1.. SPDX-License-Identifier: BSD-3-Clause 2 Copyright 2018 The DPDK contributors 3 4.. _doc_guidelines: 5 6DPDK Documentation Guidelines 7============================= 8 9This document outlines the guidelines for writing the DPDK Guides and API documentation in RST and Doxygen format. 10 11It also explains the structure of the DPDK documentation and how to build it. 12 13 14Structure of the Documentation 15------------------------------ 16 17The DPDK source code repository contains input files to build the API documentation and User Guides. 18 19The main directories that contain files related to documentation are shown below:: 20 21 lib 22 |-- acl 23 |-- cfgfile 24 |-- cmdline 25 |-- eal 26 | |-- ... 27 ... 28 doc 29 |-- api 30 +-- guides 31 |-- freebsd_gsg 32 |-- linux_gsg 33 |-- prog_guide 34 |-- sample_app_ug 35 |-- guidelines 36 |-- testpmd_app_ug 37 |-- rel_notes 38 |-- nics 39 |-- ... 40 41 42The API documentation is built from `Doxygen <http://www.doxygen.nl>`_ comments in the header files. 43These files are mainly in the ``lib/*`` directories although some of the Poll Mode Drivers in ``drivers/net`` 44are also documented with Doxygen. 45 46The configuration files that are used to control the Doxygen output are in the ``doc/api`` directory. 47 48The user guides such as *The Programmers Guide* and the *FreeBSD* and *Linux Getting Started* Guides are generated 49from RST markup text files using the `Sphinx <http://sphinx-doc.org>`_ Documentation Generator. 50 51These files are included in the ``doc/guides/`` directory. 52The output is controlled by the ``doc/guides/conf.py`` file. 53 54 55Role of the Documentation 56------------------------- 57 58The following items outline the roles of the different parts of the documentation and when they need to be updated or 59added to by the developer. 60 61* **Release Notes** 62 63 The Release Notes document which features have been added in the current and previous releases of DPDK and highlight 64 any known issues. 65 The Releases Notes also contain notifications of features that will change ABI compatibility in the next release. 66 67 Developers should include updates to the Release Notes with patch sets that relate to any of the following sections: 68 69 * New Features 70 * Resolved Issues (see below) 71 * Known Issues 72 * API Changes 73 * ABI Changes 74 * Shared Library Versions 75 76 Resolved Issues should only include issues from previous releases that have been resolved in the current release. 77 Issues that are introduced and then fixed within a release cycle do not have to be included here. 78 79 Refer to the Release Notes from the previous DPDK release for the correct format of each section. 80 81 82* **API documentation** 83 84 The API documentation explains how to use the public DPDK functions. 85 The `API index page <https://doc.dpdk.org/api/>`_ shows the generated API documentation with related groups of functions. 86 87 The API documentation should be updated via Doxygen comments when new functions are added. 88 89* **Getting Started Guides** 90 91 The Getting Started Guides show how to install and configure DPDK and how to run DPDK based applications on different OSes. 92 93 A Getting Started Guide should be added when DPDK is ported to a new OS. 94 95* **The Programmers Guide** 96 97 The Programmers Guide explains how the API components of DPDK such as the EAL, Memzone, Rings and the Hash Library work. 98 It also describes some of the higher level functionality such as Packet Distributor and Packet Framework. 99 It also shows the build system and explains how to add applications. 100 101 The Programmers Guide should be expanded when new functionality is added to DPDK. 102 103* **App Guides** 104 105 The app guides document the DPDK applications in the ``app`` directory such as ``testpmd``. 106 107 The app guides should be updated if functionality is changed or added. 108 109* **Sample App Guides** 110 111 The sample app guides document the DPDK example applications in the examples directory. 112 Generally they demonstrate a major feature such as L2 or L3 Forwarding, Multi Process or Power Management. 113 They explain the purpose of the sample application, how to run it and step through some of the code to explain the 114 major functionality. 115 116 A new sample application should be accompanied by a new sample app guide. 117 The guide for the Skeleton Forwarding app is a good starting reference. 118 119* **Network Interface Controller Drivers** 120 121 The NIC Drivers document explains the features of the individual Poll Mode Drivers, such as software requirements, 122 configuration and initialization. 123 124 New documentation should be added for new Poll Mode Drivers. 125 126* **Guidelines** 127 128 The guideline documents record community process, expectations and design directions. 129 130 They can be extended, amended or discussed by submitting a patch and getting community approval. 131 132 133Building the Documentation 134-------------------------- 135 136Dependencies 137~~~~~~~~~~~~ 138 139The following dependencies must be installed to build the documentation: 140 141* Doxygen. 142* Sphinx (also called python-sphinx). 143 144`Doxygen`_ generates documentation from commented source code. 145It can be installed as follows: 146 147.. code-block:: console 148 149 # Ubuntu/Debian. 150 sudo apt-get -y install doxygen 151 152 # Red Hat/Fedora. 153 sudo dnf -y install doxygen 154 155`Sphinx`_ is a Python documentation tool for converting RST files to HTML. 156For full support with figure and table captioning the latest version of Sphinx can be installed as follows: 157 158.. code-block:: console 159 160 # Ubuntu/Debian. 161 sudo apt-get -y install python3-sphinx python3-sphinx-rtd-theme 162 163 # Red Hat/Fedora. 164 sudo dnf -y install python3-sphinx python3-sphinx_rtd_theme 165 166For further information on getting started with Sphinx see the 167`Sphinx Getting Started <http://www.sphinx-doc.org/en/master/usage/quickstart.html>`_. 168 169.. Note:: 170 171 To get full support for Figure and Table numbering it is best to install Sphinx 1.3.1 or later. 172 173 174Build commands 175~~~~~~~~~~~~~~ 176 177The documentation is built using the standard DPDK build system. 178 179To build the documentation:: 180 181 ninja -C build doc 182 183See :doc:`../linux_gsg/build_dpdk` for more detail on compiling DPDK with meson. 184 185The output is generated in the directory ``build/doc/``, with: 186 187* HTML versions of the guide docs, e.g. Getting Started Guides, Programmers Guide, in ``build/doc/guides/html`` 188* HTML version of the API documentation in ``build/doc/api/html`` 189* Man-page version of the API documentation in ``build/doc/api/man``. 190 If not installing DPDK system-wise, these pages can be accessed by adding this directory to the ``MANPATH`` environment variable. 191 For example: 192 193.. code-block:: console 194 195 export MANPATH=:/path/to/build/doc/api/man 196 197.. Note:: 198 199 Make sure to fix any Sphinx or Doxygen warnings when adding or updating documentation. 200 201 202Document Guidelines 203------------------- 204 205Here are some guidelines in relation to the style of the documentation: 206 207* Document the obvious as well as the obscure since it won't always be obvious to the reader. 208 For example an instruction like "Set up 64 2MB Hugepages" is better when followed by a sample commandline or a link to 209 the appropriate section of the documentation. 210 211* Use American English spellings throughout. 212 This can be checked using the ``aspell`` utility:: 213 214 aspell --lang=en_US --check doc/guides/sample_app_ug/mydoc.rst 215 216 217RST Guidelines 218-------------- 219 220The RST (reStructuredText) format is a plain text markup format 221that can be converted to HTML or other formats. 222It is most closely associated with Python but it can be used to document any language. 223It is used in DPDK to document everything apart from the API. 224 225The Sphinx documentation contains a very useful `RST Primer <http://sphinx-doc.org/rest.html#rst-primer>`_ which is a 226good place to learn the minimal set of syntax required to format a document. 227 228The official `reStructuredText <http://docutils.sourceforge.net/rst.html>`_ website contains the specification for the 229RST format and also examples of how to use it. 230However, for most developers the RST Primer is a better resource. 231 232The most common guidelines for writing RST text are detailed in the 233`Documenting Python <https://docs.python.org/devguide/documenting.html>`_ guidelines. 234The additional guidelines below reiterate or expand upon those guidelines. 235 236 237Line Length 238~~~~~~~~~~~ 239 240* Lines in sentences should be less than 80 characters and wrapped at 241 words. Multiple sentences which are not separated by a blank line are joined 242 automatically into paragraphs. 243 244* Lines in literal blocks should be less than 80 characters 245 since they are not wrapped by the document formatters. 246 247 Long literal command lines can be shown wrapped with backslashes. For 248 example:: 249 250 dpdk-testpmd -l 2-3 -n 4 \ 251 --vdev=virtio_user0,path=/dev/vhost-net,queues=2,queue_size=1024 \ 252 -- -i --tx-offloads=0x0000002c --enable-lro --txq=2 --rxq=2 \ 253 --txd=1024 --rxd=1024 254 255 256Whitespace 257~~~~~~~~~~ 258 259* Standard RST indentation is 3 spaces. 260 Code can be indented 4 spaces, especially if it is copied from source files. 261 262* No tabs. 263 Convert tabs in embedded code to 4 or 8 spaces. 264 265* No trailing whitespace. 266 267* Add 2 blank lines before each section header. 268 269* Add 1 blank line after each section header. 270 271* Add 1 blank line between each line of a list. 272 273 274Section Headers 275~~~~~~~~~~~~~~~ 276 277* Section headers should use the following underline formats:: 278 279 Level 1 Heading 280 =============== 281 282 283 Level 2 Heading 284 --------------- 285 286 287 Level 3 Heading 288 ~~~~~~~~~~~~~~~ 289 290 291 Level 4 Heading 292 ^^^^^^^^^^^^^^^ 293 294 295* Level 4 headings should be used sparingly. 296 297* The underlines should match the length of the text. 298 299* In general, the heading should be less than 80 characters, for conciseness. 300 301* As noted above: 302 303 * Add 2 blank lines before each section header. 304 305 * Add 1 blank line after each section header. 306 307 308Lists 309~~~~~ 310 311* Bullet lists should be formatted with a leading ``*`` as follows:: 312 313 * Item one. 314 315 * Item two is a long line that is wrapped and then indented to match 316 the start of the previous line. 317 318 * One space character between the bullet and the text is preferred. 319 320* Numbered lists can be formatted with a leading number but the preference is to use ``#.`` which will give automatic numbering. 321 This is more convenient when adding or removing items:: 322 323 #. Item one. 324 325 #. Item two is a long line that is wrapped and then indented to match 326 the start of the previous line. 327 328 #. Item three. 329 330* Definition lists can be written with or without a bullet:: 331 332 * Item one. 333 334 Some text about item one. 335 336 * Item two. 337 338 Some text about item two. 339 340* All lists, and sub-lists, must be separated from the preceding text by a blank line. 341 This is a syntax requirement. 342 343* All list items should be separated by a blank line for readability. 344 345 346Code and Literal block sections 347~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 348 349* Inline text that is required to be rendered with a fixed width font should be enclosed in backquotes like this: 350 \`\`text\`\`, so that it appears like this: ``text``. 351 352* Fixed width, literal blocks of texts should be indented at least 3 spaces and prefixed with ``::`` like this:: 353 354 Here is some fixed width text:: 355 356 0x0001 0x0001 0x00FF 0x00FF 357 358* It is also possible to specify an encoding for a literal block using the ``.. code-block::`` directive so that syntax 359 highlighting can be applied. 360 Examples of supported highlighting are:: 361 362 .. code-block:: console 363 .. code-block:: c 364 .. code-block:: python 365 .. code-block:: diff 366 .. code-block:: none 367 368 That can be applied as follows:: 369 370 .. code-block:: c 371 372 #include<stdio.h> 373 374 int main() { 375 376 printf("Hello World\n"); 377 378 return 0; 379 } 380 381 Which would be rendered as: 382 383 .. code-block:: c 384 385 #include<stdio.h> 386 387 int main() { 388 389 printf("Hello World\n"); 390 391 return 0; 392 } 393 394* Code snippets can also be included directly from the code using the ``literalinclude`` block. 395 Using this block instead of a code block will ensure that the code snippets 396 shown in the documentation are always up to date with the code. 397 398 The following will include a snippet from the skeleton sample app:: 399 400 .. literalinclude:: ../../../examples/skeleton/basicfwd.c 401 :language: c 402 :start-after: Display the port MAC address. 403 :end-before: Enable RX in promiscuous mode for the Ethernet device. 404 :dedent: 1 405 406 This would be rendered as: 407 408 .. literalinclude:: ../../../examples/skeleton/basicfwd.c 409 :language: c 410 :start-after: Display the port MAC address. 411 :end-before: Enable RX in promiscuous mode for the Ethernet device. 412 :dedent: 1 413 414 Specifying ``:language:`` will enable syntax highlighting for the specified language. 415 ``:dedent:`` is used in this example to remove 1 leading tab from each line of the snippet. 416 417* ``start-after`` and ``end-before`` can use any text within a given file, 418 however it may be difficult to find unique text within your code to mark the 419 start and end of your snippets. In these cases, it is recommended to include 420 explicit tags in your code to denote these locations for documentation purposes. 421 The accepted format for these comments is: 422 423 * Before the code snippet, create a new comment which is a sentence explaining 424 what the code snippet contains. The comment is terminated with a scissors ``8<``. 425 * After the code snippet, create another new comment which starts with a 426 scissors ``>8``, then ``End of`` and the first comment repeated. 427 * The scissors should be orientated as shown to make it clear what code is being snipped. 428 429 This can be done as follows: 430 431 .. code-block:: c 432 433 /* Example feature being documented. 8< */ 434 foo(bar); 435 /* >8 End of example feature being documented. */ 436 437 ``foo(bar);`` could then be included in the docs using:: 438 439 .. literalinclude:: ../../../examples/sample_app/main.c 440 :language: c 441 :start-after: Example feature being documented. 8< 442 :end-before: >8 End of example feature being documented. 443 444 If a multiline comment is needed before the snippet, 445 then the last line of the multiline comment should be in the same format as 446 the first comment shown in the example. 447 448* More information about the ``literalinclude`` block can be found within the 449 `Sphinx Documentation <https://www.sphinx-doc.org/en/master/usage/restructuredtext/directives.html?highlight=literalinclude#directive-literalinclude>`_. 450 451* The default encoding for a literal block using the simplified ``::`` 452 directive is ``none``. 453 454* Lines in literal blocks should be less than 80 characters. 455 For long literal lines, try to wrap the text at sensible locations. 456 For example a long command line could be documented like this and still work if copied directly from the docs:: 457 458 ./<build_dir>/app/dpdk-testpmd -l 0-2 -n3 --vdev=net_pcap0,iface=eth0 \ 459 --vdev=net_pcap1,iface=eth1 \ 460 -- -i --nb-cores=2 --nb-ports=2 \ 461 --total-num-mbufs=2048 462 463* Long lines that cannot be wrapped, such as application output, should be truncated to be less than 80 characters. 464 465 466Images 467~~~~~~ 468 469* All images should be in SVG scalar graphics format. 470 They should be true SVG XML files and should not include binary formats embedded in a SVG wrapper. 471 472* The DPDK documentation contains some legacy images in PNG format. 473 These will be converted to SVG in time. 474 475* `Inkscape <http://inkscape.org>`_ is the recommended graphics editor for creating the images. 476 Use some of the older images in ``doc/guides/prog_guide/img/`` as a template, for example ``mbuf1.svg`` 477 or ``ring-enqueue1.svg``. 478 479* The SVG images should include a copyright notice, as an XML comment. 480 481* Images in the documentation should be formatted as follows: 482 483 * The image should be preceded by a label in the format ``.. _figure_XXXX:`` with a leading underscore and 484 where ``XXXX`` is a unique descriptive name. 485 486 * Images should be included using the ``.. figure::`` directive and the file type should be set to ``*`` (not ``.svg``). 487 This allows the format of the image to be changed if required, without updating the documentation. 488 489 * Images must have a caption as part of the ``.. figure::`` directive. 490 491* Here is an example of the previous three guidelines:: 492 493 .. _figure_mempool: 494 495 .. figure:: img/mempool.* 496 497 A mempool in memory with its associated ring. 498 499.. _mock_label: 500 501* Images can then be linked to using the ``:numref:`` directive:: 502 503 The mempool layout is shown in :numref:`figure_mempool`. 504 505 This would be rendered as: *The mempool layout is shown in* :ref:`Fig 6.3 <mock_label>`. 506 507 **Note**: The ``:numref:`` directive requires Sphinx 1.3.1 or later. 508 With earlier versions it will still be rendered as a link but won't have an automatically generated number. 509 510* The caption of the image can be generated, with a link, using the ``:ref:`` directive:: 511 512 :ref:`figure_mempool` 513 514 This would be rendered as: *A mempool in memory with its associated ring.* 515 516Tables 517~~~~~~ 518 519* RST tables should be used sparingly. 520 They are hard to format and to edit, and the same information 521 can usually be shown just as clearly with a definition or bullet list. 522 523* Tables in the documentation should be formatted as follows: 524 525 * The table should be preceded by a label in the format ``.. _table_XXXX:`` with a leading underscore and where 526 ``XXXX`` is a unique descriptive name. 527 528 * Tables should be included using the ``.. table::`` directive and must have a caption. 529 530* Here is an example of the previous two guidelines:: 531 532 .. _table_qos_pipes: 533 534 .. table:: Sample configuration for QOS pipes. 535 536 +----------+----------+----------+ 537 | Header 1 | Header 2 | Header 3 | 538 | | | | 539 +==========+==========+==========+ 540 | Text | Text | Text | 541 +----------+----------+----------+ 542 | ... | ... | ... | 543 +----------+----------+----------+ 544 545* Tables can be linked to using the ``:numref:`` and ``:ref:`` directives, as shown in the previous section for images. 546 For example:: 547 548 The QOS configuration is shown in :numref:`table_qos_pipes`. 549 550 551.. _links: 552 553Hyperlinks 554~~~~~~~~~~ 555 556* Links to external websites can be plain URLs. 557 The following is rendered as https://dpdk.org:: 558 559 https://dpdk.org 560 561* They can contain alternative text. 562 The following is rendered as `Check out DPDK <https://dpdk.org>`_:: 563 564 `Check out DPDK <https://dpdk.org>`_ 565 566* An internal link can be generated by placing labels in the document with the format ``.. _label_name``. 567 568* The following links to the top of this section: :ref:`links`:: 569 570 .. _links: 571 572 Hyperlinks 573 ~~~~~~~~~~ 574 575 * The following links to the top of this section: :ref:`links`: 576 577.. Note:: 578 579 The label must have a leading underscore but the reference to it must omit it. 580 This is a frequent cause of errors and warnings. 581 582* The use of a label is preferred since it works across files and will still work if the header text changes. 583 584 585.. _doxygen_guidelines: 586 587Doxygen Guidelines 588------------------ 589 590The DPDK API is documented using Doxygen comment annotations in the header files. 591Doxygen is a very powerful tool, it is extremely configurable and with a little effort can be used to create expressive documents. 592See the `Doxygen website <http://www.doxygen.nl>`_ for full details on how to use it. 593 594The following are some guidelines for use of Doxygen in the DPDK API documentation: 595 596* New libraries that are documented with Doxygen should be added to the Doxygen configuration file: ``doc/api/doxy-api.conf``. 597 It is only required to add the directory that contains the files. 598 It isn't necessary to explicitly name each file since the configuration matches all ``rte_*.h`` files in the directory. 599 600* Use proper capitalization and punctuation in the Doxygen comments since they will become sentences in the documentation. 601 This in particular applies to single line comments, which is the case the is most often forgotten. 602 603* Use ``@`` style Doxygen commands instead of ``\`` style commands. 604 605* Add a general description of each library at the head of the main header files: 606 607 .. code-block:: c 608 609 /** 610 * @file 611 * RTE Mempool. 612 * 613 * A memory pool is an allocator of fixed-size object. It is 614 * identified by its name, and uses a ring to store free objects. 615 * ... 616 */ 617 618* Document the purpose of a function, the parameters used and the return 619 value: 620 621 .. code-block:: c 622 623 /** 624 * Try to take the lock. 625 * 626 * @param sl 627 * A pointer to the spinlock. 628 * @return 629 * 1 if the lock is successfully taken; 0 otherwise. 630 */ 631 int rte_spinlock_trylock(rte_spinlock_t *sl); 632 633* Doxygen supports Markdown style syntax such as bold, italics, fixed width text and lists. 634 For example the second line in the ``devargs`` parameter in the previous example will be rendered as: 635 636 The strings should be a pci address like ``0000:01:00.0`` or **virtual** device name like ``net_pcap0``. 637 638* Use ``-`` instead of ``*`` for lists within the Doxygen comment since the latter can get confused with the comment delimiter. 639 640* Add an empty line between the function description, the ``@params`` and ``@return`` for readability. 641 642* Place the ``@params`` description on separate line and indent it by 2 spaces. 643 (It would be better to use no indentation since this is more common and also because checkpatch complains about leading 644 whitespace in comments. 645 However this is the convention used in the existing DPDK code.) 646 647* Documented functions can be linked to simply by adding ``()`` to the function name: 648 649 .. code-block:: c 650 651 /** 652 * The functions exported by the application Ethernet API to setup 653 * a device designated by its port identifier must be invoked in 654 * the following order: 655 * - rte_eth_dev_configure() 656 * - rte_eth_tx_queue_setup() 657 * - rte_eth_rx_queue_setup() 658 * - rte_eth_dev_start() 659 */ 660 661 In the API documentation the functions will be rendered as links, see the 662 `online section of the rte_ethdev.h docs <https://doc.dpdk.org/api/rte__ethdev_8h.html>`_ that contains the above text. 663 664* The ``@see`` keyword can be used to create a *see also* link to another file or library. 665 This directive should be placed on one line at the bottom of the documentation section. 666 667 .. code-block:: c 668 669 /** 670 * ... 671 * 672 * Some text that references mempools. 673 * 674 * @see eal_memzone.c 675 */ 676 677* Doxygen supports two types of comments for documenting variables, constants and members: prefix and postfix: 678 679 .. code-block:: c 680 681 /** This is a prefix comment. */ 682 #define RTE_FOO_ERROR 0x023. 683 684 #define RTE_BAR_ERROR 0x024. /**< This is a postfix comment. */ 685 686* Postfix comments are preferred for struct members and constants if they can be documented in the same way: 687 688 .. code-block:: c 689 690 struct rte_eth_stats { 691 uint64_t ipackets; /**< Total number of received packets. */ 692 uint64_t opackets; /**< Total number of transmitted packets.*/ 693 uint64_t ibytes; /**< Total number of received bytes. */ 694 uint64_t obytes; /**< Total number of transmitted bytes. */ 695 uint64_t imissed; /**< Total of RX missed packets. */ 696 uint64_t ibadcrc; /**< Total of RX packets with CRC error. */ 697 uint64_t ibadlen; /**< Total of RX packets with bad length. */ 698 } 699 700 Note: postfix comments should be aligned with spaces not tabs in accordance 701 with the :ref:`coding_style`. 702 703* If a single comment type can't be used, due to line length limitations then 704 prefix comments should be preferred. 705 For example this section of the code contains prefix comments, postfix comments on the same line and postfix 706 comments on a separate line: 707 708 .. code-block:: c 709 710 /** Number of elements in the elt_pa array. */ 711 uint32_t pg_num __rte_cache_aligned; 712 uint32_t pg_shift; /**< LOG2 of the physical pages. */ 713 uintptr_t pg_mask; /**< Physical page mask value. */ 714 uintptr_t elt_va_start; 715 /**< Virtual address of the first mempool object. */ 716 uintptr_t elt_va_end; 717 /**< Virtual address of the <size + 1> mempool object. */ 718 phys_addr_t elt_pa[1]; 719 /**< Array of physical page addresses for the mempool buffer. */ 720 721 This doesn't have an effect on the rendered documentation but it is confusing for the developer reading the code. 722 It this case it would be clearer to use prefix comments throughout: 723 724 .. code-block:: c 725 726 /** Number of elements in the elt_pa array. */ 727 uint32_t pg_num __rte_cache_aligned; 728 /** LOG2 of the physical pages. */ 729 uint32_t pg_shift; 730 /** Physical page mask value. */ 731 uintptr_t pg_mask; 732 /** Virtual address of the first mempool object. */ 733 uintptr_t elt_va_start; 734 /** Virtual address of the <size + 1> mempool object. */ 735 uintptr_t elt_va_end; 736 /** Array of physical page addresses for the mempool buffer. */ 737 phys_addr_t elt_pa[1]; 738 739* Read the rendered section of the documentation that you have added for correctness, clarity and consistency 740 with the surrounding text. 741