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) 1234 1235ptype mapping 1236~~~~~~~~~~~~~ 1237 1238List all items from the ptype mapping table:: 1239 1240 testpmd> ptype mapping get (port_id) (valid_only) 1241 1242Where: 1243 1244* ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0). 1245 1246Replace a specific or a group of software defined ptype with a new one:: 1247 1248 testpmd> ptype mapping replace (port_id) (target) (mask) (pkt_type) 1249 1250where: 1251 1252* ``target``: A specific software ptype or a mask to represent a group of software ptypes. 1253 1254* ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1). 1255 1256* ``pkt_type``: The new software ptype to replace the old ones. 1257 1258Update hardware defined ptype to software defined packet type mapping table:: 1259 1260 testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype) 1261 1262where: 1263 1264* ``hw_ptype``: hardware ptype as the index of the ptype mapping table. 1265 1266* ``sw_ptype``: software ptype as the value of the ptype mapping table. 1267 1268Reset ptype mapping table:: 1269 1270 testpmd> ptype mapping reset (port_id) 1271 1272Port Functions 1273-------------- 1274 1275The following sections show functions for configuring ports. 1276 1277.. note:: 1278 1279 Port configuration changes only become active when forwarding is started/restarted. 1280 1281port attach 1282~~~~~~~~~~~ 1283 1284Attach a port specified by pci address or virtual device args:: 1285 1286 testpmd> port attach (identifier) 1287 1288To attach a new pci device, the device should be recognized by kernel first. 1289Then it should be moved under DPDK management. 1290Finally the port can be attached to testpmd. 1291 1292For example, to move a pci device using ixgbe under DPDK management: 1293 1294.. code-block:: console 1295 1296 # Check the status of the available devices. 1297 ./usertools/dpdk-devbind.py --status 1298 1299 Network devices using DPDK-compatible driver 1300 ============================================ 1301 <none> 1302 1303 Network devices using kernel driver 1304 =================================== 1305 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused= 1306 1307 1308 # Bind the device to igb_uio. 1309 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0 1310 1311 1312 # Recheck the status of the devices. 1313 ./usertools/dpdk-devbind.py --status 1314 Network devices using DPDK-compatible driver 1315 ============================================ 1316 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused= 1317 1318To attach a port created by virtual device, above steps are not needed. 1319 1320For example, to attach a port whose pci address is 0000:0a:00.0. 1321 1322.. code-block:: console 1323 1324 testpmd> port attach 0000:0a:00.0 1325 Attaching a new port... 1326 EAL: PCI device 0000:0a:00.0 on NUMA socket -1 1327 EAL: probe driver: 8086:10fb rte_ixgbe_pmd 1328 EAL: PCI memory mapped at 0x7f83bfa00000 1329 EAL: PCI memory mapped at 0x7f83bfa80000 1330 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5 1331 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb 1332 Port 0 is attached. Now total ports is 1 1333 Done 1334 1335For example, to attach a port created by pcap PMD. 1336 1337.. code-block:: console 1338 1339 testpmd> port attach net_pcap0 1340 Attaching a new port... 1341 PMD: Initializing pmd_pcap for net_pcap0 1342 PMD: Creating pcap-backed ethdev on numa socket 0 1343 Port 0 is attached. Now total ports is 1 1344 Done 1345 1346In this case, identifier is ``net_pcap0``. 1347This identifier format is the same as ``--vdev`` format of DPDK applications. 1348 1349For example, to re-attach a bonded port which has been previously detached, 1350the mode and slave parameters must be given. 1351 1352.. code-block:: console 1353 1354 testpmd> port attach net_bond_0,mode=0,slave=1 1355 Attaching a new port... 1356 EAL: Initializing pmd_bond for net_bond_0 1357 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0. 1358 Port 0 is attached. Now total ports is 1 1359 Done 1360 1361 1362port detach 1363~~~~~~~~~~~ 1364 1365Detach a specific port:: 1366 1367 testpmd> port detach (port_id) 1368 1369Before detaching a port, the port should be stopped and closed. 1370 1371For example, to detach a pci device port 0. 1372 1373.. code-block:: console 1374 1375 testpmd> port stop 0 1376 Stopping ports... 1377 Done 1378 testpmd> port close 0 1379 Closing ports... 1380 Done 1381 1382 testpmd> port detach 0 1383 Detaching a port... 1384 EAL: PCI device 0000:0a:00.0 on NUMA socket -1 1385 EAL: remove driver: 8086:10fb rte_ixgbe_pmd 1386 EAL: PCI memory unmapped at 0x7f83bfa00000 1387 EAL: PCI memory unmapped at 0x7f83bfa80000 1388 Done 1389 1390 1391For example, to detach a virtual device port 0. 1392 1393.. code-block:: console 1394 1395 testpmd> port stop 0 1396 Stopping ports... 1397 Done 1398 testpmd> port close 0 1399 Closing ports... 1400 Done 1401 1402 testpmd> port detach 0 1403 Detaching a port... 1404 PMD: Closing pcap ethdev on numa socket 0 1405 Port 'net_pcap0' is detached. Now total ports is 0 1406 Done 1407 1408To remove a pci device completely from the system, first detach the port from testpmd. 1409Then the device should be moved under kernel management. 1410Finally the device can be removed using kernel pci hotplug functionality. 1411 1412For example, to move a pci device under kernel management: 1413 1414.. code-block:: console 1415 1416 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0 1417 1418 ./usertools/dpdk-devbind.py --status 1419 1420 Network devices using DPDK-compatible driver 1421 ============================================ 1422 <none> 1423 1424 Network devices using kernel driver 1425 =================================== 1426 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio 1427 1428To remove a port created by a virtual device, above steps are not needed. 1429 1430port start 1431~~~~~~~~~~ 1432 1433Start all ports or a specific port:: 1434 1435 testpmd> port start (port_id|all) 1436 1437port stop 1438~~~~~~~~~ 1439 1440Stop all ports or a specific port:: 1441 1442 testpmd> port stop (port_id|all) 1443 1444port close 1445~~~~~~~~~~ 1446 1447Close all ports or a specific port:: 1448 1449 testpmd> port close (port_id|all) 1450 1451port start/stop queue 1452~~~~~~~~~~~~~~~~~~~~~ 1453 1454Start/stop a rx/tx queue on a specific port:: 1455 1456 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop) 1457 1458Only take effect when port is started. 1459 1460port config - speed 1461~~~~~~~~~~~~~~~~~~~ 1462 1463Set the speed and duplex mode for all ports or a specific port:: 1464 1465 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \ 1466 duplex (half|full|auto) 1467 1468port config - queues/descriptors 1469~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1470 1471Set number of queues/descriptors for rxq, txq, rxd and txd:: 1472 1473 testpmd> port config all (rxq|txq|rxd|txd) (value) 1474 1475This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options. 1476 1477port config - max-pkt-len 1478~~~~~~~~~~~~~~~~~~~~~~~~~ 1479 1480Set the maximum packet length:: 1481 1482 testpmd> port config all max-pkt-len (value) 1483 1484This is equivalent to the ``--max-pkt-len`` command-line option. 1485 1486port config - CRC Strip 1487~~~~~~~~~~~~~~~~~~~~~~~ 1488 1489Set hardware CRC stripping on or off for all ports:: 1490 1491 testpmd> port config all crc-strip (on|off) 1492 1493CRC stripping is on by default. 1494 1495The ``off`` option is equivalent to the ``--disable-crc-strip`` command-line option. 1496 1497port config - scatter 1498~~~~~~~~~~~~~~~~~~~~~~~ 1499 1500Set RX scatter mode on or off for all ports:: 1501 1502 testpmd> port config all scatter (on|off) 1503 1504RX scatter mode is off by default. 1505 1506The ``on`` option is equivalent to the ``--enable-scatter`` command-line option. 1507 1508port config - TX queue flags 1509~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1510 1511Set a hexadecimal bitmap of TX queue flags for all ports:: 1512 1513 testpmd> port config all txqflags value 1514 1515This command is equivalent to the ``--txqflags`` command-line option. 1516 1517port config - RX Checksum 1518~~~~~~~~~~~~~~~~~~~~~~~~~ 1519 1520Set hardware RX checksum offload to on or off for all ports:: 1521 1522 testpmd> port config all rx-cksum (on|off) 1523 1524Checksum offload is off by default. 1525 1526The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option. 1527 1528port config - VLAN 1529~~~~~~~~~~~~~~~~~~ 1530 1531Set hardware VLAN on or off for all ports:: 1532 1533 testpmd> port config all hw-vlan (on|off) 1534 1535Hardware VLAN is on by default. 1536 1537The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option. 1538 1539port config - VLAN filter 1540~~~~~~~~~~~~~~~~~~~~~~~~~ 1541 1542Set hardware VLAN filter on or off for all ports:: 1543 1544 testpmd> port config all hw-vlan-filter (on|off) 1545 1546Hardware VLAN filter is on by default. 1547 1548The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option. 1549 1550port config - VLAN strip 1551~~~~~~~~~~~~~~~~~~~~~~~~ 1552 1553Set hardware VLAN strip on or off for all ports:: 1554 1555 testpmd> port config all hw-vlan-strip (on|off) 1556 1557Hardware VLAN strip is on by default. 1558 1559The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option. 1560 1561port config - VLAN extend 1562~~~~~~~~~~~~~~~~~~~~~~~~~ 1563 1564Set hardware VLAN extend on or off for all ports:: 1565 1566 testpmd> port config all hw-vlan-extend (on|off) 1567 1568Hardware VLAN extend is off by default. 1569 1570The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option. 1571 1572port config - Drop Packets 1573~~~~~~~~~~~~~~~~~~~~~~~~~~ 1574 1575Set packet drop for packets with no descriptors on or off for all ports:: 1576 1577 testpmd> port config all drop-en (on|off) 1578 1579Packet dropping for packets with no descriptors is off by default. 1580 1581The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option. 1582 1583port config - RSS 1584~~~~~~~~~~~~~~~~~ 1585 1586Set the RSS (Receive Side Scaling) mode on or off:: 1587 1588 testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none) 1589 1590RSS is on by default. 1591 1592The ``none`` option is equivalent to the ``--disable-rss`` command-line option. 1593 1594port config - RSS Reta 1595~~~~~~~~~~~~~~~~~~~~~~ 1596 1597Set the RSS (Receive Side Scaling) redirection table:: 1598 1599 testpmd> port config all rss reta (hash,queue)[,(hash,queue)] 1600 1601port config - DCB 1602~~~~~~~~~~~~~~~~~ 1603 1604Set the DCB mode for an individual port:: 1605 1606 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off) 1607 1608The traffic class should be 4 or 8. 1609 1610port config - Burst 1611~~~~~~~~~~~~~~~~~~~ 1612 1613Set the number of packets per burst:: 1614 1615 testpmd> port config all burst (value) 1616 1617This is equivalent to the ``--burst`` command-line option. 1618 1619port config - Threshold 1620~~~~~~~~~~~~~~~~~~~~~~~ 1621 1622Set thresholds for TX/RX queues:: 1623 1624 testpmd> port config all (threshold) (value) 1625 1626Where the threshold type can be: 1627 1628* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255. 1629 1630* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255. 1631 1632* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255. 1633 1634* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255. 1635 1636* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255. 1637 1638* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255. 1639 1640* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd. 1641 1642* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd. 1643 1644* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd. 1645 1646These threshold options are also available from the command-line. 1647 1648port config - E-tag 1649~~~~~~~~~~~~~~~~~~~ 1650 1651Set the value of ether-type for E-tag:: 1652 1653 testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value) 1654 1655Enable/disable the E-tag support:: 1656 1657 testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable) 1658 1659 1660Link Bonding Functions 1661---------------------- 1662 1663The Link Bonding functions make it possible to dynamically create and 1664manage link bonding devices from within testpmd interactive prompt. 1665 1666create bonded device 1667~~~~~~~~~~~~~~~~~~~~ 1668 1669Create a new bonding device:: 1670 1671 testpmd> create bonded device (mode) (socket) 1672 1673For example, to create a bonded device in mode 1 on socket 0:: 1674 1675 testpmd> create bonded 1 0 1676 created new bonded device (port X) 1677 1678add bonding slave 1679~~~~~~~~~~~~~~~~~ 1680 1681Adds Ethernet device to a Link Bonding device:: 1682 1683 testpmd> add bonding slave (slave id) (port id) 1684 1685For example, to add Ethernet device (port 6) to a Link Bonding device (port 10):: 1686 1687 testpmd> add bonding slave 6 10 1688 1689 1690remove bonding slave 1691~~~~~~~~~~~~~~~~~~~~ 1692 1693Removes an Ethernet slave device from a Link Bonding device:: 1694 1695 testpmd> remove bonding slave (slave id) (port id) 1696 1697For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10):: 1698 1699 testpmd> remove bonding slave 6 10 1700 1701set bonding mode 1702~~~~~~~~~~~~~~~~ 1703 1704Set the Link Bonding mode of a Link Bonding device:: 1705 1706 testpmd> set bonding mode (value) (port id) 1707 1708For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3):: 1709 1710 testpmd> set bonding mode 3 10 1711 1712set bonding primary 1713~~~~~~~~~~~~~~~~~~~ 1714 1715Set an Ethernet slave device as the primary device on a Link Bonding device:: 1716 1717 testpmd> set bonding primary (slave id) (port id) 1718 1719For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10):: 1720 1721 testpmd> set bonding primary 6 10 1722 1723set bonding mac 1724~~~~~~~~~~~~~~~ 1725 1726Set the MAC address of a Link Bonding device:: 1727 1728 testpmd> set bonding mac (port id) (mac) 1729 1730For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01:: 1731 1732 testpmd> set bonding mac 10 00:00:00:00:00:01 1733 1734set bonding xmit_balance_policy 1735~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1736 1737Set the transmission policy for a Link Bonding device when it is in Balance XOR mode:: 1738 1739 testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34) 1740 1741For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports):: 1742 1743 testpmd> set bonding xmit_balance_policy 10 l34 1744 1745 1746set bonding mon_period 1747~~~~~~~~~~~~~~~~~~~~~~ 1748 1749Set the link status monitoring polling period in milliseconds for a bonding device. 1750 1751This adds support for PMD slave devices which do not support link status interrupts. 1752When the mon_period is set to a value greater than 0 then all PMD's which do not support 1753link status ISR will be queried every polling interval to check if their link status has changed:: 1754 1755 testpmd> set bonding mon_period (port_id) (value) 1756 1757For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms:: 1758 1759 testpmd> set bonding mon_period 5 150 1760 1761 1762show bonding config 1763~~~~~~~~~~~~~~~~~~~ 1764 1765Show the current configuration of a Link Bonding device:: 1766 1767 testpmd> show bonding config (port id) 1768 1769For example, 1770to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4) 1771in balance mode with a transmission policy of layer 2+3:: 1772 1773 testpmd> show bonding config 9 1774 Bonding mode: 2 1775 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23 1776 Slaves (3): [1 3 4] 1777 Active Slaves (3): [1 3 4] 1778 Primary: [3] 1779 1780 1781Register Functions 1782------------------ 1783 1784The Register Functions can be used to read from and write to registers on the network card referenced by a port number. 1785This is mainly useful for debugging purposes. 1786Reference should be made to the appropriate datasheet for the network card for details on the register addresses 1787and fields that can be accessed. 1788 1789read reg 1790~~~~~~~~ 1791 1792Display the value of a port register:: 1793 1794 testpmd> read reg (port_id) (address) 1795 1796For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller:: 1797 1798 testpmd> read reg 0 0xEE00 1799 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241) 1800 1801read regfield 1802~~~~~~~~~~~~~ 1803 1804Display a port register bit field:: 1805 1806 testpmd> read regfield (port_id) (address) (bit_x) (bit_y) 1807 1808For example, reading the lowest two bits from the register in the example above:: 1809 1810 testpmd> read regfield 0 0xEE00 0 1 1811 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1) 1812 1813read regbit 1814~~~~~~~~~~~ 1815 1816Display a single port register bit:: 1817 1818 testpmd> read regbit (port_id) (address) (bit_x) 1819 1820For example, reading the lowest bit from the register in the example above:: 1821 1822 testpmd> read regbit 0 0xEE00 0 1823 port 0 PCI register at offset 0xEE00: bit 0=1 1824 1825write reg 1826~~~~~~~~~ 1827 1828Set the value of a port register:: 1829 1830 testpmd> write reg (port_id) (address) (value) 1831 1832For example, to clear a register:: 1833 1834 testpmd> write reg 0 0xEE00 0x0 1835 port 0 PCI register at offset 0xEE00: 0x00000000 (0) 1836 1837write regfield 1838~~~~~~~~~~~~~~ 1839 1840Set bit field of a port register:: 1841 1842 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value) 1843 1844For example, writing to the register cleared in the example above:: 1845 1846 testpmd> write regfield 0 0xEE00 0 1 2 1847 port 0 PCI register at offset 0xEE00: 0x00000002 (2) 1848 1849write regbit 1850~~~~~~~~~~~~ 1851 1852Set single bit value of a port register:: 1853 1854 testpmd> write regbit (port_id) (address) (bit_x) (value) 1855 1856For example, to set the high bit in the register from the example above:: 1857 1858 testpmd> write regbit 0 0xEE00 31 1 1859 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658) 1860 1861 1862Filter Functions 1863---------------- 1864 1865This section details the available filter functions that are available. 1866 1867Note these functions interface the deprecated legacy filtering framework, 1868superseded by *rte_flow*. See `Flow rules management`_. 1869 1870ethertype_filter 1871~~~~~~~~~~~~~~~~~~~~ 1872 1873Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue:: 1874 1875 ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \ 1876 ethertype (ether_type) (drop|fwd) queue (queue_id) 1877 1878The available information parameters are: 1879 1880* ``port_id``: The port which the Ethertype filter assigned on. 1881 1882* ``mac_addr``: Compare destination mac address. 1883 1884* ``mac_ignr``: Ignore destination mac address match. 1885 1886* ``mac_address``: Destination mac address to match. 1887 1888* ``ether_type``: The EtherType value want to match, 1889 for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid. 1890 1891* ``queue_id``: The receive queue associated with this EtherType filter. 1892 It is meaningless when deleting or dropping. 1893 1894Example, to add/remove an ethertype filter rule:: 1895 1896 testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \ 1897 ethertype 0x0806 fwd queue 3 1898 1899 testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \ 1900 ethertype 0x0806 fwd queue 3 1901 19022tuple_filter 1903~~~~~~~~~~~~~~~~~ 1904 1905Add or delete a 2-tuple filter, 1906which identifies packets by specific protocol and destination TCP/UDP port 1907and forwards packets into one of the receive queues:: 1908 1909 2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \ 1910 protocol (protocol_value) mask (mask_value) \ 1911 tcp_flags (tcp_flags_value) priority (prio_value) \ 1912 queue (queue_id) 1913 1914The available information parameters are: 1915 1916* ``port_id``: The port which the 2-tuple filter assigned on. 1917 1918* ``dst_port_value``: Destination port in L4. 1919 1920* ``protocol_value``: IP L4 protocol. 1921 1922* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate. 1923 1924* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP). 1925 1926* ``prio_value``: Priority of this filter. 1927 1928* ``queue_id``: The receive queue associated with this 2-tuple filter. 1929 1930Example, to add/remove an 2tuple filter rule:: 1931 1932 testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \ 1933 tcp_flags 0x02 priority 3 queue 3 1934 1935 testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \ 1936 tcp_flags 0x02 priority 3 queue 3 1937 19385tuple_filter 1939~~~~~~~~~~~~~~~~~ 1940 1941Add or delete a 5-tuple filter, 1942which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port) 1943and routes packets into one of the receive queues:: 1944 1945 5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \ 1946 (src_address) dst_port (dst_port_value) \ 1947 src_port (src_port_value) protocol (protocol_value) \ 1948 mask (mask_value) tcp_flags (tcp_flags_value) \ 1949 priority (prio_value) queue (queue_id) 1950 1951The available information parameters are: 1952 1953* ``port_id``: The port which the 5-tuple filter assigned on. 1954 1955* ``dst_address``: Destination IP address. 1956 1957* ``src_address``: Source IP address. 1958 1959* ``dst_port_value``: TCP/UDP destination port. 1960 1961* ``src_port_value``: TCP/UDP source port. 1962 1963* ``protocol_value``: L4 protocol. 1964 1965* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate 1966 1967* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP). 1968 1969* ``prio_value``: The priority of this filter. 1970 1971* ``queue_id``: The receive queue associated with this 5-tuple filter. 1972 1973Example, to add/remove an 5tuple filter rule:: 1974 1975 testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \ 1976 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \ 1977 flags 0x0 priority 3 queue 3 1978 1979 testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \ 1980 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \ 1981 flags 0x0 priority 3 queue 3 1982 1983syn_filter 1984~~~~~~~~~~ 1985 1986Using the SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue:: 1987 1988 syn_filter (port_id) (add|del) priority (high|low) queue (queue_id) 1989 1990The available information parameters are: 1991 1992* ``port_id``: The port which the SYN filter assigned on. 1993 1994* ``high``: This SYN filter has higher priority than other filters. 1995 1996* ``low``: This SYN filter has lower priority than other filters. 1997 1998* ``queue_id``: The receive queue associated with this SYN filter 1999 2000Example:: 2001 2002 testpmd> syn_filter 0 add priority high queue 3 2003 2004flex_filter 2005~~~~~~~~~~~ 2006 2007With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet 2008and routed into one of the receive queues:: 2009 2010 flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \ 2011 mask (mask_value) priority (prio_value) queue (queue_id) 2012 2013The available information parameters are: 2014 2015* ``port_id``: The port which the Flex filter is assigned on. 2016 2017* ``len_value``: Filter length in bytes, no greater than 128. 2018 2019* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match. 2020 2021* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match. 2022 2023* ``prio_value``: The priority of this filter. 2024 2025* ``queue_id``: The receive queue associated with this Flex filter. 2026 2027Example:: 2028 2029 testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \ 2030 mask 000C priority 3 queue 3 2031 2032 testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \ 2033 mask 000C priority 3 queue 3 2034 2035 2036.. _testpmd_flow_director: 2037 2038flow_director_filter 2039~~~~~~~~~~~~~~~~~~~~ 2040 2041The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues. 2042 2043Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and 2044Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter: 2045 2046* Perfect match filters. 2047 The hardware checks a match between the masked fields of the received packets and the programmed filters. 2048 The masked fields are for IP flow. 2049 2050* Signature filters. 2051 The hardware checks a match between a hash-based signature of the masked fields of the received packet. 2052 2053* Perfect-mac-vlan match filters. 2054 The hardware checks a match between the masked fields of the received packets and the programmed filters. 2055 The masked fields are for MAC VLAN flow. 2056 2057* Perfect-tunnel match filters. 2058 The hardware checks a match between the masked fields of the received packets and the programmed filters. 2059 The masked fields are for tunnel flow. 2060 2061The Flow Director filters can match the different fields for different type of packet: flow type, specific input set 2062per flow type and the flexible payload. 2063 2064The Flow Director can also mask out parts of all of these fields so that filters 2065are only applied to certain fields or parts of the fields. 2066 2067Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information. 2068 2069# Commands to add flow director filters of different flow types:: 2070 2071 flow_director_filter (port_id) mode IP (add|del|update) \ 2072 flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \ 2073 src (src_ip_address) dst (dst_ip_address) \ 2074 tos (tos_value) proto (proto_value) ttl (ttl_value) \ 2075 vlan (vlan_value) flexbytes (flexbytes_value) \ 2076 (drop|fwd) pf|vf(vf_id) queue (queue_id) \ 2077 fd_id (fd_id_value) 2078 2079 flow_director_filter (port_id) mode IP (add|del|update) \ 2080 flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \ 2081 src (src_ip_address) (src_port) \ 2082 dst (dst_ip_address) (dst_port) \ 2083 tos (tos_value) ttl (ttl_value) \ 2084 vlan (vlan_value) flexbytes (flexbytes_value) \ 2085 (drop|fwd) queue pf|vf(vf_id) (queue_id) \ 2086 fd_id (fd_id_value) 2087 2088 flow_director_filter (port_id) mode IP (add|del|update) \ 2089 flow (ipv4-sctp|ipv6-sctp) \ 2090 src (src_ip_address) (src_port) \ 2091 dst (dst_ip_address) (dst_port) \ 2092 tos (tos_value) ttl (ttl_value) \ 2093 tag (verification_tag) vlan (vlan_value) \ 2094 flexbytes (flexbytes_value) (drop|fwd) \ 2095 pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value) 2096 2097 flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \ 2098 ether (ethertype) flexbytes (flexbytes_value) \ 2099 (drop|fwd) pf|vf(vf_id) queue (queue_id) 2100 fd_id (fd_id_value) 2101 2102 flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \ 2103 mac (mac_address) vlan (vlan_value) \ 2104 flexbytes (flexbytes_value) (drop|fwd) \ 2105 queue (queue_id) fd_id (fd_id_value) 2106 2107 flow_director_filter (port_id) mode Tunnel (add|del|update) \ 2108 mac (mac_address) vlan (vlan_value) \ 2109 tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \ 2110 flexbytes (flexbytes_value) (drop|fwd) \ 2111 queue (queue_id) fd_id (fd_id_value) 2112 2113For example, to add an ipv4-udp flow type filter:: 2114 2115 testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \ 2116 dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \ 2117 fwd pf queue 1 fd_id 1 2118 2119For example, add an ipv4-other flow type filter:: 2120 2121 testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \ 2122 dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \ 2123 flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1 2124 2125flush_flow_director 2126~~~~~~~~~~~~~~~~~~~ 2127 2128Flush all flow director filters on a device:: 2129 2130 testpmd> flush_flow_director (port_id) 2131 2132Example, to flush all flow director filter on port 0:: 2133 2134 testpmd> flush_flow_director 0 2135 2136flow_director_mask 2137~~~~~~~~~~~~~~~~~~ 2138 2139Set flow director's input masks:: 2140 2141 flow_director_mask (port_id) mode IP vlan (vlan_value) \ 2142 src_mask (ipv4_src) (ipv6_src) (src_port) \ 2143 dst_mask (ipv4_dst) (ipv6_dst) (dst_port) 2144 2145 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value) 2146 2147 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \ 2148 mac (mac_value) tunnel-type (tunnel_type_value) \ 2149 tunnel-id (tunnel_id_value) 2150 2151Example, to set flow director mask on port 0:: 2152 2153 testpmd> flow_director_mask 0 mode IP vlan 0xefff \ 2154 src_mask 255.255.255.255 \ 2155 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \ 2156 dst_mask 255.255.255.255 \ 2157 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF 2158 2159flow_director_flex_mask 2160~~~~~~~~~~~~~~~~~~~~~~~ 2161 2162set masks of flow director's flexible payload based on certain flow type:: 2163 2164 testpmd> flow_director_flex_mask (port_id) \ 2165 flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2166 ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \ 2167 l2_payload|all) (mask) 2168 2169Example, to set flow director's flex mask for all flow type on port 0:: 2170 2171 testpmd> flow_director_flex_mask 0 flow all \ 2172 (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0) 2173 2174 2175flow_director_flex_payload 2176~~~~~~~~~~~~~~~~~~~~~~~~~~ 2177 2178Configure flexible payload selection:: 2179 2180 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config) 2181 2182For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload:: 2183 2184 testpmd> flow_director_flex_payload 0 l4 \ 2185 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19) 2186 2187get_sym_hash_ena_per_port 2188~~~~~~~~~~~~~~~~~~~~~~~~~ 2189 2190Get symmetric hash enable configuration per port:: 2191 2192 get_sym_hash_ena_per_port (port_id) 2193 2194For example, to get symmetric hash enable configuration of port 1:: 2195 2196 testpmd> get_sym_hash_ena_per_port 1 2197 2198set_sym_hash_ena_per_port 2199~~~~~~~~~~~~~~~~~~~~~~~~~ 2200 2201Set symmetric hash enable configuration per port to enable or disable:: 2202 2203 set_sym_hash_ena_per_port (port_id) (enable|disable) 2204 2205For example, to set symmetric hash enable configuration of port 1 to enable:: 2206 2207 testpmd> set_sym_hash_ena_per_port 1 enable 2208 2209get_hash_global_config 2210~~~~~~~~~~~~~~~~~~~~~~ 2211 2212Get the global configurations of hash filters:: 2213 2214 get_hash_global_config (port_id) 2215 2216For example, to get the global configurations of hash filters of port 1:: 2217 2218 testpmd> get_hash_global_config 1 2219 2220set_hash_global_config 2221~~~~~~~~~~~~~~~~~~~~~~ 2222 2223Set the global configurations of hash filters:: 2224 2225 set_hash_global_config (port_id) (toeplitz|simple_xor|default) \ 2226 (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \ 2227 ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \ 2228 (enable|disable) 2229 2230For example, to enable simple_xor for flow type of ipv6 on port 2:: 2231 2232 testpmd> set_hash_global_config 2 simple_xor ipv6 enable 2233 2234set_hash_input_set 2235~~~~~~~~~~~~~~~~~~ 2236 2237Set the input set for hash:: 2238 2239 set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2240 ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \ 2241 l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \ 2242 ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \ 2243 tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \ 2244 udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \ 2245 fld-8th|none) (select|add) 2246 2247For example, to add source IP to hash input set for flow type of ipv4-udp on port 0:: 2248 2249 testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add 2250 2251set_fdir_input_set 2252~~~~~~~~~~~~~~~~~~ 2253 2254The Flow Director filters can match the different fields for different type of packet, i.e. specific input set 2255on per flow type and the flexible payload. This command can be used to change input set for each flow type. 2256 2257Set the input set for flow director:: 2258 2259 set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \ 2260 ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \ 2261 l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \ 2262 ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \ 2263 tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \ 2264 sctp-dst-port|sctp-veri-tag|none) (select|add) 2265 2266For example to add source IP to FD input set for flow type of ipv4-udp on port 0:: 2267 2268 testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add 2269 2270global_config 2271~~~~~~~~~~~~~ 2272 2273Set different GRE key length for input set:: 2274 2275 global_config (port_id) gre-key-len (number in bytes) 2276 2277For example to set GRE key length for input set to 4 bytes on port 0:: 2278 2279 testpmd> global_config 0 gre-key-len 4 2280 2281 2282.. _testpmd_rte_flow: 2283 2284Flow rules management 2285--------------------- 2286 2287Control of the generic flow API (*rte_flow*) is fully exposed through the 2288``flow`` command (validation, creation, destruction, queries and operation 2289modes). 2290 2291Considering *rte_flow* overlaps with all `Filter Functions`_, using both 2292features simultaneously may cause undefined side-effects and is therefore 2293not recommended. 2294 2295``flow`` syntax 2296~~~~~~~~~~~~~~~ 2297 2298Because the ``flow`` command uses dynamic tokens to handle the large number 2299of possible flow rules combinations, its behavior differs slightly from 2300other commands, in particular: 2301 2302- Pressing *?* or the *<tab>* key displays contextual help for the current 2303 token, not that of the entire command. 2304 2305- Optional and repeated parameters are supported (provided they are listed 2306 in the contextual help). 2307 2308The first parameter stands for the operation mode. Possible operations and 2309their general syntax are described below. They are covered in detail in the 2310following sections. 2311 2312- Check whether a flow rule can be created:: 2313 2314 flow validate {port_id} 2315 [group {group_id}] [priority {level}] [ingress] [egress] 2316 pattern {item} [/ {item} [...]] / end 2317 actions {action} [/ {action} [...]] / end 2318 2319- Create a flow rule:: 2320 2321 flow create {port_id} 2322 [group {group_id}] [priority {level}] [ingress] [egress] 2323 pattern {item} [/ {item} [...]] / end 2324 actions {action} [/ {action} [...]] / end 2325 2326- Destroy specific flow rules:: 2327 2328 flow destroy {port_id} rule {rule_id} [...] 2329 2330- Destroy all flow rules:: 2331 2332 flow flush {port_id} 2333 2334- Query an existing flow rule:: 2335 2336 flow query {port_id} {rule_id} {action} 2337 2338- List existing flow rules sorted by priority, filtered by group 2339 identifiers:: 2340 2341 flow list {port_id} [group {group_id}] [...] 2342 2343- Restrict ingress traffic to the defined flow rules:: 2344 2345 flow isolate {port_id} {boolean} 2346 2347Validating flow rules 2348~~~~~~~~~~~~~~~~~~~~~ 2349 2350``flow validate`` reports whether a flow rule would be accepted by the 2351underlying device in its current state but stops short of creating it. It is 2352bound to ``rte_flow_validate()``:: 2353 2354 flow validate {port_id} 2355 [group {group_id}] [priority {level}] [ingress] [egress] 2356 pattern {item} [/ {item} [...]] / end 2357 actions {action} [/ {action} [...]] / end 2358 2359If successful, it will show:: 2360 2361 Flow rule validated 2362 2363Otherwise it will show an error message of the form:: 2364 2365 Caught error type [...] ([...]): [...] 2366 2367This command uses the same parameters as ``flow create``, their format is 2368described in `Creating flow rules`_. 2369 2370Check whether redirecting any Ethernet packet received on port 0 to RX queue 2371index 6 is supported:: 2372 2373 testpmd> flow validate 0 ingress pattern eth / end 2374 actions queue index 6 / end 2375 Flow rule validated 2376 testpmd> 2377 2378Port 0 does not support TCPv6 rules:: 2379 2380 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end 2381 actions drop / end 2382 Caught error type 9 (specific pattern item): Invalid argument 2383 testpmd> 2384 2385Creating flow rules 2386~~~~~~~~~~~~~~~~~~~ 2387 2388``flow create`` validates and creates the specified flow rule. It is bound 2389to ``rte_flow_create()``:: 2390 2391 flow create {port_id} 2392 [group {group_id}] [priority {level}] [ingress] [egress] 2393 pattern {item} [/ {item} [...]] / end 2394 actions {action} [/ {action} [...]] / end 2395 2396If successful, it will return a flow rule ID usable with other commands:: 2397 2398 Flow rule #[...] created 2399 2400Otherwise it will show an error message of the form:: 2401 2402 Caught error type [...] ([...]): [...] 2403 2404Parameters describe in the following order: 2405 2406- Attributes (*group*, *priority*, *ingress*, *egress* tokens). 2407- A matching pattern, starting with the *pattern* token and terminated by an 2408 *end* pattern item. 2409- Actions, starting with the *actions* token and terminated by an *end* 2410 action. 2411 2412These translate directly to *rte_flow* objects provided as-is to the 2413underlying functions. 2414 2415The shortest valid definition only comprises mandatory tokens:: 2416 2417 testpmd> flow create 0 pattern end actions end 2418 2419Note that PMDs may refuse rules that essentially do nothing such as this 2420one. 2421 2422**All unspecified object values are automatically initialized to 0.** 2423 2424Attributes 2425^^^^^^^^^^ 2426 2427These tokens affect flow rule attributes (``struct rte_flow_attr``) and are 2428specified before the ``pattern`` token. 2429 2430- ``group {group id}``: priority group. 2431- ``priority {level}``: priority level within group. 2432- ``ingress``: rule applies to ingress traffic. 2433- ``egress``: rule applies to egress traffic. 2434 2435Each instance of an attribute specified several times overrides the previous 2436value as shown below (group 4 is used):: 2437 2438 testpmd> flow create 0 group 42 group 24 group 4 [...] 2439 2440Note that once enabled, ``ingress`` and ``egress`` cannot be disabled. 2441 2442While not specifying a direction is an error, some rules may allow both 2443simultaneously. 2444 2445Most rules affect RX therefore contain the ``ingress`` token:: 2446 2447 testpmd> flow create 0 ingress pattern [...] 2448 2449Matching pattern 2450^^^^^^^^^^^^^^^^ 2451 2452A matching pattern starts after the ``pattern`` token. It is made of pattern 2453items and is terminated by a mandatory ``end`` item. 2454 2455Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum 2456rte_flow_item_type``). 2457 2458The ``/`` token is used as a separator between pattern items as shown 2459below:: 2460 2461 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...] 2462 2463Note that protocol items like these must be stacked from lowest to highest 2464layer to make sense. For instance, the following rule is either invalid or 2465unlikely to match any packet:: 2466 2467 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...] 2468 2469More information on these restrictions can be found in the *rte_flow* 2470documentation. 2471 2472Several items support additional specification structures, for example 2473``ipv4`` allows specifying source and destination addresses as follows:: 2474 2475 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1 2476 dst is 10.2.0.0 / end [...] 2477 2478This rule matches all IPv4 traffic with the specified properties. 2479 2480In this example, ``src`` and ``dst`` are field names of the underlying 2481``struct rte_flow_item_ipv4`` object. All item properties can be specified 2482in a similar fashion. 2483 2484The ``is`` token means that the subsequent value must be matched exactly, 2485and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item`` 2486accordingly. Possible assignment tokens are: 2487 2488- ``is``: match value perfectly (with full bit-mask). 2489- ``spec``: match value according to configured bit-mask. 2490- ``last``: specify upper bound to establish a range. 2491- ``mask``: specify bit-mask with relevant bits set to one. 2492- ``prefix``: generate bit-mask from a prefix length. 2493 2494These yield identical results:: 2495 2496 ipv4 src is 10.1.1.1 2497 2498:: 2499 2500 ipv4 src spec 10.1.1.1 src mask 255.255.255.255 2501 2502:: 2503 2504 ipv4 src spec 10.1.1.1 src prefix 32 2505 2506:: 2507 2508 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value 2509 2510:: 2511 2512 ipv4 src is 10.1.1.1 src last 0 # 0 disables range 2513 2514Inclusive ranges can be defined with ``last``:: 2515 2516 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4 2517 2518Note that ``mask`` affects both ``spec`` and ``last``:: 2519 2520 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0 2521 # matches 10.1.0.0 to 10.2.255.255 2522 2523Properties can be modified multiple times:: 2524 2525 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4 2526 2527:: 2528 2529 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16 2530 2531Pattern items 2532^^^^^^^^^^^^^ 2533 2534This section lists supported pattern items and their attributes, if any. 2535 2536- ``end``: end list of pattern items. 2537 2538- ``void``: no-op pattern item. 2539 2540- ``invert``: perform actions when pattern does not match. 2541 2542- ``any``: match any protocol for the current layer. 2543 2544 - ``num {unsigned}``: number of layers covered. 2545 2546- ``pf``: match packets addressed to the physical function. 2547 2548- ``vf``: match packets addressed to a virtual function ID. 2549 2550 - ``id {unsigned}``: destination VF ID. 2551 2552- ``port``: device-specific physical port index to use. 2553 2554 - ``index {unsigned}``: physical port index. 2555 2556- ``raw``: match an arbitrary byte string. 2557 2558 - ``relative {boolean}``: look for pattern after the previous item. 2559 - ``search {boolean}``: search pattern from offset (see also limit). 2560 - ``offset {integer}``: absolute or relative offset for pattern. 2561 - ``limit {unsigned}``: search area limit for start of pattern. 2562 - ``pattern {string}``: byte string to look for. 2563 2564- ``eth``: match Ethernet header. 2565 2566 - ``dst {MAC-48}``: destination MAC. 2567 - ``src {MAC-48}``: source MAC. 2568 - ``type {unsigned}``: EtherType. 2569 2570- ``vlan``: match 802.1Q/ad VLAN tag. 2571 2572 - ``tpid {unsigned}``: tag protocol identifier. 2573 - ``tci {unsigned}``: tag control information. 2574 - ``pcp {unsigned}``: priority code point. 2575 - ``dei {unsigned}``: drop eligible indicator. 2576 - ``vid {unsigned}``: VLAN identifier. 2577 2578- ``ipv4``: match IPv4 header. 2579 2580 - ``tos {unsigned}``: type of service. 2581 - ``ttl {unsigned}``: time to live. 2582 - ``proto {unsigned}``: next protocol ID. 2583 - ``src {ipv4 address}``: source address. 2584 - ``dst {ipv4 address}``: destination address. 2585 2586- ``ipv6``: match IPv6 header. 2587 2588 - ``tc {unsigned}``: traffic class. 2589 - ``flow {unsigned}``: flow label. 2590 - ``proto {unsigned}``: protocol (next header). 2591 - ``hop {unsigned}``: hop limit. 2592 - ``src {ipv6 address}``: source address. 2593 - ``dst {ipv6 address}``: destination address. 2594 2595- ``icmp``: match ICMP header. 2596 2597 - ``type {unsigned}``: ICMP packet type. 2598 - ``code {unsigned}``: ICMP packet code. 2599 2600- ``udp``: match UDP header. 2601 2602 - ``src {unsigned}``: UDP source port. 2603 - ``dst {unsigned}``: UDP destination port. 2604 2605- ``tcp``: match TCP header. 2606 2607 - ``src {unsigned}``: TCP source port. 2608 - ``dst {unsigned}``: TCP destination port. 2609 2610- ``sctp``: match SCTP header. 2611 2612 - ``src {unsigned}``: SCTP source port. 2613 - ``dst {unsigned}``: SCTP destination port. 2614 - ``tag {unsigned}``: validation tag. 2615 - ``cksum {unsigned}``: checksum. 2616 2617- ``vxlan``: match VXLAN header. 2618 2619 - ``vni {unsigned}``: VXLAN identifier. 2620 2621- ``e_tag``: match IEEE 802.1BR E-Tag header. 2622 2623 - ``grp_ecid_b {unsigned}``: GRP and E-CID base. 2624 2625- ``nvgre``: match NVGRE header. 2626 2627 - ``tni {unsigned}``: virtual subnet ID. 2628 2629- ``mpls``: match MPLS header. 2630 2631 - ``label {unsigned}``: MPLS label. 2632 2633- ``gre``: match GRE header. 2634 2635 - ``protocol {unsigned}``: protocol type. 2636 2637- ``fuzzy``: fuzzy pattern match, expect faster than default. 2638 2639 - ``thresh {unsigned}``: accuracy threshold. 2640 2641Actions list 2642^^^^^^^^^^^^ 2643 2644A list of actions starts after the ``actions`` token in the same fashion as 2645`Matching pattern`_; actions are separated by ``/`` tokens and the list is 2646terminated by a mandatory ``end`` action. 2647 2648Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum 2649rte_flow_action_type``). 2650 2651Dropping all incoming UDPv4 packets can be expressed as follows:: 2652 2653 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2654 actions drop / end 2655 2656Several actions have configurable properties which must be specified when 2657there is no valid default value. For example, ``queue`` requires a target 2658queue index. 2659 2660This rule redirects incoming UDPv4 traffic to queue index 6:: 2661 2662 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2663 actions queue index 6 / end 2664 2665While this one could be rejected by PMDs (unspecified queue index):: 2666 2667 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end 2668 actions queue / end 2669 2670As defined by *rte_flow*, the list is not ordered, all actions of a given 2671rule are performed simultaneously. These are equivalent:: 2672 2673 queue index 6 / void / mark id 42 / end 2674 2675:: 2676 2677 void / mark id 42 / queue index 6 / end 2678 2679All actions in a list should have different types, otherwise only the last 2680action of a given type is taken into account:: 2681 2682 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6 2683 2684:: 2685 2686 drop / drop / drop / end # drop is performed only once 2687 2688:: 2689 2690 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24 2691 2692Considering they are performed simultaneously, opposite and overlapping 2693actions can sometimes be combined when the end result is unambiguous:: 2694 2695 drop / queue index 6 / end # drop has no effect 2696 2697:: 2698 2699 drop / dup index 6 / end # same as above 2700 2701:: 2702 2703 queue index 6 / rss queues 6 7 8 / end # queue has no effect 2704 2705:: 2706 2707 drop / passthru / end # drop has no effect 2708 2709Note that PMDs may still refuse such combinations. 2710 2711Actions 2712^^^^^^^ 2713 2714This section lists supported actions and their attributes, if any. 2715 2716- ``end``: end list of actions. 2717 2718- ``void``: no-op action. 2719 2720- ``passthru``: let subsequent rule process matched packets. 2721 2722- ``mark``: attach 32 bit value to packets. 2723 2724 - ``id {unsigned}``: 32 bit value to return with packets. 2725 2726- ``flag``: flag packets. 2727 2728- ``queue``: assign packets to a given queue index. 2729 2730 - ``index {unsigned}``: queue index to use. 2731 2732- ``drop``: drop packets (note: passthru has priority). 2733 2734- ``count``: enable counters for this rule. 2735 2736- ``dup``: duplicate packets to a given queue index. 2737 2738 - ``index {unsigned}``: queue index to duplicate packets to. 2739 2740- ``rss``: spread packets among several queues. 2741 2742 - ``queues [{unsigned} [...]] end``: queue indices to use. 2743 2744- ``pf``: redirect packets to physical device function. 2745 2746- ``vf``: redirect packets to virtual device function. 2747 2748 - ``original {boolean}``: use original VF ID if possible. 2749 - ``id {unsigned}``: VF ID to redirect packets to. 2750 2751Destroying flow rules 2752~~~~~~~~~~~~~~~~~~~~~ 2753 2754``flow destroy`` destroys one or more rules from their rule ID (as returned 2755by ``flow create``), this command calls ``rte_flow_destroy()`` as many 2756times as necessary:: 2757 2758 flow destroy {port_id} rule {rule_id} [...] 2759 2760If successful, it will show:: 2761 2762 Flow rule #[...] destroyed 2763 2764It does not report anything for rule IDs that do not exist. The usual error 2765message is shown when a rule cannot be destroyed:: 2766 2767 Caught error type [...] ([...]): [...] 2768 2769``flow flush`` destroys all rules on a device and does not take extra 2770arguments. It is bound to ``rte_flow_flush()``:: 2771 2772 flow flush {port_id} 2773 2774Any errors are reported as above. 2775 2776Creating several rules and destroying them:: 2777 2778 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2779 actions queue index 2 / end 2780 Flow rule #0 created 2781 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2782 actions queue index 3 / end 2783 Flow rule #1 created 2784 testpmd> flow destroy 0 rule 0 rule 1 2785 Flow rule #1 destroyed 2786 Flow rule #0 destroyed 2787 testpmd> 2788 2789The same result can be achieved using ``flow flush``:: 2790 2791 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2792 actions queue index 2 / end 2793 Flow rule #0 created 2794 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2795 actions queue index 3 / end 2796 Flow rule #1 created 2797 testpmd> flow flush 0 2798 testpmd> 2799 2800Non-existent rule IDs are ignored:: 2801 2802 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2803 actions queue index 2 / end 2804 Flow rule #0 created 2805 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2806 actions queue index 3 / end 2807 Flow rule #1 created 2808 testpmd> flow destroy 0 rule 42 rule 10 rule 2 2809 testpmd> 2810 testpmd> flow destroy 0 rule 0 2811 Flow rule #0 destroyed 2812 testpmd> 2813 2814Querying flow rules 2815~~~~~~~~~~~~~~~~~~~ 2816 2817``flow query`` queries a specific action of a flow rule having that 2818ability. Such actions collect information that can be reported using this 2819command. It is bound to ``rte_flow_query()``:: 2820 2821 flow query {port_id} {rule_id} {action} 2822 2823If successful, it will display either the retrieved data for known actions 2824or the following message:: 2825 2826 Cannot display result for action type [...] ([...]) 2827 2828Otherwise, it will complain either that the rule does not exist or that some 2829error occurred:: 2830 2831 Flow rule #[...] not found 2832 2833:: 2834 2835 Caught error type [...] ([...]): [...] 2836 2837Currently only the ``count`` action is supported. This action reports the 2838number of packets that hit the flow rule and the total number of bytes. Its 2839output has the following format:: 2840 2841 count: 2842 hits_set: [...] # whether "hits" contains a valid value 2843 bytes_set: [...] # whether "bytes" contains a valid value 2844 hits: [...] # number of packets 2845 bytes: [...] # number of bytes 2846 2847Querying counters for TCPv6 packets redirected to queue 6:: 2848 2849 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end 2850 actions queue index 6 / count / end 2851 Flow rule #4 created 2852 testpmd> flow query 0 4 count 2853 count: 2854 hits_set: 1 2855 bytes_set: 0 2856 hits: 386446 2857 bytes: 0 2858 testpmd> 2859 2860Listing flow rules 2861~~~~~~~~~~~~~~~~~~ 2862 2863``flow list`` lists existing flow rules sorted by priority and optionally 2864filtered by group identifiers:: 2865 2866 flow list {port_id} [group {group_id}] [...] 2867 2868This command only fails with the following message if the device does not 2869exist:: 2870 2871 Invalid port [...] 2872 2873Output consists of a header line followed by a short description of each 2874flow rule, one per line. There is no output at all when no flow rules are 2875configured on the device:: 2876 2877 ID Group Prio Attr Rule 2878 [...] [...] [...] [...] [...] 2879 2880``Attr`` column flags: 2881 2882- ``i`` for ``ingress``. 2883- ``e`` for ``egress``. 2884 2885Creating several flow rules and listing them:: 2886 2887 testpmd> flow create 0 ingress pattern eth / ipv4 / end 2888 actions queue index 6 / end 2889 Flow rule #0 created 2890 testpmd> flow create 0 ingress pattern eth / ipv6 / end 2891 actions queue index 2 / end 2892 Flow rule #1 created 2893 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end 2894 actions rss queues 6 7 8 end / end 2895 Flow rule #2 created 2896 testpmd> flow list 0 2897 ID Group Prio Attr Rule 2898 0 0 0 i- ETH IPV4 => QUEUE 2899 1 0 0 i- ETH IPV6 => QUEUE 2900 2 0 5 i- ETH IPV4 UDP => RSS 2901 testpmd> 2902 2903Rules are sorted by priority (i.e. group ID first, then priority level):: 2904 2905 testpmd> flow list 1 2906 ID Group Prio Attr Rule 2907 0 0 0 i- ETH => COUNT 2908 6 0 500 i- ETH IPV6 TCP => DROP COUNT 2909 5 0 1000 i- ETH IPV6 ICMP => QUEUE 2910 1 24 0 i- ETH IPV4 UDP => QUEUE 2911 4 24 10 i- ETH IPV4 TCP => DROP 2912 3 24 20 i- ETH IPV4 => DROP 2913 2 24 42 i- ETH IPV4 UDP => QUEUE 2914 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE 2915 testpmd> 2916 2917Output can be limited to specific groups:: 2918 2919 testpmd> flow list 1 group 0 group 63 2920 ID Group Prio Attr Rule 2921 0 0 0 i- ETH => COUNT 2922 6 0 500 i- ETH IPV6 TCP => DROP COUNT 2923 5 0 1000 i- ETH IPV6 ICMP => QUEUE 2924 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE 2925 testpmd> 2926 2927Toggling isolated mode 2928~~~~~~~~~~~~~~~~~~~~~~ 2929 2930``flow isolate`` can be used to tell the underlying PMD that ingress traffic 2931must only be injected from the defined flow rules; that no default traffic 2932is expected outside those rules and the driver is free to assign more 2933resources to handle them. It is bound to ``rte_flow_isolate()``:: 2934 2935 flow isolate {port_id} {boolean} 2936 2937If successful, enabling or disabling isolated mode shows either:: 2938 2939 Ingress traffic on port [...] 2940 is now restricted to the defined flow rules 2941 2942Or:: 2943 2944 Ingress traffic on port [...] 2945 is not restricted anymore to the defined flow rules 2946 2947Otherwise, in case of error:: 2948 2949 Caught error type [...] ([...]): [...] 2950 2951Mainly due to its side effects, PMDs supporting this mode may not have the 2952ability to toggle it more than once without reinitializing affected ports 2953first (e.g. by exiting testpmd). 2954 2955Enabling isolated mode:: 2956 2957 testpmd> flow isolate 0 true 2958 Ingress traffic on port 0 is now restricted to the defined flow rules 2959 testpmd> 2960 2961Disabling isolated mode:: 2962 2963 testpmd> flow isolate 0 false 2964 Ingress traffic on port 0 is not restricted anymore to the defined flow rules 2965 testpmd> 2966 2967Sample QinQ flow rules 2968~~~~~~~~~~~~~~~~~~~~~~ 2969 2970Before creating QinQ rule(s) the following commands should be issued to enable QinQ:: 2971 2972 testpmd> port stop 0 2973 testpmd> vlan set qinq on 0 2974 2975The above command sets the inner and outer TPID's to 0x8100. 2976 2977To change the TPID's the following commands should be used:: 2978 2979 testpmd> vlan set outer tpid 0xa100 0 2980 testpmd> vlan set inner tpid 0x9100 0 2981 testpmd> port start 0 2982 2983Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM. 2984 2985:: 2986 2987 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 / 2988 vlan tci is 456 / end actions vf id 1 / queue index 0 / end 2989 Flow rule #0 validated 2990 2991 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 / 2992 vlan tci is 456 / end actions vf id 123 / queue index 0 / end 2993 Flow rule #0 created 2994 2995 testpmd> flow list 0 2996 ID Group Prio Attr Rule 2997 0 0 0 i- ETH VLAN VLAN=>VF QUEUE 2998 2999Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host. 3000 3001:: 3002 3003 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 / 3004 vlan tci is 654 / end actions pf / queue index 0 / end 3005 Flow rule #1 validated 3006 3007 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 / 3008 vlan tci is 654 / end actions pf / queue index 1 / end 3009 Flow rule #1 created 3010 3011 testpmd> flow list 0 3012 ID Group Prio Attr Rule 3013 0 0 0 i- ETH VLAN VLAN=>VF QUEUE 3014 1 0 0 i- ETH VLAN VLAN=>PF QUEUE 3015