1.. BSD LICENSE 2 Copyright(c) 2015-2016 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 31Internet Protocol (IP) Pipeline Application 32=========================================== 33 34Application overview 35-------------------- 36 37The *Internet Protocol (IP) Pipeline* application is intended to be a vehicle for rapid development of packet processing 38applications running on multi-core CPUs. 39 40The application provides a library of reusable functional blocks called pipelines. 41These pipelines can be seen as prefabricated blocks that can be instantiated and inter-connected through packet queues 42to create complete applications (super-pipelines). 43 44Pipelines are created and inter-connected through the application configuration file. 45By using different configuration files, different applications are effectively created, therefore this application 46can be seen as an application generator. 47The configuration of each pipeline can be updated at run-time through the application Command Line Interface (CLI). 48 49Main application components are: 50 51**A Library of reusable pipelines** 52 53 * Each pipeline represents a functional block, e.g. flow classification, firewall, routing, master, etc. 54 55 * Each pipeline type can be instantiated several times in the same application, which each instance configured 56 separately and mapped to a single CPU core. 57 Each CPU core can run one or several pipeline instances, which can be of same or different type. 58 59 * Pipeline instances are inter-connected through packet queues (for packet processing) and message queues 60 (for run-time configuration). 61 62 * Pipelines are implemented using DPDK Packet Framework. 63 64 * More pipeline types can always be built and added to the existing pipeline types. 65 66**The Configuration file** 67 68 * The configuration file defines the application structure. 69 By using different configuration files, different applications are created. 70 71 * All the application resources are created and configured through the application configuration file: 72 pipeline instances, buffer pools, links (i.e. network interfaces), hardware device RX/TX queues, 73 software queues, traffic manager devices, EAL startup arguments, etc. 74 75 * The configuration file syntax is “define by reference”, meaning that resources are defined as they are referenced. 76 First time a resource name is detected, it is registered with default parameters. 77 Optionally, the resource parameters can be further refined through a configuration file section dedicated to 78 that resource. 79 80 * Command Line Interface (CLI) 81 82**Global CLI commands: link configuration, etc.** 83 84 * Common pipeline CLI commands: ping (keep-alive), statistics, etc. 85 86 * Pipeline type specific CLI commands: used to configure instances of specific pipeline type. 87 These commands are registered with the application when the pipeline type is registered. 88 For example, the commands for routing pipeline instances include: route add, route delete, route list, etc. 89 90 * CLI commands can be grouped into scripts that can be invoked at initialization and at runtime. 91 92 93Design goals 94------------ 95 96 97Rapid development 98~~~~~~~~~~~~~~~~~ 99 100This application enables rapid development through quick connectivity of standard components called pipelines. 101These components are built using DPDK Packet Framework and encapsulate packet processing features at different levels: 102ports, tables, actions, pipelines and complete applications. 103 104Pipeline instances are instantiated, configured and inter-connected through low complexity configuration files loaded 105during application initialization. 106Each pipeline instance is mapped to a single CPU core, with each CPU core able to run one or multiple pipeline 107instances of same or different types. By loading a different configuration file, a different application is 108effectively started. 109 110 111Flexibility 112~~~~~~~~~~~ 113 114Each packet processing application is typically represented as a chain of functional stages which is often called 115the functional pipeline of the application. 116These stages are mapped to CPU cores to create chains of CPU cores (pipeline model), clusters of CPU cores 117(run-to-completion model) or chains of clusters of CPU cores (hybrid model). 118 119This application allows all the above programming models. 120By applying changes to the configuration file, the application provides the flexibility to reshuffle its 121building blocks in different ways until the configuration providing the best performance is identified. 122 123 124Move pipelines around 125^^^^^^^^^^^^^^^^^^^^^ 126 127The mapping of pipeline instances to CPU cores can be reshuffled through the configuration file. 128One or several pipeline instances can be mapped to the same CPU core. 129 130.. _figure_ip_pipelines_1: 131 132.. figure:: img/ip_pipelines_1.* 133 134 Example of moving pipeline instances across different CPU cores 135 136 137Move tables around 138^^^^^^^^^^^^^^^^^^ 139 140There is some degree of flexibility for moving tables from one pipeline instance to another. 141Based on the configuration arguments passed to each pipeline instance in the configuration file, specific tables 142can be enabled or disabled. 143This way, a specific table can be “moved” from pipeline instance A to pipeline instance B by simply disabling its 144associated functionality for pipeline instance A while enabling it for pipeline instance B. 145 146Due to requirement to have simple syntax for the configuration file, moving tables across different pipeline 147instances is not as flexible as the mapping of pipeline instances to CPU cores, or mapping actions to pipeline tables. 148Complete flexibility in moving tables from one pipeline to another could be achieved through a complex pipeline 149description language that would detail the structural elements of the pipeline (ports, tables and actions) and 150their connectivity, resulting in complex syntax for the configuration file, which is not acceptable. 151Good configuration file readability through simple syntax is preferred. 152 153*Example*: the IP routing pipeline can run the routing function only (with ARP function run by a different 154pipeline instance), or it can run both the routing and ARP functions as part of the same pipeline instance. 155 156 157.. _figure_ip_pipelines_2: 158 159.. figure:: img/ip_pipelines_2.* 160 161 Example of moving tables across different pipeline instances 162 163 164Move actions around 165^^^^^^^^^^^^^^^^^^^ 166 167When it makes sense, packet processing actions can be moved from one pipeline instance to another. 168Based on the configuration arguments passed to each pipeline instance in the configuration file, specific actions 169can be enabled or disabled. 170This way, a specific action can be "moved" from pipeline instance A to pipeline instance B by simply disabling its 171associated functionality for pipeline instance A while enabling it for pipeline instance B. 172 173*Example*: The flow actions of accounting, traffic metering, application identification, NAT, etc can be run as part 174of the flow classification pipeline instance or split across several flow actions pipeline instances, depending on 175the number of flow instances and their compute requirements. 176 177 178.. _figure_ip_pipelines_3: 179 180.. figure:: img/ip_pipelines_3.* 181 182 Example of moving actions across different tables and pipeline instances 183 184 185Performance 186~~~~~~~~~~~ 187 188Performance of the application is the highest priority requirement. 189Flexibility is not provided at the expense of performance. 190 191The purpose of flexibility is to provide an incremental development methodology that allows monitoring the 192performance evolution: 193 194* Apply incremental changes in the configuration (e.g. mapping on pipeline instances to CPU cores) 195 in order to identify the configuration providing the best performance for a given application; 196 197* Add more processing incrementally (e.g. by enabling more actions for specific pipeline instances) until 198 the application is feature complete while checking the performance impact at each step. 199 200 201Debug capabilities 202~~~~~~~~~~~~~~~~~~ 203 204The application provides a significant set of debug capabilities: 205 206* Command Line Interface (CLI) support for statistics polling: pipeline instance ping (keep-alive checks), 207 pipeline instance statistics per input port/output port/table, link statistics, etc; 208 209* Logging: Turn on/off application log messages based on priority level; 210 211Running the application 212----------------------- 213 214The application startup command line is:: 215 216 ip_pipeline [-f CONFIG_FILE] [-s SCRIPT_FILE] -p PORT_MASK [-l LOG_LEVEL] 217 218The application startup arguments are: 219 220``-f CONFIG_FILE`` 221 222 * Optional: Yes 223 224 * Default: ``./config/ip_pipeline.cfg`` 225 226 * Argument: Path to the configuration file to be loaded by the application. 227 Please refer to the :ref:`ip_pipeline_configuration_file` for details on how to write the configuration file. 228 229``-s SCRIPT_FILE`` 230 231 * Optional: Yes 232 233 * Default: Not present 234 235 * Argument: Path to the CLI script file to be run by the master pipeline at application startup. 236 No CLI script file will be run at startup of this argument is not present. 237 238``-p PORT_MASK`` 239 240 * Optional: No 241 242 * Default: N/A 243 244 * Argument: Hexadecimal mask of NIC port IDs to be used by the application. 245 First port enabled in this mask will be referenced as LINK0 as part of the application configuration file, 246 next port as LINK1, etc. 247 248``-l LOG_LEVEL`` 249 250 * Optional: Yes 251 252 * Default: 1 (High priority) 253 254 * Argument: Log level to determine which application messages are to be printed to standard output. 255 Available log levels are: 0 (None), 1 (High priority), 2 (Low priority). 256 Only application messages whose priority is higher than or equal to the application log level will be printed. 257 258 259Application stages 260------------------ 261 262 263Configuration 264~~~~~~~~~~~~~ 265 266During this stage, the application configuration file is parsed and its content is loaded into the application data 267structures. 268In case of any configuration file parse error, an error message is displayed and the application is terminated. 269Please refer to the :ref:`ip_pipeline_configuration_file` for a description of the application configuration file format. 270 271 272Configuration checking 273~~~~~~~~~~~~~~~~~~~~~~ 274 275In the absence of any parse errors, the loaded content of application data structures is checked for overall consistency. 276In case of any configuration check error, an error message is displayed and the application is terminated. 277 278 279Initialization 280~~~~~~~~~~~~~~ 281 282During this stage, the application resources are initialized and the handles to access them are saved into the 283application data structures. 284In case of any initialization error, an error message is displayed and the application is terminated. 285 286The typical resources to be initialized are: pipeline instances, buffer pools, links (i.e. network interfaces), 287hardware device RX/TX queues, software queues, traffic management devices, etc. 288 289 290.. _ip_pipeline_runtime: 291 292Run-time 293~~~~~~~~ 294 295Each CPU core runs the pipeline instances assigned to it in time sharing mode and in round robin order: 296 2971. *Packet processing task*: The pipeline run-time code is typically a packet *processing* task built on top of 298 DPDK Packet Framework rte_pipeline library, which reads bursts of packets from the pipeline input ports, 299 performs table lookups and executes the identified actions for all tables in the pipeline, with packet 300 eventually written to pipeline output ports or dropped. 301 3022. *Message handling task*: Each CPU core will also periodically execute the *message handling* code of each 303 of the pipelines mapped to it. 304 The pipeline message handling code is processing the messages that are pending in the pipeline input message 305 queues, which are typically sent by the master CPU core for the on-the-fly pipeline configuration: check 306 that pipeline is still alive (ping), add/delete entries in the pipeline tables, get statistics, etc. 307 The frequency of executing the message handling code is usually much smaller than the frequency of executing 308 the packet processing work. 309 310Please refer to the :ref:`ip_pipeline_pipeline_section` for more details about the application pipeline module encapsulation. 311 312.. _ip_pipeline_configuration_file: 313 314Configuration file syntax 315------------------------- 316 317Syntax overview 318~~~~~~~~~~~~~~~ 319 320The syntax of the configuration file is designed to be simple, which favors readability. 321The configuration file is parsed using the DPDK library librte_cfgfile, which supports simple 322`INI file format <http://en.wikipedia.org/wiki/INI_file>`__ for configuration files. 323 324As result, the configuration file is split into several sections, with each section containing one or more entries. 325The scope of each entry is its section, and each entry specifies a variable that is assigned a specific value. 326Any text after the ``;`` character is considered a comment and is therefore ignored. 327 328The following are application specific: number of sections, name of each section, number of entries of each section, 329name of the variables used for each section entry, the value format (e.g. signed/unsigned integer, string, etc) 330and range of each section entry variable. 331 332Generic example of configuration file section: 333 334.. code-block:: ini 335 336 [<section_name>] 337 338 <variable_name_1> = <value_1> 339 340 ; ... 341 342 <variable_name_N> = <value_N> 343 344 345Application resources present in the configuration file 346~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 347 348.. _table_ip_pipelines_resource_name: 349 350.. tabularcolumns:: |p{4cm}|p{6cm}|p{6cm}| 351 352.. table:: Application resource names in the configuration file 353 354 +----------------------------+-----------------------------+-------------------------------------------------+ 355 | Resource type | Format | Examples | 356 +============================+=============================+=================================================+ 357 | Pipeline | ``PIPELINE<ID>`` | ``PIPELINE0``, ``PIPELINE1`` | 358 +----------------------------+-----------------------------+-------------------------------------------------+ 359 | Mempool | ``MEMPOOL<ID>`` | ``MEMPOOL0``, ``MEMPOOL1`` | 360 +----------------------------+-----------------------------+-------------------------------------------------+ 361 | Link (network interface) | ``LINK<ID>`` | ``LINK0``, ``LINK1`` | 362 +----------------------------+-----------------------------+-------------------------------------------------+ 363 | Link RX queue | ``RXQ<LINK_ID>.<QUEUE_ID>`` | ``RXQ0.0``, ``RXQ1.5`` | 364 +----------------------------+-----------------------------+-------------------------------------------------+ 365 | Link TX queue | ``TXQ<LINK_ID>.<QUEUE_ID>`` | ``TXQ0.0``, ``TXQ1.5`` | 366 +----------------------------+-----------------------------+-------------------------------------------------+ 367 | Software queue | ``SWQ<ID>`` | ``SWQ0``, ``SWQ1`` | 368 +----------------------------+-----------------------------+-------------------------------------------------+ 369 | Traffic Manager | ``TM<LINK_ID>`` | ``TM0``, ``TM1`` | 370 +----------------------------+-----------------------------+-------------------------------------------------+ 371 | KNI (kernel NIC interface) | ``KNI<LINK_ID>`` | ``KNI0``, ``KNI1`` | 372 +----------------------------+-----------------------------+-------------------------------------------------+ 373 | Source | ``SOURCE<ID>`` | ``SOURCE0``, ``SOURCE1`` | 374 +----------------------------+-----------------------------+-------------------------------------------------+ 375 | Sink | ``SINK<ID>`` | ``SINK0``, ``SINK1`` | 376 +----------------------------+-----------------------------+-------------------------------------------------+ 377 | Message queue | ``MSGQ<ID>`` | ``MSGQ0``, ``MSGQ1``, | 378 | | ``MSGQ-REQ-PIPELINE<ID>`` | ``MSGQ-REQ-PIPELINE2``, ``MSGQ-RSP-PIPELINE2,`` | 379 | | ``MSGQ-RSP-PIPELINE<ID>`` | ``MSGQ-REQ-CORE-s0c1``, ``MSGQ-RSP-CORE-s0c1`` | 380 | | ``MSGQ-REQ-CORE-<CORE_ID>`` | | 381 | | ``MSGQ-RSP-CORE-<CORE_ID>`` | | 382 +----------------------------+-----------------------------+-------------------------------------------------+ 383 384``LINK`` instances are created implicitly based on the ``PORT_MASK`` application startup argument. 385``LINK0`` is the first port enabled in the ``PORT_MASK``, port 1 is the next one, etc. 386The LINK ID is different than the DPDK PMD-level NIC port ID, which is the actual position in the bitmask mentioned above. 387For example, if bit 5 is the first bit set in the bitmask, then ``LINK0`` is having the PMD ID of 5. 388This mechanism creates a contiguous LINK ID space and isolates the configuration file against changes in the board 389PCIe slots where NICs are plugged in. 390 391``RXQ``, ``TXQ``, ``TM`` and ``KNI`` instances have the LINK ID as part of their name. 392For example, ``RXQ2.1``, ``TXQ2.1`` and ``TM2`` are all associated with ``LINK2``. 393 394 395Rules to parse the configuration file 396~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 397 398The main rules used to parse the configuration file are: 399 4001. Application resource name determines the type of resource based on the name prefix. 401 402 *Example*: all software queues need to start with ``SWQ`` prefix, so ``SWQ0`` and ``SWQ5`` are valid software 403 queue names. 404 4052. An application resource is defined by creating a configuration file section with its name. 406 The configuration file section allows fine tuning on any of the resource parameters. 407 Some resource parameters are mandatory, in which case it is required to have them specified as part of the 408 section, while some others are optional, in which case they get assigned their default value when not present. 409 410 *Example*: section ``SWQ0`` defines a software queue named SWQ0, whose parameters are detailed as part of this section. 411 4123. An application resource can also be defined by referencing it. 413 Referencing a resource takes place by simply using its name as part of the value assigned to a variable in any 414 configuration file section. 415 In this case, the resource is registered with all its parameters having their default values. 416 Optionally, a section with the resource name can be added to the configuration file to fine tune some or all 417 of the resource parameters. 418 419 *Example*: in section ``PIPELINE3``, variable ``pktq_in`` includes ``SWQ5`` as part of its list, which results 420 in defining a software queue named ``SWQ5``; when there is no ``SWQ5`` section present in the configuration file, 421 ``SWQ5`` gets registered with default parameters. 422 423 424.. _ip_pipeline_pipeline_section: 425 426PIPELINE section 427~~~~~~~~~~~~~~~~ 428 429.. _table_ip_pipelines_pipeline_section_1: 430 431.. tabularcolumns:: |p{2.5cm}|p{7cm}|p{1.5cm}|p{1.5cm}|p{1.5cm}| 432 433.. table:: Configuration file PIPELINE section (1/2) 434 435 +---------------+-----------------------------------------------------------+---------------+------------------------+----------------+ 436 | Section | Description | Optional | Range | Default value | 437 +===============+===========================================================+===============+========================+================+ 438 | type | Pipeline type. Defines the functionality to be | NO | See "List | N/A | 439 | | executed. | | of pipeline types" | | 440 +---------------+-----------------------------------------------------------+---------------+------------------------+----------------+ 441 | core | CPU core to run the current pipeline. | YES | See "CPU Core | CPU socket 0, | 442 | | | | notation" | core 0, | 443 | | | | | hyper-thread 0 | 444 +---------------+-----------------------------------------------------------+---------------+------------------------+----------------+ 445 | pktq_in | Packet queues to serve as input ports for the | YES | List of input | Empty list | 446 | | current pipeline instance. The acceptable packet | | packet queue IDs | | 447 | | queue types are: ``RXQ``, ``SWQ``, ``TM`` and ``SOURCE``. | | | | 448 | | First device in this list is used as pipeline input port | | | | 449 | | 0, second as pipeline input port 1, etc. | | | | 450 +---------------+-----------------------------------------------------------+---------------+------------------------+----------------+ 451 | pktq_out | Packet queues to serve as output ports for the | YES | List of output | Empty list | 452 | | current pipeline instance. The acceptable packet | | packet queue IDs. | | 453 | | queue types are: ``TXQ``, ``SWQ``, ``TM`` and ``SINK``. | | | | 454 | | First device in this list is used as pipeline output | | | | 455 | | port 0, second as pipeline output port 1, etc. | | | | 456 +---------------+-----------------------------------------------------------+---------------+------------------------+----------------+ 457 458.. _table_ip_pipelines_pipeline_section_2: 459 460.. tabularcolumns:: |p{2.5cm}|p{7cm}|p{1.5cm}|p{1.5cm}|p{1.5cm}| 461 462.. table:: Configuration file PIPELINE section (2/2) 463 464 +---------------+-----------------------------------------------------------+---------------+------------------------+----------------+ 465 | Section | Description | Optional | Range | Default value | 466 +===============+===========================================================+===============+========================+================+ 467 | msgq_in | Input message queues. These queues contain | YES | List of message | Empty list | 468 | | request messages that need to be handled by the | | queue IDs | | 469 | | current pipeline instance. The type and format of | | | | 470 | | request messages is defined by the pipeline type. | | | | 471 | | For each pipeline instance, there is an input | | | | 472 | | message queue defined implicitly, whose name is: | | | | 473 | | ``MSGQ-REQ-<PIPELINE_ID>``. This message queue | | | | 474 | | should not be mentioned as part of msgq_in list. | | | | 475 +---------------+-----------------------------------------------------------+---------------+------------------------+----------------+ 476 | msgq_out | Output message queues. These queues are used by | YES | List of message | Empty list | 477 | | the current pipeline instance to write response | | queue IDs | | 478 | | messages as result of request messages being | | | | 479 | | handled. The type and format of response | | | | 480 | | messages is defined by the pipeline type. | | | | 481 | | For each pipeline instance, there is an output | | | | 482 | | message queue defined implicitly, whose name is: | | | | 483 | | ``MSGQ-RSP-<PIPELINE_ID>``. This message queue | | | | 484 | | should not be mentioned as part of msgq_out list. | | | | 485 +---------------+-----------------------------------------------------------+---------------+------------------------+----------------+ 486 | timer_period | Time period, measured in milliseconds, | YES | milliseconds | 1 ms | 487 | | for handling the input message queues. | | | | 488 +---------------+-----------------------------------------------------------+---------------+------------------------+----------------+ 489 | <any other> | Arguments to be passed to the current pipeline | Depends on | Depends on | Depends on | 490 | | instance. Format of the arguments, their type, | pipeline type | pipeline type | pipeline type | 491 | | whether each argument is optional or mandatory | | | | 492 | | and its default value (when optional) are defined | | | | 493 | | by the pipeline type. | | | | 494 | | The value of the arguments is applicable to the | | | | 495 | | current pipeline instance only. | | | | 496 +---------------+-----------------------------------------------------------+---------------+------------------------+----------------+ 497 498 499CPU core notation 500^^^^^^^^^^^^^^^^^ 501 502The CPU Core notation is:: 503 504 <CPU core> ::= [s|S<CPU socket ID>][c|C]<CPU core ID>[h|H] 505 506For example:: 507 508 CPU socket 0, core 0, hyper-thread 0: 0, c0, s0c0 509 510 CPU socket 0, core 0, hyper-thread 1: 0h, c0h, s0c0h 511 512 CPU socket 3, core 9, hyper-thread 1: s3c9h 513 514 515MEMPOOL section 516~~~~~~~~~~~~~~~ 517 518.. _table_ip_pipelines_mempool_section: 519 520.. tabularcolumns:: |p{2.5cm}|p{6cm}|p{1.5cm}|p{1.5cm}|p{3cm}| 521 522.. table:: Configuration file MEMPOOL section 523 524 +---------------+-----------------------------------------------+----------+----------+---------------------------+ 525 | Section | Description | Optional | Type | Default value | 526 +===============+===============================================+==========+==========+===========================+ 527 | buffer_size | Buffer size (in bytes) for the current | YES | uint32_t | 2048 | 528 | | buffer pool. | | | + sizeof(struct rte_mbuf) | 529 | | | | | + HEADROOM | 530 +---------------+-----------------------------------------------+----------+----------+---------------------------+ 531 | pool_size | Number of buffers in the current buffer pool. | YES | uint32_t | 32K | 532 +---------------+-----------------------------------------------+----------+----------+---------------------------+ 533 | cache_size | Per CPU thread cache size (in number of | YES | uint32_t | 256 | 534 | | buffers) for the current buffer pool. | | | | 535 +---------------+-----------------------------------------------+----------+----------+---------------------------+ 536 | cpu | CPU socket ID where to allocate memory for | YES | uint32_t | 0 | 537 | | the current buffer pool. | | | | 538 +---------------+-----------------------------------------------+----------+----------+---------------------------+ 539 540 541LINK section 542~~~~~~~~~~~~ 543 544.. _table_ip_pipelines_link_section: 545 546.. tabularcolumns:: |p{3cm}|p{7cm}|p{1.5cm}|p{1.5cm}|p{2cm}| 547 548.. table:: Configuration file LINK section 549 550 +-----------------+----------------------------------------------+----------+----------+-------------------+ 551 | Section entry | Description | Optional | Type | Default value | 552 +=================+==============================================+==========+==========+===================+ 553 | arp_q | NIC RX queue where ARP packets should | YES | 0 .. 127 | 0 (default queue) | 554 | | be filtered. | | | | 555 +-----------------+----------------------------------------------+----------+----------+-------------------+ 556 | tcp_syn_local_q | NIC RX queue where TCP packets with SYN | YES | 0 .. 127 | 0 (default queue) | 557 | | flag should be filtered. | | | | 558 +-----------------+----------------------------------------------+----------+----------+-------------------+ 559 | ip_local_q | NIC RX queue where IP packets with local | YES | 0 .. 127 | 0 (default queue) | 560 | | destination should be filtered. | | | | 561 | | When TCP, UDP and SCTP local queues are | | | | 562 | | defined, they take higher priority than this | | | | 563 | | queue. | | | | 564 +-----------------+----------------------------------------------+----------+----------+-------------------+ 565 | tcp_local_q | NIC RX queue where TCP packets with local | YES | 0 .. 127 | 0 (default queue) | 566 | | destination should be filtered. | | | | 567 +-----------------+----------------------------------------------+----------+----------+-------------------+ 568 | udp_local_q | NIC RX queue where TCP packets with local | YES | 0 .. 127 | 0 (default queue) | 569 | | destination should be filtered. | | | | 570 +-----------------+----------------------------------------------+----------+----------+-------------------+ 571 | sctp_local_q | NIC RX queue where TCP packets with local | YES | 0 .. 127 | 0 (default queue) | 572 | | destination should be filtered. | | | | 573 +-----------------+----------------------------------------------+----------+----------+-------------------+ 574 | promisc | Indicates whether current link should be | YES | YES/NO | YES | 575 | | started in promiscuous mode. | | | | 576 +-----------------+----------------------------------------------+----------+----------+-------------------+ 577 578 579RXQ section 580~~~~~~~~~~~ 581 582.. _table_ip_pipelines_rxq_section: 583 584.. tabularcolumns:: |p{3cm}|p{7cm}|p{1.5cm}|p{1.5cm}|p{2cm}| 585 586.. table:: Configuration file RXQ section 587 588 +---------------+--------------------------------------------+----------+----------+---------------+ 589 | Section | Description | Optional | Type | Default value | 590 +===============+============================================+==========+==========+===============+ 591 | mempool | Mempool to use for buffer allocation for | YES | uint32_t | MEMPOOL0 | 592 | | current NIC RX queue. The mempool ID has | | | | 593 | | to be associated with a valid instance | | | | 594 | | defined in the mempool entry of the global | | | | 595 | | section. | | | | 596 +---------------+--------------------------------------------+----------+----------+---------------+ 597 | Size | NIC RX queue size (number of descriptors) | YES | uint32_t | 128 | 598 +---------------+--------------------------------------------+----------+----------+---------------+ 599 | burst | Read burst size (number of descriptors) | YES | uint32_t | 32 | 600 +---------------+--------------------------------------------+----------+----------+---------------+ 601 602 603TXQ section 604~~~~~~~~~~~ 605 606.. _table_ip_pipelines_txq_section: 607 608.. tabularcolumns:: |p{2.5cm}|p{7cm}|p{1.5cm}|p{2cm}|p{1.5cm}| 609 610.. table:: Configuration file TXQ section 611 612 +---------------+----------------------------------------------+----------+------------------+---------------+ 613 | Section | Description | Optional | Type | Default value | 614 +===============+==============================================+==========+==================+===============+ 615 | size | NIC TX queue size (number of descriptors) | YES | uint32_t | 512 | 616 | | | | power of 2 | | 617 | | | | > 0 | | 618 +---------------+----------------------------------------------+----------+------------------+---------------+ 619 | burst | Write burst size (number of descriptors) | YES | uint32_t | 32 | 620 | | | | power of 2 | | 621 | | | | 0 < burst < size | | 622 +---------------+----------------------------------------------+----------+------------------+---------------+ 623 | dropless | When dropless is set to NO, packets can be | YES | YES/NO | NO | 624 | | dropped if not enough free slots are | | | | 625 | | currently available in the queue, so the | | | | 626 | | write operation to the queue is non- | | | | 627 | | blocking. | | | | 628 | | When dropless is set to YES, packets cannot | | | | 629 | | be dropped if not enough free slots are | | | | 630 | | currently available in the queue, so the | | | | 631 | | write operation to the queue is blocking, as | | | | 632 | | the write operation is retried until enough | | | | 633 | | free slots become available and all the | | | | 634 | | packets are successfully written to the | | | | 635 | | queue. | | | | 636 +---------------+----------------------------------------------+----------+------------------+---------------+ 637 | n_retries | Number of retries. Valid only when dropless | YES | uint32_t | 0 | 638 | | is set to YES. When set to 0, it indicates | | | | 639 | | unlimited number of retries. | | | | 640 +---------------+----------------------------------------------+----------+------------------+---------------+ 641 642 643SWQ section 644~~~~~~~~~~~ 645 646.. _table_ip_pipelines_swq_section: 647 648.. tabularcolumns:: |p{2.5cm}|p{7cm}|p{1.5cm}|p{1.5cm}|p{1.5cm}| 649 650.. table:: Configuration file SWQ section 651 652 +---------------+----------------------------------------------+----------+------------------+---------------+ 653 | Section | Description | Optional | Type | Default value | 654 +===============+==============================================+==========+==================+===============+ 655 | size | Queue size (number of packets) | YES | uint32_t | 256 | 656 | | | | power of 2 | | 657 +---------------+----------------------------------------------+----------+------------------+---------------+ 658 | burst_read | Read burst size (number of packets) | YES | uint32_t | 32 | 659 | | | | power of 2 | | 660 | | | | 0 < burst < size | | 661 +---------------+----------------------------------------------+----------+------------------+---------------+ 662 | burst_write | Write burst size (number of packets) | YES | uint32_t | 32 | 663 | | | | power of 2 | | 664 | | | | 0 < burst < size | | 665 +---------------+----------------------------------------------+----------+------------------+---------------+ 666 | dropless | When dropless is set to NO, packets can be | YES | YES/NO | NO | 667 | | dropped if not enough free slots are | | | | 668 | | currently available in the queue, so the | | | | 669 | | write operation to the queue is non- | | | | 670 | | blocking. | | | | 671 | | When dropless is set to YES, packets cannot | | | | 672 | | be dropped if not enough free slots are | | | | 673 | | currently available in the queue, so the | | | | 674 | | write operation to the queue is blocking, as | | | | 675 | | the write operation is retried until enough | | | | 676 | | free slots become available and all the | | | | 677 | | packets are successfully written to the | | | | 678 | | queue. | | | | 679 +---------------+----------------------------------------------+----------+------------------+---------------+ 680 | n_retries | Number of retries. Valid only when dropless | YES | uint32_t | 0 | 681 | | is set to YES. When set to 0, it indicates | | | | 682 | | unlimited number of retries. | | | | 683 +---------------+----------------------------------------------+----------+------------------+---------------+ 684 | cpu | CPU socket ID where to allocate memory | YES | uint32_t | 0 | 685 | | for this SWQ. | | | | 686 +---------------+----------------------------------------------+----------+------------------+---------------+ 687 688 689TM section 690~~~~~~~~~~ 691 692.. _table_ip_pipelines_tm_section: 693 694.. tabularcolumns:: |p{2.5cm}|p{7cm}|p{1.5cm}|p{1.5cm}|p{1.5cm}| 695 696.. table:: Configuration file TM section 697 698 +---------------+---------------------------------------------+----------+----------+---------------+ 699 | Section | Description | Optional | Type | Default value | 700 +===============+=============================================+==========+==========+===============+ 701 | Cfg | File name to parse for the TM configuration | YES | string | tm_profile | 702 | | to be applied. The syntax of this file is | | | | 703 | | described in the examples/qos_sched DPDK | | | | 704 | | application documentation. | | | | 705 +---------------+---------------------------------------------+----------+----------+---------------+ 706 | burst_read | Read burst size (number of packets) | YES | uint32_t | 64 | 707 +---------------+---------------------------------------------+----------+----------+---------------+ 708 | burst_write | Write burst size (number of packets) | YES | uint32_t | 32 | 709 +---------------+---------------------------------------------+----------+----------+---------------+ 710 711 712KNI section 713~~~~~~~~~~~ 714 715.. _table_ip_pipelines_kni_section: 716 717.. tabularcolumns:: |p{2.5cm}|p{7cm}|p{1.5cm}|p{1.5cm}|p{1.5cm}| 718 719.. table:: Configuration file KNI section 720 721 +---------------+----------------------------------------------+----------+------------------+---------------+ 722 | Section | Description | Optional | Type | Default value | 723 +===============+==============================================+==========+==================+===============+ 724 | core | CPU core to run the KNI kernel thread. | YES | See "CPU Core | Not set | 725 | | When core config is set, the KNI kernel | | notation" | | 726 | | thread will be bound to the particular core. | | | | 727 | | When core config is not set, the KNI kernel | | | | 728 | | thread will be scheduled by the OS. | | | | 729 +---------------+----------------------------------------------+----------+------------------+---------------+ 730 | mempool | Mempool to use for buffer allocation for | YES | uint32_t | MEMPOOL0 | 731 | | current KNI port. The mempool ID has | | | | 732 | | to be associated with a valid instance | | | | 733 | | defined in the mempool entry of the global | | | | 734 | | section. | | | | 735 +---------------+----------------------------------------------+----------+------------------+---------------+ 736 | burst_read | Read burst size (number of packets) | YES | uint32_t | 32 | 737 | | | | power of 2 | | 738 | | | | 0 < burst < size | | 739 +---------------+----------------------------------------------+----------+------------------+---------------+ 740 | burst_write | Write burst size (number of packets) | YES | uint32_t | 32 | 741 | | | | power of 2 | | 742 | | | | 0 < burst < size | | 743 +---------------+----------------------------------------------+----------+------------------+---------------+ 744 | dropless | When dropless is set to NO, packets can be | YES | YES/NO | NO | 745 | | dropped if not enough free slots are | | | | 746 | | currently available in the queue, so the | | | | 747 | | write operation to the queue is non- | | | | 748 | | blocking. | | | | 749 | | When dropless is set to YES, packets cannot | | | | 750 | | be dropped if not enough free slots are | | | | 751 | | currently available in the queue, so the | | | | 752 | | write operation to the queue is blocking, as | | | | 753 | | the write operation is retried until enough | | | | 754 | | free slots become available and all the | | | | 755 | | packets are successfully written to the | | | | 756 | | queue. | | | | 757 +---------------+----------------------------------------------+----------+------------------+---------------+ 758 | n_retries | Number of retries. Valid only when dropless | YES | uint64_t | 0 | 759 | | is set to YES. When set to 0, it indicates | | | | 760 | | unlimited number of retries. | | | | 761 +---------------+----------------------------------------------+----------+------------------+---------------+ 762 763 764SOURCE section 765~~~~~~~~~~~~~~ 766 767.. _table_ip_pipelines_source_section: 768 769.. tabularcolumns:: |p{2.5cm}|p{7cm}|p{1.5cm}|p{1.5cm}|p{2cm}| 770 771.. table:: Configuration file SOURCE section 772 773 +---------------+---------------------------------------+----------+----------+---------------+ 774 | Section | Description | Optional | Type | Default value | 775 +===============+=======================================+==========+==========+===============+ 776 | Mempool | Mempool to use for buffer allocation. | YES | uint32_t | MEMPOOL0 | 777 +---------------+---------------------------------------+----------+----------+---------------+ 778 | Burst | Read burst size (number of packets) | | uint32_t | 32 | 779 +---------------+---------------------------------------+----------+----------+---------------+ 780 781 782SINK section 783~~~~~~~~~~~~ 784 785Currently, there are no parameters to be passed to a sink device, so 786SINK section is not allowed. 787 788MSGQ section 789~~~~~~~~~~~~ 790 791.. _table_ip_pipelines_msgq_section: 792 793.. tabularcolumns:: |p{2.5cm}|p{7cm}|p{1.5cm}|p{1.5cm}|p{1.5cm}| 794 795.. table:: Configuration file MSGQ section 796 797 +---------+--------------------------------------------+----------+------------+---------------+ 798 | Section | Description | Optional | Type | Default value | 799 +=========+============================================+==========+============+===============+ 800 | size | Queue size (number of packets) | YES | uint32_t | 64 | 801 | | | | != 0 | | 802 | | | | power of 2 | | 803 +---------+--------------------------------------------+----------+------------+---------------+ 804 | cpu | CPU socket ID where to allocate memory for | YES | uint32_t | 0 | 805 | | the current queue. | | | | 806 +---------+--------------------------------------------+----------+------------+---------------+ 807 808 809EAL section 810~~~~~~~~~~~ 811 812The application generates the EAL parameters rather than reading them from the command line. 813 814The CPU core mask parameter is generated based on the core entry of all PIPELINE sections. 815All the other EAL parameters can be set from this section of the application configuration file. 816 817 818Library of pipeline types 819------------------------- 820 821Pipeline module 822~~~~~~~~~~~~~~~ 823 824A pipeline is a self-contained module that implements a packet processing function and is typically implemented on 825top of the DPDK Packet Framework *librte_pipeline* library. 826The application provides a run-time mechanism to register different pipeline types. 827 828Depending on the required configuration, each registered pipeline type (pipeline class) is instantiated one or 829several times, with each pipeline instance (pipeline object) assigned to one of the available CPU cores. 830Each CPU core can run one or more pipeline instances, which might be of same or different types. 831For more information of the CPU core threading model, please refer to the :ref:`ip_pipeline_runtime` section. 832 833 834Pipeline type 835^^^^^^^^^^^^^ 836 837Each pipeline type is made up of a back-end and a front-end. The back-end represents the packet processing engine 838of the pipeline, typically implemented using the DPDK Packet Framework libraries, which reads packets from the 839input packet queues, handles them and eventually writes them to the output packet queues or drops them. 840The front-end represents the run-time configuration interface of the pipeline, which is exposed as CLI commands. 841The front-end communicates with the back-end through message queues. 842 843.. _table_ip_pipelines_back_end: 844 845.. tabularcolumns:: |p{1cm}|p{2cm}|p{12cm}| 846 847.. table:: Pipeline back-end 848 849 +------------+------------------+--------------------------------------------------------------------+ 850 | Field name | Field type | Description | 851 +============+==================+====================================================================+ 852 | f_init | Function pointer | Function to initialize the back-end of the current pipeline | 853 | | | instance. Typical work implemented by this function for the | 854 | | | current pipeline instance: | 855 | | | Memory allocation; | 856 | | | Parse the pipeline type specific arguments; | 857 | | | Initialize the pipeline input ports, output ports and tables, | 858 | | | interconnect input ports to tables; | 859 | | | Set the message handlers. | 860 +------------+------------------+--------------------------------------------------------------------+ 861 | f_free | Function pointer | Function to free the resources allocated by the back-end of the | 862 | | | current pipeline instance. | 863 +------------+------------------+--------------------------------------------------------------------+ 864 | f_run | Function pointer | Set to NULL for pipelines implemented using the DPDK library | 865 | | | librte_pipeline (typical case), and to non-NULL otherwise. This | 866 | | | mechanism is made available to support quick integration of | 867 | | | legacy code. | 868 | | | This function is expected to provide the packet processing | 869 | | | related code to be called as part of the CPU thread dispatch | 870 | | | loop, so this function is not allowed to contain an infinite loop. | 871 +------------+------------------+--------------------------------------------------------------------+ 872 | f_timer | Function pointer | Function to read the pipeline input message queues, handle | 873 | | | the request messages, create response messages and write | 874 | | | the response queues. The format of request and response | 875 | | | messages is defined by each pipeline type, with the exception | 876 | | | of some requests which are mandatory for all pipelines (e.g. | 877 | | | ping, statistics). | 878 +------------+------------------+--------------------------------------------------------------------+ 879 | f_track | Function pointer | See section Tracking pipeline output port to physical link | 880 +------------+------------------+--------------------------------------------------------------------+ 881 882 883.. _table_ip_pipelines_front_end: 884 885.. tabularcolumns:: |p{1cm}|p{2cm}|p{12cm}| 886 887.. table:: Pipeline front-end 888 889 +------------+-----------------------+-------------------------------------------------------------------+ 890 | Field name | Field type | Description | 891 +============+=======================+===================================================================+ 892 | f_init | Function pointer | Function to initialize the front-end of the current pipeline | 893 | | | instance. | 894 +------------+-----------------------+-------------------------------------------------------------------+ 895 | f_free | Function pointer | Function to free the resources allocated by the front-end of | 896 | | | the current pipeline instance. | 897 +------------+-----------------------+-------------------------------------------------------------------+ 898 | cmds | Array of CLI commands | Array of CLI commands to be registered to the application CLI | 899 | | | for the current pipeline type. Even though the CLI is executed | 900 | | | by a different pipeline (typically, this is the master pipeline), | 901 | | | from modularity perspective is more efficient to keep the | 902 | | | message client side (part of the front-end) together with the | 903 | | | message server side (part of the back-end). | 904 +------------+-----------------------+-------------------------------------------------------------------+ 905 906 907Tracking pipeline output port to physical link 908^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 909 910Each pipeline instance is a standalone block that does not have visibility into the other pipeline instances or 911the application-level pipeline inter-connectivity. 912In some cases, it is useful for a pipeline instance to get application level information related to pipeline 913connectivity, such as to identify the output link (e.g. physical NIC port) where one of its output ports connected, 914either directly or indirectly by traversing other pipeline instances. 915 916Tracking can be successful or unsuccessful. 917Typically, tracking for a specific pipeline instance is successful when each one of its input ports can be mapped 918to a single output port, meaning that all packets read from the current input port can only go out on a single 919output port. 920Depending on the pipeline type, some exceptions may be allowed: a small portion of the packets, considered exception 921packets, are sent out on an output port that is pre-configured for this purpose. 922 923For pass-through pipeline type, the tracking is always successful. 924For pipeline types as flow classification, firewall or routing, the tracking is only successful when the number of 925output ports for the current pipeline instance is 1. 926 927This feature is used by the IP routing pipeline for adding/removing implicit routes every time a link is brought 928up/down. 929 930 931Table copies 932^^^^^^^^^^^^ 933 934Fast table copy: pipeline table used by pipeline for the packet processing task, updated through messages, table 935data structures are optimized for lookup operation. 936 937Slow table copy: used by the configuration layer, typically updated through CLI commands, kept in sync with the fast 938copy (its update triggers the fast copy update). 939Required for executing advanced table queries without impacting the packet processing task, therefore the slow copy 940is typically organized using different criteria than the fast copy. 941 942Examples: 943 944* Flow classification: Search through current set of flows (e.g. list all flows with a specific source IP address); 945 946* Firewall: List rules in descending order of priority; 947 948* Routing table: List routes sorted by prefix depth and their type (local, remote, default); 949 950* ARP: List entries sorted per output interface. 951 952 953Packet meta-data 954^^^^^^^^^^^^^^^^ 955 956Packet meta-data field offsets provided as argument to pipeline instances are essentially defining the data structure 957for the packet meta-data used by the current application use-case. 958It is very useful to put it in the configuration file as a comment in order to facilitate the readability of the 959configuration file. 960 961The reason to use field offsets for defining the data structure for the packet meta-data is due to the C language 962limitation of not being able to define data structures at run-time. 963Feature to consider: have the configuration file parser automatically generate and print the data structure defining 964the packet meta-data for the current application use-case. 965 966Packet meta-data typically contains: 967 9681. Pure meta-data: intermediate data per packet that is computed internally, passed between different tables of 969 the same pipeline instance (e.g. lookup key for the ARP table is obtained from the routing table), or between 970 different pipeline instances (e.g. flow ID, traffic metering color, etc); 971 9722. Packet fields: typically, packet header fields that are read directly from the packet, or read from the packet 973 and saved (duplicated) as a working copy at a different location within the packet meta-data (e.g. Diffserv 974 5-tuple, IP destination address, etc). 975 976Several strategies are used to design the packet meta-data, as described in the next subsections. 977 978 979Store packet meta-data in a different cache line as the packet headers 980"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""" 981 982This approach is able to support protocols with variable header length, like MPLS, where the offset of IP header 983from the start of the packet (and, implicitly, the offset of the IP header in the packet buffer) is not fixed. 984Since the pipelines typically require the specification of a fixed offset to the packet fields (e.g. Diffserv 9855-tuple, used by the flow classification pipeline, or the IP destination address, used by the IP routing pipeline), 986the workaround is to have the packet RX pipeline copy these fields at fixed offsets within the packet meta-data. 987 988As this approach duplicates some of the packet fields, it requires accessing more cache lines per packet for filling 989in selected packet meta-data fields (on RX), as well as flushing selected packet meta-data fields into the 990packet (on TX). 991 992Example: 993 994.. code-block:: ini 995 996 997 ; struct app_pkt_metadata { 998 ; uint32_t ip_da; 999 ; uint32_t hash; 1000 ; uint32_t flow_id; 1001 ; uint32_t color; 1002 ; } __attribute__((__packed__)); 1003 ; 1004 1005 [PIPELINE1] 1006 ; Packet meta-data offsets 1007 ip_da_offset = 0; Used by: routing 1008 hash_offset = 4; Used by: RX, flow classification 1009 flow_id_offset = 8; Used by: flow classification, flow actions 1010 color_offset = 12; Used by: flow actions, routing 1011 1012 1013Overlay the packet meta-data in the same cache line with the packet headers 1014""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""" 1015 1016This approach is minimizing the number of cache line accessed per packet by storing the packet metadata in the 1017same cache line with the packet headers. 1018To enable this strategy, either some headroom is reserved for meta-data at the beginning of the packet headers 1019cache line (e.g. if 16 bytes are needed for meta-data, then the packet headroom can be set to 128+16 bytes, so 1020that NIC writes the first byte of the packet at offset 16 from the start of the first packet cache line), 1021or meta-data is reusing the space of some packet headers that are discarded from the packet (e.g. input Ethernet 1022header). 1023 1024Example: 1025 1026.. code-block:: ini 1027 1028 ; struct app_pkt_metadata { 1029 ; uint8_t headroom[RTE_PKTMBUF_HEADROOM]; /* 128 bytes (default) */ 1030 ; union { 1031 ; struct { 1032 ; struct ether_hdr ether; /* 14 bytes */ 1033 ; struct qinq_hdr qinq; /* 8 bytes */ 1034 ; }; 1035 ; struct { 1036 ; uint32_t hash; 1037 ; uint32_t flow_id; 1038 ; uint32_t color; 1039 ; }; 1040 ; }; 1041 ; struct ipv4_hdr ip; /* 20 bytes */ 1042 ; } __attribute__((__packed__)); 1043 ; 1044 [PIPELINE2] 1045 ; Packet meta-data offsets 1046 qinq_offset = 142; Used by: RX, flow classification 1047 ip_da_offset = 166; Used by: routing 1048 hash_offset = 128; Used by: RX, flow classification 1049 flow_id_offset = 132; Used by: flow classification, flow actions 1050 color_offset = 136; Used by: flow actions, routing 1051 1052 1053List of pipeline types 1054~~~~~~~~~~~~~~~~~~~~~~ 1055 1056.. _table_ip_pipelines_types: 1057 1058.. tabularcolumns:: |p{3cm}|p{5cm}|p{4cm}|p{4cm}| 1059 1060.. table:: List of pipeline types provided with the application 1061 1062 +-----------------------+-----------------------------+-----------------------+------------------------------------------+ 1063 | Name | Table(s) | Actions | Messages | 1064 +=======================+=============================+=======================+==========================================+ 1065 | Pass-through | Passthrough | 1. Pkt metadata build | 1. Ping | 1066 | | | 2. Flow hash | 2. Stats | 1067 | Note: depending on | | 3. Pkt checks | | 1068 | port type, can be | | 4. Load balancing | | 1069 | used for RX, TX, IP | | | | 1070 | fragmentation, IP | | | | 1071 | reassembly or Traffic | | | | 1072 | Management | | | | 1073 +-----------------------+-----------------------------+-----------------------+------------------------------------------+ 1074 | Flow classification | Exact match | 1. Flow ID | 1. Ping | 1075 | | | | | 1076 | | * Key = byte array | 2. Flow stats | 2. Stats | 1077 | | (source: pkt metadata) | 3. Metering | 3. Flow stats | 1078 | | * Data = action dependent | 4. Network Address | 4. Action stats | 1079 | | | 5. Translation (NAT) | 5. Flow add/ update/ delete | 1080 | | | | 6. Default flow add/ update/ delete | 1081 | | | | 7. Action update | 1082 +-----------------------+-----------------------------+-----------------------+------------------------------------------+ 1083 | Flow actions | Array | 1. Flow stats | 1. Ping | 1084 | | | | | 1085 | | * Key = Flow ID | 2. Metering | 2. Stats | 1086 | | (source: pkt metadata) | 3. Network Address | 3. Action stats | 1087 | | * Data = action dependent | 4. Translation (NAT) | 4. Action update | 1088 +-----------------------+-----------------------------+-----------------------+------------------------------------------+ 1089 | Firewall | ACL | 1. Allow/Drop | 1. Ping | 1090 | | | | | 1091 | | * Key = n-tuple | | 2. Stats | 1092 | | (source: pkt headers) | | 3. Rule add/ update/ delete | 1093 | | * Data = none | | 4. Default rule add/ update/ delete | 1094 +-----------------------+-----------------------------+-----------------------+------------------------------------------+ 1095 | IP routing | LPM (IPv4 or IPv6, | 1. TTL decrement and | 1. Ping | 1096 | | depending on pipeline type) | 2. IPv4 checksum | 2. Stats | 1097 | | | | | 1098 | | * Key = IP destination | 3. update | 3. Route add/ update/ delete | 1099 | | (source: pkt metadata) | 4. Header | 4. Default route add/ update/ delete | 1100 | | * Data = Dependent on | 5. encapsulation | 5. ARP entry add/ update/ delete | 1101 | | actions and next hop | 6. (based on next hop | 6. Default ARP entry add/ update/ delete | 1102 | | type | 7. type) | | 1103 | | | | | 1104 | | Hash table (for ARP, only | | | 1105 | | | | | 1106 | | when ARP is enabled) | | | 1107 | | | | | 1108 | | * Key = (Port ID, | | | 1109 | | next hop IP address) | | | 1110 | | (source: pkt meta-data) | | | 1111 | | * Data: MAC address | | | 1112 +-----------------------+-----------------------------+-----------------------+------------------------------------------+ 1113 1114 1115 1116Command Line Interface (CLI) 1117---------------------------- 1118 1119Global CLI commands 1120~~~~~~~~~~~~~~~~~~~ 1121 1122.. _table_ip_pipelines_cli_commands: 1123 1124.. tabularcolumns:: |p{3cm}|p{6cm}|p{6cm}| 1125 1126.. table:: Global CLI commands 1127 1128 +---------+---------------------------------------+--------------------------------------------+ 1129 | Command | Description | Syntax | 1130 +=========+=======================================+============================================+ 1131 | run | Run CLI commands script file. | run <file> | 1132 | | | <file> = path to file with CLI commands to | 1133 | | | execute | 1134 +---------+---------------------------------------+--------------------------------------------+ 1135 | quit | Gracefully terminate the application. | quit | 1136 +---------+---------------------------------------+--------------------------------------------+ 1137 1138 1139CLI commands for link configuration 1140~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1141 1142.. _table_ip_pipelines_runtime_config: 1143 1144.. tabularcolumns:: |p{3cm}|p{6cm}|p{6cm}| 1145 1146.. table:: List of run-time configuration commands for link configuration 1147 1148 +-------------+--------------------+--------------------------------------------+ 1149 | Command | Description | Syntax | 1150 +=============+====================+============================================+ 1151 | link config | Link configuration | link <link ID> config <IP address> <depth> | 1152 +-------------+--------------------+--------------------------------------------+ 1153 | link up | Link up | link <link ID> up | 1154 +-------------+--------------------+--------------------------------------------+ 1155 | link down | Link down | link <link ID> down | 1156 +-------------+--------------------+--------------------------------------------+ 1157 | link ls | Link list | link ls | 1158 +-------------+--------------------+--------------------------------------------+ 1159 1160 1161CLI commands common for all pipeline types 1162~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1163 1164.. _table_ip_pipelines_mandatory: 1165 1166.. tabularcolumns:: |p{3cm}|p{6cm}|p{6cm}| 1167 1168.. table:: CLI commands mandatory for all pipelines 1169 1170 +--------------------+------------------------------------------------------+----------------------------------------------+ 1171 | Command | Description | Syntax | 1172 +====================+======================================================+==============================================+ 1173 | ping | Check whether specific pipeline instance is alive. | p <pipeline ID> ping | 1174 | | The master pipeline sends a ping request | | 1175 | | message to given pipeline instance and waits for | | 1176 | | a response message back. | | 1177 | | Timeout message is displayed when the response | | 1178 | | message is not received before the timer | | 1179 | | expires. | | 1180 +--------------------+------------------------------------------------------+----------------------------------------------+ 1181 | stats | Display statistics for specific pipeline input port, | p <pipeline ID> stats port in <port in ID> | 1182 | | output port or table. | p <pipeline ID> stats port out <port out ID> | 1183 | | | p <pipeline ID> stats table <table ID> | 1184 +--------------------+------------------------------------------------------+----------------------------------------------+ 1185 | input port enable | Enable given input port for specific pipeline | p <pipeline ID> port in <port ID> enable | 1186 | | instance. | | 1187 +--------------------+------------------------------------------------------+----------------------------------------------+ 1188 | input port disable | Disable given input port for specific pipeline | p <pipeline ID> port in <port ID> disable | 1189 | | instance. | | 1190 +--------------------+------------------------------------------------------+----------------------------------------------+ 1191 1192Pipeline type specific CLI commands 1193~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1194 1195The pipeline specific CLI commands are part of the pipeline type front-end. 1196