [...] ([...]): [...] 2419 2420Parameters describe in the following order: 2421 2422- Attributes (*group*, *priority*, *ingress*, *egress* tokens). 2423- A matching pattern, starting with the *pattern* token and terminated by an 2424 *end* pattern item. 2425- Actions, starting with the *actions* token and terminated by an *end* 2426 action. 2427 2428These translate directly to *rte_flow* objects provided as-is to the 2429underlying functions. 2430 2431The shortest valid definition only comprises mandatory tokens:: 2432 2433 testpmd> flow create 0 pattern end actions end 2434 2435Note that PMDs may refuse rules that essentially do nothing such as this 2436one. 2437 2438**All unspecified object values are automatically initialized to 0.** 2439 2440Attributes 2441^^^^^^^^^^ 2442 2443These tokens affect flow rule attributes (``struct rte_flow_attr``) and are 2444specified before the ``pattern`` token. 2445 2446- ``group {group id}``: priority group. 2447- ``priority {level}``: priority level within group. 2448- ``ingress``: rule applies to ingress traffic. 2449- ``egress``: rule applies to egress traffic. 2450 2451Each instance of an attribute specified several times overrides the previous 2452value as shown below (group 4 is used):: 2453 2454 testpmd> flow create 0 group 42 group 24 group 4 [...] 2455 2456Note that once enabled, ``ingress`` and ``egress`` cannot be disabled. 2457 2458While not specifying a direction is an error, some rules may allow both 2459simultaneously. 2460 2461Most rules affect RX therefore contain the ``ingress`` token:: 2462 2463 testpmd> flow create 0 ingress pattern [...] 2464 2465Matching pattern 2466^^^^^^^^^^^^^^^^ 2467 2468A matching pattern starts after the ``pattern`` token. It is made of pattern 2469items and is terminated by a mandatory ``end`` item. 2470 2471Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum 2472rte_flow_item_type``). 2473 2474The ``/`` token is used as a separator between pattern items as shown 2475below:: 2476 2477 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...] 2478 2479Note that protocol items like these must be stacked from lowest to highest 2480layer to make sense. For instance, the following rule is either invalid or 2481unlikely to match any packet:: 2482 2483 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...] 2484 2485More information on these restrictions can be found in the *rte_flow* 2486documentation. 2487 2488Several items support additional specification structures, for example 2489``ipv4`` allows specifying source and destination addresses as follows:: 2490 2491 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1 2492 dst is 10.2.0.0 / end [...] 2493 2494This rule matches all IPv4 traffic with the specified properties. 2495 2496In this example, ``src`` and ``dst`` are field names of the underlying 2497``struct rte_flow_item_ipv4`` object. All item properties can be specified 2498in a similar fashion. 2499 2500The ``is`` token means that the subsequent value must be matched exactly, 2501and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item`` 2502accordingly. Possible assignment tokens are: 2503 2504- ``is``: match value perfectly (with full bit-mask). 2505- ``spec``: match value according to configured bit-mask. 2506- ``last``: specify upper bound to establish a range. 2507- ``mask``: specify bit-mask with relevant bits set to one. 2508- ``prefix``: generate bit-mask from a prefix length. 2509 2510These yield identical results:: 2511 2512 ipv4 src is 10.1.1.1 2513 2514:: 2515 2516 ipv4 src spec 10.1.1.1 src mask 255.255.255.255 2517 2518:: 2519 2520 ipv4 src spec 10.1.1.1 src prefix 32 2521 2522:: 2523 2524 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value 2525 2526:: 2527 2528 ipv4 src is 10.1.1.1 src last 0 # 0 disables range 2529 2530Inclusive ranges can be defined with ``last``:: 2531 2532 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4 2533 2534Note that ``mask`` affects both ``spec`` and ``last``:: 2535 2536 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0 2537 # matches 10.1.0.0 to 10.2.255.255 2538 2539Properties can be modified multiple times:: 2540 2541 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4 2542 2543:: 2544 2545 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16 2546 2547Pattern items 2548^^^^^^^^^^^^^ 2549 2550This section lists supported pattern items and their attributes, if any. 2551 2552- ``end``: end list of pattern items. 2553 2554- ``void``: no-op pattern item. 2555 2556- ``invert``: perform actions when pattern does not match. 2557 2558- ``any``: match any protocol for the current layer. 2559 2560 - ``num {unsigned}``: number of layers covered. 2561 2562- ``pf``: match packets addressed to the physical function. 2563 2564- ``vf``: match packets addressed to a virtual function ID. 2565 2566 - ``id {unsigned}``: destination VF ID. 2567 2568- ``port``: device-specific physical port index to use. 2569 2570 - ``index {unsigned}``: physical port index. 2571 2572- ``raw``: match an arbitrary byte string. 2573 2574 - ``relative {boolean}``: look for pattern after the previous item. 2575 - ``search {boolean}``: search pattern from offset (see also limit). 2576 - ``offset {integer}``: absolute or relative offset for pattern. 2577 - ``limit {unsigned}``: search area limit for start of pattern. 2578 - ``pattern {string}``: byte string to look for. 2579 2580- ``eth``: match Ethernet header. 2581 2582 - ``dst {MAC-48}``: destination MAC. 2583 - ``src {MAC-48}``: source MAC. 2584 - ``type {unsigned}``: EtherType. 2585 2586- ``vlan``: match 802.1Q/ad VLAN tag. 2587 2588 - ``tpid {unsigned}``: tag protocol identifier. 2589 - ``tci {unsigned}``: tag control information. 2590 - ``pcp {unsigned}``: priority code point. 2591 - ``dei {unsigned}``: drop eligible indicator. 2592 - ``vid {unsigned}``: VLAN identifier. 2593 2594- ``ipv4``: match IPv4 header. 2595 2596 - ``tos {unsigned}``: type of service. 2597 - ``ttl {unsigned}``: time to live. 2598 - ``proto {unsigned}``: next protocol ID. 2599 - ``src {ipv4 address}``: source address. 2600 - ``dst {ipv4 address}``: destination address. 2601 2602- ``ipv6``: match IPv6 header. 2603 2604 - ``tc {unsigned}``: traffic class. 2605 - ``flow {unsigned}``: flow label. 2606 - ``proto {unsigned}``: protocol (next header). 2607 - ``hop {unsigned}``: hop limit. 2608 - ``src {ipv6 address}``: source address. 2609 - ``dst {ipv6 address}``: destination address. 2610 2611- ``icmp``: match ICMP header. 2612 2613 - ``type {unsigned}``: ICMP packet type. 2614 - ``code {unsigned}``: ICMP packet code. 2615 2616- ``udp``: match UDP header. 2617 2618 - ``src {unsigned}``: UDP source port. 2619 - ``dst {unsigned}``: UDP destination port. 2620 2621- ``tcp``: match TCP header. 2622 2623 - ``src {unsigned}``: TCP source port. 2624 - ``dst {unsigned}``: TCP destination port. 2625 2626- ``sctp``: match SCTP header. 2627 2628 - ``src {unsigned}``: SCTP source port. 2629 - ``dst {unsigned}``: SCTP destination port. 2630 - ``tag {unsigned}``: validation tag. 2631 - ``cksum {unsigned}``: checksum. 2632 2633- ``vxlan``: match VXLAN header. 2634 2635 - ``vni {unsigned}``: VXLAN identifier. 2636 2637- ``e_tag``: match IEEE 802.1BR E-Tag header. 2638 2639 - ``grp_ecid_b {unsigned}``: GRP and E-CID base. 2640 2641- ``nvgre``: match NVGRE header. 2642 2643 - ``tni {unsigned}``: virtual subnet ID. 2644 2645- ``mpls``: match MPLS header. 2646 2647 - ``label {unsigned}``: MPLS label. 2648 2649- ``gre``: match GRE header. 2650 2651 - ``protocol {unsigned}``: protocol type. 2652 2653- ``fuzzy``: fuzzy pattern match, expect faster than default. 2654 2655 - ``thresh {unsigned}``: accuracy threshold. 2656 2657Actions list 2658^^^^^^^^^^^^ 2659 2660A list of actions starts after the ``actions`` token in the same fashion as 2661`Matching pattern`_; actions are separated by ``/`` tokens and the list is 2662terminated by a mandatory ``end`` action. 2663 2664Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum 2665rte_flow_action_type``). 2666 2667Dropping all incoming UDPv4 packets can be expressed as follows:: 2668 2669 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2670 actions drop / end 2671 2672Several actions have configurable properties which must be specified when 2673there is no valid default value. For example, ``queue`` requires a target 2674queue index. 2675 2676This rule redirects incoming UDPv4 traffic to queue index 6:: 2677 2678 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2679 actions queue index 6 / end 2680 2681While this one could be rejected by PMDs (unspecified queue index):: 2682 2683 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2684 actions queue / end 2685 2686As defined by *rte_flow*, the list is not ordered, all actions of a given 2687rule are performed simultaneously. These are equivalent:: 2688 2689 queue index 6 / void / mark id 42 / end 2690 2691:: 2692 2693 void / mark id 42 / queue index 6 / end 2694 2695All actions in a list should have different types, otherwise only the last 2696action of a given type is taken into account:: 2697 2698 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6 2699 2700:: 2701 2702 drop / drop / drop / end # drop is performed only once 2703 2704:: 2705 2706 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24 2707 2708Considering they are performed simultaneously, opposite and overlapping 2709actions can sometimes be combined when the end result is unambiguous:: 2710 2711 drop / queue index 6 / end # drop has no effect 2712 2713:: 2714 2715 drop / dup index 6 / end # same as above 2716 2717:: 2718 2719 queue index 6 / rss queues 6 7 8 / end # queue has no effect 2720 2721:: 2722 2723 drop / passthru / end # drop has no effect 2724 2725Note that PMDs may still refuse such combinations. 2726 2727Actions 2728^^^^^^^ 2729 2730This section lists supported actions and their attributes, if any. 2731 2732- ``end``: end list of actions. 2733 2734- ``void``: no-op action. 2735 2736- ``passthru``: let subsequent rule process matched packets. 2737 2738- ``mark``: attach 32 bit value to packets. 2739 2740 - ``id {unsigned}``: 32 bit value to return with packets. 2741 2742- ``flag``: flag packets. 2743 2744- ``queue``: assign packets to a given queue index. 2745 2746 - ``index {unsigned}``: queue index to use. 2747 2748- ``drop``: drop packets (note: passthru has priority). 2749 2750- ``count``: enable counters for this rule. 2751 2752- ``dup``: duplicate packets to a given queue index. 2753 2754 - ``index {unsigned}``: queue index to duplicate packets to. 2755 2756- ``rss``: spread packets among several queues. 2757 2758 - ``queues [{unsigned} [...]] end``: queue indices to use. 2759 2760- ``pf``: redirect packets to physical device function. 2761 2762- ``vf``: redirect packets to virtual device function. 2763 2764 - ``original {boolean}``: use original VF ID if possible. 2765 - ``id {unsigned}``: VF ID to redirect packets to. 2766 2767Destroying flow rules 2768~~~~~~~~~~~~~~~~~~~~~ 2769 2770``flow destroy`` destroys one or more rules from their rule ID (as returned 2771by ``flow create``), this command calls ``rte_flow_destroy()`` as many 2772times as necessary:: 2773 2774 flow destroy {port_id} rule {rule_id} [...] 2775 2776If successful, it will show:: 2777 2778 Flow rule #[...] destroyed 2779 2780It does not report anything for rule IDs that do not exist. The usual error 2781message is shown when a rule cannot be destroyed:: 2782 2783 Caught error type [...] ([...]): [...] 2784 2785``flow flush`` destroys all rules on a device and does not take extra 2786arguments. It is bound to ``rte_flow_flush()``:: 2787 2788 flow flush {port_id} 2789 2790Any errors are reported as above. 2791 2792Creating several rules and destroying them:: 2793 2794 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2795 actions queue index 2 / end 2796 Flow rule #0 created 2797 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2798 actions queue index 3 / end 2799 Flow rule #1 created 2800 testpmd> flow destroy 0 rule 0 rule 1 2801 Flow rule #1 destroyed 2802 Flow rule #0 destroyed 2803 testpmd> 2804 2805The same result can be achieved using ``flow flush``:: 2806 2807 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2808 actions queue index 2 / end 2809 Flow rule #0 created 2810 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2811 actions queue index 3 / end 2812 Flow rule #1 created 2813 testpmd> flow flush 0 2814 testpmd> 2815 2816Non-existent rule IDs are ignored:: 2817 2818 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2819 actions queue index 2 / end 2820 Flow rule #0 created 2821 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2822 actions queue index 3 / end 2823 Flow rule #1 created 2824 testpmd> flow destroy 0 rule 42 rule 10 rule 2 2825 testpmd> 2826 testpmd> flow destroy 0 rule 0 2827 Flow rule #0 destroyed 2828 testpmd> 2829 2830Querying flow rules 2831~~~~~~~~~~~~~~~~~~~ 2832 2833``flow query`` queries a specific action of a flow rule having that 2834ability. Such actions collect information that can be reported using this 2835command. It is bound to ``rte_flow_query()``:: 2836 2837 flow query {port_id} {rule_id} {action} 2838 2839If successful, it will display either the retrieved data for known actions 2840or the following message:: 2841 2842 Cannot display result for action type [...] ([...]) 2843 2844Otherwise, it will complain either that the rule does not exist or that some 2845error occurred:: 2846 2847 Flow rule #[...] not found 2848 2849:: 2850 2851 Caught error type [...] ([...]): [...] 2852 2853Currently only the ``count`` action is supported. This action reports the 2854number of packets that hit the flow rule and the total number of bytes. Its 2855output has the following format:: 2856 2857 count: 2858 hits_set: [...] # whether "hits" contains a valid value 2859 bytes_set: [...] # whether "bytes" contains a valid value 2860 hits: [...] # number of packets 2861 bytes: [...] # number of bytes 2862 2863Querying counters for TCPv6 packets redirected to queue 6:: 2864 2865 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end 2866 actions queue index 6 / count / end 2867 Flow rule #4 created 2868 testpmd> flow query 0 4 count 2869 count: 2870 hits_set: 1 2871 bytes_set: 0 2872 hits: 386446 2873 bytes: 0 2874 testpmd> 2875 2876Listing flow rules 2877~~~~~~~~~~~~~~~~~~ 2878 2879``flow list`` lists existing flow rules sorted by priority and optionally 2880filtered by group identifiers:: 2881 2882 flow list {port_id} [group {group_id}] [...] 2883 2884This command only fails with the following message if the device does not 2885exist:: 2886 2887 Invalid port [...] 2888 2889Output consists of a header line followed by a short description of each 2890flow rule, one per line. There is no output at all when no flow rules are 2891configured on the device:: 2892 2893 ID Group Prio Attr Rule 2894 [...] [...] [...] [...] [...] 2895 2896``Attr`` column flags: 2897 2898- ``i`` for ``ingress``. 2899- ``e`` for ``egress``. 2900 2901Creating several flow rules and listing them:: 2902 2903 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2904 actions queue index 6 / end 2905 Flow rule #0 created 2906 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2907 actions queue index 2 / end 2908 Flow rule #1 created 2909 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end 2910 actions rss queues 6 7 8 end / end 2911 Flow rule #2 created 2912 testpmd> flow list 0 2913 ID Group Prio Attr Rule 2914 0 0 0 i- ETH IPV4 => QUEUE 2915 1 0 0 i- ETH IPV6 => QUEUE 2916 2 0 5 i- ETH IPV4 UDP => RSS 2917 testpmd> 2918 2919Rules are sorted by priority (i.e. group ID first, then priority level):: 2920 2921 testpmd> flow list 1 2922 ID Group Prio Attr Rule 2923 0 0 0 i- ETH => COUNT 2924 6 0 500 i- ETH IPV6 TCP => DROP COUNT 2925 5 0 1000 i- ETH IPV6 ICMP => QUEUE 2926 1 24 0 i- ETH IPV4 UDP => QUEUE 2927 4 24 10 i- ETH IPV4 TCP => DROP 2928 3 24 20 i- ETH IPV4 => DROP 2929 2 24 42 i- ETH IPV4 UDP => QUEUE 2930 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE 2931 testpmd> 2932 2933Output can be limited to specific groups:: 2934 2935 testpmd> flow list 1 group 0 group 63 2936 ID Group Prio Attr Rule 2937 0 0 0 i- ETH => COUNT 2938 6 0 500 i- ETH IPV6 TCP => DROP COUNT 2939 5 0 1000 i- ETH IPV6 ICMP => QUEUE 2940 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE 2941 testpmd> 2942 2943Toggling isolated mode 2944~~~~~~~~~~~~~~~~~~~~~~ 2945 2946``flow isolate`` can be used to tell the underlying PMD that ingress traffic 2947must only be injected from the defined flow rules; that no default traffic 2948is expected outside those rules and the driver is free to assign more 2949resources to handle them. It is bound to ``rte_flow_isolate()``:: 2950 2951 flow isolate {port_id} {boolean} 2952 2953If successful, enabling or disabling isolated mode shows either:: 2954 2955 Ingress traffic on port [...] 2956 is now restricted to the defined flow rules 2957 2958Or:: 2959 2960 Ingress traffic on port [...] 2961 is not restricted anymore to the defined flow rules 2962 2963Otherwise, in case of error:: 2964 2965 Caught error type [...] ([...]): [...] 2966 2967Mainly due to its side effects, PMDs supporting this mode may not have the 2968ability to toggle it more than once without reinitializing affected ports 2969first (e.g. by exiting testpmd). 2970 2971Enabling isolated mode:: 2972 2973 testpmd> flow isolate 0 true 2974 Ingress traffic on port 0 is now restricted to the defined flow rules 2975 testpmd> 2976 2977Disabling isolated mode:: 2978 2979 testpmd> flow isolate 0 false 2980 Ingress traffic on port 0 is not restricted anymore to the defined flow rules 2981 testpmd> 2982 2983Sample QinQ flow rules 2984~~~~~~~~~~~~~~~~~~~~~~ 2985 2986Before creating QinQ rule(s) the following commands should be issued to enable QinQ:: 2987 2988 testpmd> port stop 0 2989 testpmd> vlan set qinq on 0 2990 2991The above command sets the inner and outer TPID's to 0x8100. 2992 2993To change the TPID's the following commands should be used:: 2994 2995 testpmd> vlan set outer tpid 0xa100 0 2996 testpmd> vlan set inner tpid 0x9100 0 2997 testpmd> port start 0 2998 2999Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM. 3000 3001:: 3002 3003 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 / 3004 vlan tci is 456 / end actions vf id 1 / queue index 0 / end 3005 Flow rule #0 validated 3006 3007 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 / 3008 vlan tci is 456 / end actions vf id 123 / queue index 0 / end 3009 Flow rule #0 created 3010 3011 testpmd> flow list 0 3012 ID Group Prio Attr Rule 3013 0 0 0 i- ETH VLAN VLAN=>VF QUEUE 3014 3015Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host. 3016 3017:: 3018 3019 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 / 3020 vlan tci is 654 / end actions pf / queue index 0 / end 3021 Flow rule #1 validated 3022 3023 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 / 3024 vlan tci is 654 / end actions pf / queue index 1 / end 3025 Flow rule #1 created 3026 3027 testpmd> flow list 0 3028 ID Group Prio Attr Rule 3029 0 0 0 i- ETH VLAN VLAN=>VF QUEUE 3030 1 0 0 i- ETH VLAN VLAN=>PF QUEUE 3031