developer.texi 32 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880881882883884885886887
  1. \input texinfo @c -*- texinfo -*-
  2. @documentencoding UTF-8
  3. @settitle Developer Documentation
  4. @titlepage
  5. @center @titlefont{Developer Documentation}
  6. @end titlepage
  7. @top
  8. @contents
  9. @chapter Notes for external developers
  10. This document is mostly useful for internal FFmpeg developers.
  11. External developers who need to use the API in their application should
  12. refer to the API doxygen documentation in the public headers, and
  13. check the examples in @file{doc/examples} and in the source code to
  14. see how the public API is employed.
  15. You can use the FFmpeg libraries in your commercial program, but you
  16. are encouraged to @emph{publish any patch you make}. In this case the
  17. best way to proceed is to send your patches to the ffmpeg-devel
  18. mailing list following the guidelines illustrated in the remainder of
  19. this document.
  20. For more detailed legal information about the use of FFmpeg in
  21. external programs read the @file{LICENSE} file in the source tree and
  22. consult @url{https://ffmpeg.org/legal.html}.
  23. @chapter Contributing
  24. There are 2 ways by which code gets into FFmpeg:
  25. @itemize @bullet
  26. @item Submitting patches to the ffmpeg-devel mailing list.
  27. See @ref{Submitting patches} for details.
  28. @item Directly committing changes to the main tree.
  29. @end itemize
  30. Whichever way, changes should be reviewed by the maintainer of the code
  31. before they are committed. And they should follow the @ref{Coding Rules}.
  32. The developer making the commit and the author are responsible for their changes
  33. and should try to fix issues their commit causes.
  34. @anchor{Coding Rules}
  35. @chapter Coding Rules
  36. @section Code formatting conventions
  37. There are the following guidelines regarding the indentation in files:
  38. @itemize @bullet
  39. @item
  40. Indent size is 4.
  41. @item
  42. The TAB character is forbidden outside of Makefiles as is any
  43. form of trailing whitespace. Commits containing either will be
  44. rejected by the git repository.
  45. @item
  46. You should try to limit your code lines to 80 characters; however, do so if
  47. and only if this improves readability.
  48. @item
  49. K&R coding style is used.
  50. @end itemize
  51. The presentation is one inspired by 'indent -i4 -kr -nut'.
  52. The main priority in FFmpeg is simplicity and small code size in order to
  53. minimize the bug count.
  54. @section Comments
  55. Use the JavaDoc/Doxygen format (see examples below) so that code documentation
  56. can be generated automatically. All nontrivial functions should have a comment
  57. above them explaining what the function does, even if it is just one sentence.
  58. All structures and their member variables should be documented, too.
  59. Avoid Qt-style and similar Doxygen syntax with @code{!} in it, i.e. replace
  60. @code{//!} with @code{///} and similar. Also @@ syntax should be employed
  61. for markup commands, i.e. use @code{@@param} and not @code{\param}.
  62. @example
  63. /**
  64. * @@file
  65. * MPEG codec.
  66. * @@author ...
  67. */
  68. /**
  69. * Summary sentence.
  70. * more text ...
  71. * ...
  72. */
  73. typedef struct Foobar @{
  74. int var1; /**< var1 description */
  75. int var2; ///< var2 description
  76. /** var3 description */
  77. int var3;
  78. @} Foobar;
  79. /**
  80. * Summary sentence.
  81. * more text ...
  82. * ...
  83. * @@param my_parameter description of my_parameter
  84. * @@return return value description
  85. */
  86. int myfunc(int my_parameter)
  87. ...
  88. @end example
  89. @section C language features
  90. FFmpeg is programmed in the ISO C90 language with a few additional
  91. features from ISO C99, namely:
  92. @itemize @bullet
  93. @item
  94. the @samp{inline} keyword;
  95. @item
  96. @samp{//} comments;
  97. @item
  98. designated struct initializers (@samp{struct s x = @{ .i = 17 @};});
  99. @item
  100. compound literals (@samp{x = (struct s) @{ 17, 23 @};}).
  101. @item
  102. for loops with variable definition (@samp{for (int i = 0; i < 8; i++)});
  103. @item
  104. Variadic macros (@samp{#define ARRAY(nb, ...) (int[nb + 1])@{ nb, __VA_ARGS__ @}});
  105. @item
  106. Implementation defined behavior for signed integers is assumed to match the
  107. expected behavior for two's complement. Non representable values in integer
  108. casts are binary truncated. Shift right of signed values uses sign extension.
  109. @end itemize
  110. These features are supported by all compilers we care about, so we will not
  111. accept patches to remove their use unless they absolutely do not impair
  112. clarity and performance.
  113. All code must compile with recent versions of GCC and a number of other
  114. currently supported compilers. To ensure compatibility, please do not use
  115. additional C99 features or GCC extensions. Especially watch out for:
  116. @itemize @bullet
  117. @item
  118. mixing statements and declarations;
  119. @item
  120. @samp{long long} (use @samp{int64_t} instead);
  121. @item
  122. @samp{__attribute__} not protected by @samp{#ifdef __GNUC__} or similar;
  123. @item
  124. GCC statement expressions (@samp{(x = (@{ int y = 4; y; @})}).
  125. @end itemize
  126. @section Naming conventions
  127. All names should be composed with underscores (_), not CamelCase. For example,
  128. @samp{avfilter_get_video_buffer} is an acceptable function name and
  129. @samp{AVFilterGetVideo} is not. The exception from this are type names, like
  130. for example structs and enums; they should always be in CamelCase.
  131. There are the following conventions for naming variables and functions:
  132. @itemize @bullet
  133. @item
  134. For local variables no prefix is required.
  135. @item
  136. For file-scope variables and functions declared as @code{static}, no prefix
  137. is required.
  138. @item
  139. For variables and functions visible outside of file scope, but only used
  140. internally by a library, an @code{ff_} prefix should be used,
  141. e.g. @samp{ff_w64_demuxer}.
  142. @item
  143. For variables and functions visible outside of file scope, used internally
  144. across multiple libraries, use @code{avpriv_} as prefix, for example,
  145. @samp{avpriv_report_missing_feature}.
  146. @item
  147. Each library has its own prefix for public symbols, in addition to the
  148. commonly used @code{av_} (@code{avformat_} for libavformat,
  149. @code{avcodec_} for libavcodec, @code{swr_} for libswresample, etc).
  150. Check the existing code and choose names accordingly.
  151. Note that some symbols without these prefixes are also exported for
  152. retro-compatibility reasons. These exceptions are declared in the
  153. @code{lib<name>/lib<name>.v} files.
  154. @end itemize
  155. Furthermore, name space reserved for the system should not be invaded.
  156. Identifiers ending in @code{_t} are reserved by
  157. @url{http://pubs.opengroup.org/onlinepubs/007904975/functions/xsh_chap02_02.html#tag_02_02_02, POSIX}.
  158. Also avoid names starting with @code{__} or @code{_} followed by an uppercase
  159. letter as they are reserved by the C standard. Names starting with @code{_}
  160. are reserved at the file level and may not be used for externally visible
  161. symbols. If in doubt, just avoid names starting with @code{_} altogether.
  162. @section Miscellaneous conventions
  163. @itemize @bullet
  164. @item
  165. fprintf and printf are forbidden in libavformat and libavcodec,
  166. please use av_log() instead.
  167. @item
  168. Casts should be used only when necessary. Unneeded parentheses
  169. should also be avoided if they don't make the code easier to understand.
  170. @end itemize
  171. @section Editor configuration
  172. In order to configure Vim to follow FFmpeg formatting conventions, paste
  173. the following snippet into your @file{.vimrc}:
  174. @example
  175. " indentation rules for FFmpeg: 4 spaces, no tabs
  176. set expandtab
  177. set shiftwidth=4
  178. set softtabstop=4
  179. set cindent
  180. set cinoptions=(0
  181. " Allow tabs in Makefiles.
  182. autocmd FileType make,automake set noexpandtab shiftwidth=8 softtabstop=8
  183. " Trailing whitespace and tabs are forbidden, so highlight them.
  184. highlight ForbiddenWhitespace ctermbg=red guibg=red
  185. match ForbiddenWhitespace /\s\+$\|\t/
  186. " Do not highlight spaces at the end of line while typing on that line.
  187. autocmd InsertEnter * match ForbiddenWhitespace /\t\|\s\+\%#\@@<!$/
  188. @end example
  189. For Emacs, add these roughly equivalent lines to your @file{.emacs.d/init.el}:
  190. @lisp
  191. (c-add-style "ffmpeg"
  192. '("k&r"
  193. (c-basic-offset . 4)
  194. (indent-tabs-mode . nil)
  195. (show-trailing-whitespace . t)
  196. (c-offsets-alist
  197. (statement-cont . (c-lineup-assignments +)))
  198. )
  199. )
  200. (setq c-default-style "ffmpeg")
  201. @end lisp
  202. @chapter Development Policy
  203. @section Patches/Committing
  204. @subheading Licenses for patches must be compatible with FFmpeg.
  205. Contributions should be licensed under the
  206. @uref{http://www.gnu.org/licenses/lgpl-2.1.html, LGPL 2.1},
  207. including an "or any later version" clause, or, if you prefer
  208. a gift-style license, the
  209. @uref{http://opensource.org/licenses/isc-license.txt, ISC} or
  210. @uref{http://mit-license.org/, MIT} license.
  211. @uref{http://www.gnu.org/licenses/gpl-2.0.html, GPL 2} including
  212. an "or any later version" clause is also acceptable, but LGPL is
  213. preferred.
  214. If you add a new file, give it a proper license header. Do not copy and
  215. paste it from a random place, use an existing file as template.
  216. @subheading You must not commit code which breaks FFmpeg!
  217. This means unfinished code which is enabled and breaks compilation,
  218. or compiles but does not work/breaks the regression tests. Code which
  219. is unfinished but disabled may be permitted under-circumstances, like
  220. missing samples or an implementation with a small subset of features.
  221. Always check the mailing list for any reviewers with issues and test
  222. FATE before you push.
  223. @subheading Keep the main commit message short with an extended description below.
  224. The commit message should have a short first line in the form of
  225. a @samp{topic: short description} as a header, separated by a newline
  226. from the body consisting of an explanation of why the change is necessary.
  227. If the commit fixes a known bug on the bug tracker, the commit message
  228. should include its bug ID. Referring to the issue on the bug tracker does
  229. not exempt you from writing an excerpt of the bug in the commit message.
  230. @subheading Testing must be adequate but not excessive.
  231. If it works for you, others, and passes FATE then it should be OK to commit
  232. it, provided it fits the other committing criteria. You should not worry about
  233. over-testing things. If your code has problems (portability, triggers
  234. compiler bugs, unusual environment etc) they will be reported and eventually
  235. fixed.
  236. @subheading Do not commit unrelated changes together.
  237. They should be split them into self-contained pieces. Also do not forget
  238. that if part B depends on part A, but A does not depend on B, then A can
  239. and should be committed first and separate from B. Keeping changes well
  240. split into self-contained parts makes reviewing and understanding them on
  241. the commit log mailing list easier. This also helps in case of debugging
  242. later on.
  243. Also if you have doubts about splitting or not splitting, do not hesitate to
  244. ask/discuss it on the developer mailing list.
  245. @subheading Ask before you change the build system (configure, etc).
  246. Do not commit changes to the build system (Makefiles, configure script)
  247. which change behavior, defaults etc, without asking first. The same
  248. applies to compiler warning fixes, trivial looking fixes and to code
  249. maintained by other developers. We usually have a reason for doing things
  250. the way we do. Send your changes as patches to the ffmpeg-devel mailing
  251. list, and if the code maintainers say OK, you may commit. This does not
  252. apply to files you wrote and/or maintain.
  253. @subheading Cosmetic changes should be kept in separate patches.
  254. We refuse source indentation and other cosmetic changes if they are mixed
  255. with functional changes, such commits will be rejected and removed. Every
  256. developer has his own indentation style, you should not change it. Of course
  257. if you (re)write something, you can use your own style, even though we would
  258. prefer if the indentation throughout FFmpeg was consistent (Many projects
  259. force a given indentation style - we do not.). If you really need to make
  260. indentation changes (try to avoid this), separate them strictly from real
  261. changes.
  262. NOTE: If you had to put if()@{ .. @} over a large (> 5 lines) chunk of code,
  263. then either do NOT change the indentation of the inner part within (do not
  264. move it to the right)! or do so in a separate commit
  265. @subheading Commit messages should always be filled out properly.
  266. Always fill out the commit log message. Describe in a few lines what you
  267. changed and why. You can refer to mailing list postings if you fix a
  268. particular bug. Comments such as "fixed!" or "Changed it." are unacceptable.
  269. Recommended format:
  270. @example
  271. area changed: Short 1 line description
  272. details describing what and why and giving references.
  273. @end example
  274. @subheading Credit the author of the patch.
  275. Make sure the author of the commit is set correctly. (see git commit --author)
  276. If you apply a patch, send an
  277. answer to ffmpeg-devel (or wherever you got the patch from) saying that
  278. you applied the patch.
  279. @subheading Complex patches should refer to discussion surrounding them.
  280. When applying patches that have been discussed (at length) on the mailing
  281. list, reference the thread in the log message.
  282. @subheading Always wait long enough before pushing changes
  283. Do NOT commit to code actively maintained by others without permission.
  284. Send a patch to ffmpeg-devel. If no one answers within a reasonable
  285. time-frame (12h for build failures and security fixes, 3 days small changes,
  286. 1 week for big patches) then commit your patch if you think it is OK.
  287. Also note, the maintainer can simply ask for more time to review!
  288. @section Code
  289. @subheading API/ABI changes should be discussed before they are made.
  290. Do not change behavior of the programs (renaming options etc) or public
  291. API or ABI without first discussing it on the ffmpeg-devel mailing list.
  292. Do not remove widely used functionality or features (redundant code can be removed).
  293. @subheading Remember to check if you need to bump versions for libav*.
  294. Depending on the change, you may need to change the version integer.
  295. Incrementing the first component means no backward compatibility to
  296. previous versions (e.g. removal of a function from the public API).
  297. Incrementing the second component means backward compatible change
  298. (e.g. addition of a function to the public API or extension of an
  299. existing data structure).
  300. Incrementing the third component means a noteworthy binary compatible
  301. change (e.g. encoder bug fix that matters for the decoder). The third
  302. component always starts at 100 to distinguish FFmpeg from Libav.
  303. @subheading Warnings for correct code may be disabled if there is no other option.
  304. Compiler warnings indicate potential bugs or code with bad style. If a type of
  305. warning always points to correct and clean code, that warning should
  306. be disabled, not the code changed.
  307. Thus the remaining warnings can either be bugs or correct code.
  308. If it is a bug, the bug has to be fixed. If it is not, the code should
  309. be changed to not generate a warning unless that causes a slowdown
  310. or obfuscates the code.
  311. @subheading Check untrusted input properly.
  312. Never write to unallocated memory, never write over the end of arrays,
  313. always check values read from some untrusted source before using them
  314. as array index or other risky things.
  315. @section Documentation/Other
  316. @subheading Subscribe to the ffmpeg-devel mailing list.
  317. It is important to be subscribed to the
  318. @uref{https://lists.ffmpeg.org/mailman/listinfo/ffmpeg-devel, ffmpeg-devel}
  319. mailing list. Almost any non-trivial patch is to be sent there for review.
  320. Other developers may have comments about your contribution. We expect you see
  321. those comments, and to improve it if requested. (N.B. Experienced committers
  322. have other channels, and may sometimes skip review for trivial fixes.) Also,
  323. discussion here about bug fixes and FFmpeg improvements by other developers may
  324. be helpful information for you. Finally, by being a list subscriber, your
  325. contribution will be posted immediately to the list, without the moderation
  326. hold which messages from non-subscribers experience.
  327. However, it is more important to the project that we receive your patch than
  328. that you be subscribed to the ffmpeg-devel list. If you have a patch, and don't
  329. want to subscribe and discuss the patch, then please do send it to the list
  330. anyway.
  331. @subheading Subscribe to the ffmpeg-cvslog mailing list.
  332. Diffs of all commits are sent to the
  333. @uref{https://lists.ffmpeg.org/mailman/listinfo/ffmpeg-cvslog, ffmpeg-cvslog}
  334. mailing list. Some developers read this list to review all code base changes
  335. from all sources. Subscribing to this list is not mandatory.
  336. @subheading Keep the documentation up to date.
  337. Update the documentation if you change behavior or add features. If you are
  338. unsure how best to do this, send a patch to ffmpeg-devel, the documentation
  339. maintainer(s) will review and commit your stuff.
  340. @subheading Important discussions should be accessible to all.
  341. Try to keep important discussions and requests (also) on the public
  342. developer mailing list, so that all developers can benefit from them.
  343. @subheading Check your entries in MAINTAINERS.
  344. Make sure that no parts of the codebase that you maintain are missing from the
  345. @file{MAINTAINERS} file. If something that you want to maintain is missing add it with
  346. your name after it.
  347. If at some point you no longer want to maintain some code, then please help in
  348. finding a new maintainer and also don't forget to update the @file{MAINTAINERS} file.
  349. We think our rules are not too hard. If you have comments, contact us.
  350. @chapter Code of conduct
  351. Be friendly and respectful towards others and third parties.
  352. Treat others the way you yourself want to be treated.
  353. Be considerate. Not everyone shares the same viewpoint and priorities as you do.
  354. Different opinions and interpretations help the project.
  355. Looking at issues from a different perspective assists development.
  356. Do not assume malice for things that can be attributed to incompetence. Even if
  357. it is malice, it's rarely good to start with that as initial assumption.
  358. Stay friendly even if someone acts contrarily. Everyone has a bad day
  359. once in a while.
  360. If you yourself have a bad day or are angry then try to take a break and reply
  361. once you are calm and without anger if you have to.
  362. Try to help other team members and cooperate if you can.
  363. The goal of software development is to create technical excellence, not for any
  364. individual to be better and "win" against the others. Large software projects
  365. are only possible and successful through teamwork.
  366. If someone struggles do not put them down. Give them a helping hand
  367. instead and point them in the right direction.
  368. Finally, keep in mind the immortal words of Bill and Ted,
  369. "Be excellent to each other."
  370. @anchor{Submitting patches}
  371. @chapter Submitting patches
  372. First, read the @ref{Coding Rules} above if you did not yet, in particular
  373. the rules regarding patch submission.
  374. When you submit your patch, please use @code{git format-patch} or
  375. @code{git send-email}. We cannot read other diffs :-).
  376. Also please do not submit a patch which contains several unrelated changes.
  377. Split it into separate, self-contained pieces. This does not mean splitting
  378. file by file. Instead, make the patch as small as possible while still
  379. keeping it as a logical unit that contains an individual change, even
  380. if it spans multiple files. This makes reviewing your patches much easier
  381. for us and greatly increases your chances of getting your patch applied.
  382. Use the patcheck tool of FFmpeg to check your patch.
  383. The tool is located in the tools directory.
  384. Run the @ref{Regression tests} before submitting a patch in order to verify
  385. it does not cause unexpected problems.
  386. It also helps quite a bit if you tell us what the patch does (for example
  387. 'replaces lrint by lrintf'), and why (for example '*BSD isn't C99 compliant
  388. and has no lrint()')
  389. Also please if you send several patches, send each patch as a separate mail,
  390. do not attach several unrelated patches to the same mail.
  391. Patches should be posted to the
  392. @uref{https://lists.ffmpeg.org/mailman/listinfo/ffmpeg-devel, ffmpeg-devel}
  393. mailing list. Use @code{git send-email} when possible since it will properly
  394. send patches without requiring extra care. If you cannot, then send patches
  395. as base64-encoded attachments, so your patch is not trashed during
  396. transmission. Also ensure the correct mime type is used
  397. (text/x-diff or text/x-patch or at least text/plain) and that only one
  398. patch is inline or attached per mail.
  399. You can check @url{https://patchwork.ffmpeg.org}, if your patch does not show up, its mime type
  400. likely was wrong.
  401. @subheading How to setup git send-email?
  402. Please see @url{https://git-send-email.io/}.
  403. For gmail additionally see @url{https://shallowsky.com/blog/tech/email/gmail-app-passwds.html}.
  404. @subheading Sending patches from email clients
  405. Using @code{git send-email} might not be desirable for everyone. The
  406. following trick allows to send patches via email clients in a safe
  407. way. It has been tested with Outlook and Thunderbird (with X-Unsent
  408. extension) and might work with other applications.
  409. Create your patch like this:
  410. @verbatim
  411. git format-patch -s -o "outputfolder" --add-header "X-Unsent: 1" --suffix .eml --to ffmpeg-devel@ffmpeg.org -1 1a2b3c4d
  412. @end verbatim
  413. Now you'll just need to open the eml file with the email application
  414. and execute 'Send'.
  415. @subheading Reviews
  416. Your patch will be reviewed on the mailing list. You will likely be asked
  417. to make some changes and are expected to send in an improved version that
  418. incorporates the requests from the review. This process may go through
  419. several iterations. Once your patch is deemed good enough, some developer
  420. will pick it up and commit it to the official FFmpeg tree.
  421. Give us a few days to react. But if some time passes without reaction,
  422. send a reminder by email. Your patch should eventually be dealt with.
  423. @chapter New codecs or formats checklist
  424. @enumerate
  425. @item
  426. Did you use av_cold for codec initialization and close functions?
  427. @item
  428. Did you add a long_name under NULL_IF_CONFIG_SMALL to the AVCodec or
  429. AVInputFormat/AVOutputFormat struct?
  430. @item
  431. Did you bump the minor version number (and reset the micro version
  432. number) in @file{libavcodec/version.h} or @file{libavformat/version.h}?
  433. @item
  434. Did you register it in @file{allcodecs.c} or @file{allformats.c}?
  435. @item
  436. Did you add the AVCodecID to @file{avcodec.h}?
  437. When adding new codec IDs, also add an entry to the codec descriptor
  438. list in @file{libavcodec/codec_desc.c}.
  439. @item
  440. If it has a FourCC, did you add it to @file{libavformat/riff.c},
  441. even if it is only a decoder?
  442. @item
  443. Did you add a rule to compile the appropriate files in the Makefile?
  444. Remember to do this even if you're just adding a format to a file that is
  445. already being compiled by some other rule, like a raw demuxer.
  446. @item
  447. Did you add an entry to the table of supported formats or codecs in
  448. @file{doc/general.texi}?
  449. @item
  450. Did you add an entry in the Changelog?
  451. @item
  452. If it depends on a parser or a library, did you add that dependency in
  453. configure?
  454. @item
  455. Did you @code{git add} the appropriate files before committing?
  456. @item
  457. Did you make sure it compiles standalone, i.e. with
  458. @code{configure --disable-everything --enable-decoder=foo}
  459. (or @code{--enable-demuxer} or whatever your component is)?
  460. @end enumerate
  461. @chapter Patch submission checklist
  462. @enumerate
  463. @item
  464. Does @code{make fate} pass with the patch applied?
  465. @item
  466. Was the patch generated with git format-patch or send-email?
  467. @item
  468. Did you sign-off your patch? (@code{git commit -s})
  469. See @uref{https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/plain/Documentation/process/submitting-patches.rst, Sign your work} for the meaning
  470. of @dfn{sign-off}.
  471. @item
  472. Did you provide a clear git commit log message?
  473. @item
  474. Is the patch against latest FFmpeg git master branch?
  475. @item
  476. Are you subscribed to ffmpeg-devel?
  477. (the list is subscribers only due to spam)
  478. @item
  479. Have you checked that the changes are minimal, so that the same cannot be
  480. achieved with a smaller patch and/or simpler final code?
  481. @item
  482. If the change is to speed critical code, did you benchmark it?
  483. @item
  484. If you did any benchmarks, did you provide them in the mail?
  485. @item
  486. Have you checked that the patch does not introduce buffer overflows or
  487. other security issues?
  488. @item
  489. Did you test your decoder or demuxer against damaged data? If no, see
  490. tools/trasher, the noise bitstream filter, and
  491. @uref{http://caca.zoy.org/wiki/zzuf, zzuf}. Your decoder or demuxer
  492. should not crash, end in a (near) infinite loop, or allocate ridiculous
  493. amounts of memory when fed damaged data.
  494. @item
  495. Did you test your decoder or demuxer against sample files?
  496. Samples may be obtained at @url{https://samples.ffmpeg.org}.
  497. @item
  498. Does the patch not mix functional and cosmetic changes?
  499. @item
  500. Did you add tabs or trailing whitespace to the code? Both are forbidden.
  501. @item
  502. Is the patch attached to the email you send?
  503. @item
  504. Is the mime type of the patch correct? It should be text/x-diff or
  505. text/x-patch or at least text/plain and not application/octet-stream.
  506. @item
  507. If the patch fixes a bug, did you provide a verbose analysis of the bug?
  508. @item
  509. If the patch fixes a bug, did you provide enough information, including
  510. a sample, so the bug can be reproduced and the fix can be verified?
  511. Note please do not attach samples >100k to mails but rather provide a
  512. URL, you can upload to @url{https://streams.videolan.org/upload/}.
  513. @item
  514. Did you provide a verbose summary about what the patch does change?
  515. @item
  516. Did you provide a verbose explanation why it changes things like it does?
  517. @item
  518. Did you provide a verbose summary of the user visible advantages and
  519. disadvantages if the patch is applied?
  520. @item
  521. Did you provide an example so we can verify the new feature added by the
  522. patch easily?
  523. @item
  524. If you added a new file, did you insert a license header? It should be
  525. taken from FFmpeg, not randomly copied and pasted from somewhere else.
  526. @item
  527. You should maintain alphabetical order in alphabetically ordered lists as
  528. long as doing so does not break API/ABI compatibility.
  529. @item
  530. Lines with similar content should be aligned vertically when doing so
  531. improves readability.
  532. @item
  533. Consider adding a regression test for your code.
  534. @item
  535. If you added YASM code please check that things still work with --disable-yasm.
  536. @item
  537. Make sure you check the return values of function and return appropriate
  538. error codes. Especially memory allocation functions like @code{av_malloc()}
  539. are notoriously left unchecked, which is a serious problem.
  540. @item
  541. Test your code with valgrind and or Address Sanitizer to ensure it's free
  542. of leaks, out of array accesses, etc.
  543. @end enumerate
  544. @chapter Patch review process
  545. All patches posted to ffmpeg-devel will be reviewed, unless they contain a
  546. clear note that the patch is not for the git master branch.
  547. Reviews and comments will be posted as replies to the patch on the
  548. mailing list. The patch submitter then has to take care of every comment,
  549. that can be by resubmitting a changed patch or by discussion. Resubmitted
  550. patches will themselves be reviewed like any other patch. If at some point
  551. a patch passes review with no comments then it is approved, that can for
  552. simple and small patches happen immediately while large patches will generally
  553. have to be changed and reviewed many times before they are approved.
  554. After a patch is approved it will be committed to the repository.
  555. We will review all submitted patches, but sometimes we are quite busy so
  556. especially for large patches this can take several weeks.
  557. If you feel that the review process is too slow and you are willing to try to
  558. take over maintainership of the area of code you change then just clone
  559. git master and maintain the area of code there. We will merge each area from
  560. where its best maintained.
  561. When resubmitting patches, please do not make any significant changes
  562. not related to the comments received during review. Such patches will
  563. be rejected. Instead, submit significant changes or new features as
  564. separate patches.
  565. Everyone is welcome to review patches. Also if you are waiting for your patch
  566. to be reviewed, please consider helping to review other patches, that is a great
  567. way to get everyone's patches reviewed sooner.
  568. @anchor{Regression tests}
  569. @chapter Regression tests
  570. Before submitting a patch (or committing to the repository), you should at least
  571. test that you did not break anything.
  572. Running 'make fate' accomplishes this, please see @url{fate.html} for details.
  573. [Of course, some patches may change the results of the regression tests. In
  574. this case, the reference results of the regression tests shall be modified
  575. accordingly].
  576. @section Adding files to the fate-suite dataset
  577. When there is no muxer or encoder available to generate test media for a
  578. specific test then the media has to be included in the fate-suite.
  579. First please make sure that the sample file is as small as possible to test the
  580. respective decoder or demuxer sufficiently. Large files increase network
  581. bandwidth and disk space requirements.
  582. Once you have a working fate test and fate sample, provide in the commit
  583. message or introductory message for the patch series that you post to
  584. the ffmpeg-devel mailing list, a direct link to download the sample media.
  585. @section Visualizing Test Coverage
  586. The FFmpeg build system allows visualizing the test coverage in an easy
  587. manner with the coverage tools @code{gcov}/@code{lcov}. This involves
  588. the following steps:
  589. @enumerate
  590. @item
  591. Configure to compile with instrumentation enabled:
  592. @code{configure --toolchain=gcov}.
  593. @item
  594. Run your test case, either manually or via FATE. This can be either
  595. the full FATE regression suite, or any arbitrary invocation of any
  596. front-end tool provided by FFmpeg, in any combination.
  597. @item
  598. Run @code{make lcov} to generate coverage data in HTML format.
  599. @item
  600. View @code{lcov/index.html} in your preferred HTML viewer.
  601. @end enumerate
  602. You can use the command @code{make lcov-reset} to reset the coverage
  603. measurements. You will need to rerun @code{make lcov} after running a
  604. new test.
  605. @section Using Valgrind
  606. The configure script provides a shortcut for using valgrind to spot bugs
  607. related to memory handling. Just add the option
  608. @code{--toolchain=valgrind-memcheck} or @code{--toolchain=valgrind-massif}
  609. to your configure line, and reasonable defaults will be set for running
  610. FATE under the supervision of either the @strong{memcheck} or the
  611. @strong{massif} tool of the valgrind suite.
  612. In case you need finer control over how valgrind is invoked, use the
  613. @code{--target-exec='valgrind <your_custom_valgrind_options>} option in
  614. your configure line instead.
  615. @anchor{Release process}
  616. @chapter Release process
  617. FFmpeg maintains a set of @strong{release branches}, which are the
  618. recommended deliverable for system integrators and distributors (such as
  619. Linux distributions, etc.). At regular times, a @strong{release
  620. manager} prepares, tests and publishes tarballs on the
  621. @url{https://ffmpeg.org} website.
  622. There are two kinds of releases:
  623. @enumerate
  624. @item
  625. @strong{Major releases} always include the latest and greatest
  626. features and functionality.
  627. @item
  628. @strong{Point releases} are cut from @strong{release} branches,
  629. which are named @code{release/X}, with @code{X} being the release
  630. version number.
  631. @end enumerate
  632. Note that we promise to our users that shared libraries from any FFmpeg
  633. release never break programs that have been @strong{compiled} against
  634. previous versions of @strong{the same release series} in any case!
  635. However, from time to time, we do make API changes that require adaptations
  636. in applications. Such changes are only allowed in (new) major releases and
  637. require further steps such as bumping library version numbers and/or
  638. adjustments to the symbol versioning file. Please discuss such changes
  639. on the @strong{ffmpeg-devel} mailing list in time to allow forward planning.
  640. @anchor{Criteria for Point Releases}
  641. @section Criteria for Point Releases
  642. Changes that match the following criteria are valid candidates for
  643. inclusion into a point release:
  644. @enumerate
  645. @item
  646. Fixes a security issue, preferably identified by a @strong{CVE
  647. number} issued by @url{http://cve.mitre.org/}.
  648. @item
  649. Fixes a documented bug in @url{https://trac.ffmpeg.org}.
  650. @item
  651. Improves the included documentation.
  652. @item
  653. Retains both source code and binary compatibility with previous
  654. point releases of the same release branch.
  655. @end enumerate
  656. The order for checking the rules is (1 OR 2 OR 3) AND 4.
  657. @section Release Checklist
  658. The release process involves the following steps:
  659. @enumerate
  660. @item
  661. Ensure that the @file{RELEASE} file contains the version number for
  662. the upcoming release.
  663. @item
  664. Add the release at @url{https://trac.ffmpeg.org/admin/ticket/versions}.
  665. @item
  666. Announce the intent to do a release to the mailing list.
  667. @item
  668. Make sure all relevant security fixes have been backported. See
  669. @url{https://ffmpeg.org/security.html}.
  670. @item
  671. Ensure that the FATE regression suite still passes in the release
  672. branch on at least @strong{i386} and @strong{amd64}
  673. (cf. @ref{Regression tests}).
  674. @item
  675. Prepare the release tarballs in @code{bz2} and @code{gz} formats, and
  676. supplementing files that contain @code{gpg} signatures
  677. @item
  678. Publish the tarballs at @url{https://ffmpeg.org/releases}. Create and
  679. push an annotated tag in the form @code{nX}, with @code{X}
  680. containing the version number.
  681. @item
  682. Propose and send a patch to the @strong{ffmpeg-devel} mailing list
  683. with a news entry for the website.
  684. @item
  685. Publish the news entry.
  686. @item
  687. Send an announcement to the mailing list.
  688. @end enumerate
  689. @bye