11 This document contains the LLVM Developer Policy which defines the project's
12 policy towards developers and their contributions. The intent of this policy is
13 to eliminate miscommunication, rework, and confusion that might arise from the
14 distributed nature of LLVM's development. By stating the policy in clear terms,
15 we hope each developer can know ahead of time what to expect when making LLVM
16 contributions. This policy covers all llvm.org subprojects, including Clang,
19 This policy is also designed to accomplish the following objectives:
21 #. Attract both users and developers to the LLVM project.
23 #. Make life as simple and easy for contributors as possible.
25 #. Keep the top of Subversion trees as stable as possible.
27 #. Establish awareness of the project's :ref:`copyright, license, and patent
28 policies <copyright-license-patents>` with contributors to the project.
30 This policy is aimed at frequent contributors to LLVM. People interested in
31 contributing one-off patches can do so in an informal way by sending them to the
32 `llvm-commits mailing list
33 <http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits>`_ and engaging another
34 developer to see it through the process.
39 This section contains policies that pertain to frequent LLVM developers. We
40 always welcome `one-off patches`_ from people who do not routinely contribute to
41 LLVM, but we expect more from frequent contributors to keep the system as
42 efficient as possible for everyone. Frequent LLVM contributors are expected to
43 meet the following requirements in order for LLVM to maintain a high standard of
49 Developers should stay informed by reading at least the "dev" mailing list for
50 the projects you are interested in, such as `llvmdev
51 <http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev>`_ for LLVM, `cfe-dev
52 <http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev>`_ for Clang, or `lldb-dev
53 <http://lists.cs.uiuc.edu/mailman/listinfo/lldb-dev>`_ for LLDB. If you are
54 doing anything more than just casual work on LLVM, it is suggested that you also
55 subscribe to the "commits" mailing list for the subproject you're interested in,
57 <http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits>`_, `cfe-commits
58 <http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits>`_, or `lldb-commits
59 <http://lists.cs.uiuc.edu/mailman/listinfo/lldb-commits>`_. Reading the
60 "commits" list and paying attention to changes being made by others is a good
61 way to see what other people are interested in and watching the flow of the
64 We recommend that active developers register an email account with `LLVM
65 Bugzilla <http://llvm.org/bugs/>`_ and preferably subscribe to the `llvm-bugs
66 <http://lists.cs.uiuc.edu/mailman/listinfo/llvmbugs>`_ email list to keep track
67 of bugs and enhancements occurring in LLVM. We really appreciate people who are
68 proactive at catching incoming bugs in their components and dealing with them
77 When making a patch for review, the goal is to make it as easy for the reviewer
78 to read it as possible. As such, we recommend that you:
80 #. Make your patch against the Subversion trunk, not a branch, and not an old
81 version of LLVM. This makes it easy to apply the patch. For information on
82 how to check out SVN trunk, please see the `Getting Started
83 Guide <GettingStarted.html#checkout>`_.
85 #. Similarly, patches should be submitted soon after they are generated. Old
86 patches may not apply correctly if the underlying code changes between the
87 time the patch was created and the time it is applied.
89 #. Patches should be made with ``svn diff``, or similar. If you use a
90 different tool, make sure it uses the ``diff -u`` format and that it
91 doesn't contain clutter which makes it hard to read.
93 #. If you are modifying generated files, such as the top-level ``configure``
94 script, please separate out those changes into a separate patch from the rest
97 When sending a patch to a mailing list, it is a good idea to send it as an
98 *attachment* to the message, not embedded into the text of the message. This
99 ensures that your mailer will not mangle the patch when it sends it (e.g. by
100 making whitespace changes or by wrapping lines).
102 *For Thunderbird users:* Before submitting a patch, please open *Preferences >
103 Advanced > General > Config Editor*, find the key
104 ``mail.content_disposition_type``, and set its value to ``1``. Without this
105 setting, Thunderbird sends your attachment using ``Content-Disposition: inline``
106 rather than ``Content-Disposition: attachment``. Apple Mail gamely displays such
107 a file inline, making it difficult to work with for reviewers using that
115 LLVM has a code review policy. Code review is one way to increase the quality of
116 software. We generally follow these policies:
118 #. All developers are required to have significant changes reviewed before they
119 are committed to the repository.
121 #. Code reviews are conducted by email, usually on the llvm-commits list.
123 #. Code can be reviewed either before it is committed or after. We expect major
124 changes to be reviewed before being committed, but smaller changes (or
125 changes where the developer owns the component) can be reviewed after commit.
127 #. The developer responsible for a code change is also responsible for making
128 all necessary review-related changes.
130 #. Code review can be an iterative process, which continues until the patch is
131 ready to be committed.
133 Developers should participate in code reviews as both reviewers and
134 reviewees. If someone is kind enough to review your code, you should return the
135 favor for someone else. Note that anyone is welcome to review and give feedback
136 on a patch, but only people with Subversion write access can approve it.
138 There is a web based code review tool that can optionally be used
139 for code reviews. See :doc:`Phabricator`.
144 The LLVM Project relies on two features of its process to maintain rapid
145 development in addition to the high quality of its source base: the combination
146 of code review plus post-commit review for trusted maintainers. Having both is
147 a great way for the project to take advantage of the fact that most people do
148 the right thing most of the time, and only commit patches without pre-commit
149 review when they are confident they are right.
151 The trick to this is that the project has to guarantee that all patches that are
152 committed are reviewed after they go in: you don't want everyone to assume
153 someone else will review it, allowing the patch to go unreviewed. To solve this
154 problem, we have a notion of an 'owner' for a piece of the code. The sole
155 responsibility of a code owner is to ensure that a commit to their area of the
156 code is appropriately reviewed, either by themself or by someone else. The list
157 of current code owners can be found in the file
158 `CODE_OWNERS.TXT <http://llvm.org/viewvc/llvm-project/llvm/trunk/CODE_OWNERS.TXT?view=markup>`_
159 in the root of the LLVM source tree.
161 Note that code ownership is completely different than reviewers: anyone can
162 review a piece of code, and we welcome code review from anyone who is
163 interested. Code owners are the "last line of defense" to guarantee that all
164 patches that are committed are actually reviewed.
166 Being a code owner is a somewhat unglamorous position, but it is incredibly
167 important for the ongoing success of the project. Because people get busy,
168 interests change, and unexpected things happen, code ownership is purely opt-in,
169 and anyone can choose to resign their "title" at any time. For now, we do not
170 have an official policy on how one gets elected to be a code owner.
172 .. _include a testcase:
177 Developers are required to create test cases for any bugs fixed and any new
178 features added. Some tips for getting your testcase approved:
180 * All feature and regression test cases are added to the ``llvm/test``
181 directory. The appropriate sub-directory should be selected (see the
182 :doc:`Testing Guide <TestingGuide>` for details).
184 * Test cases should be written in `LLVM assembly language <LangRef.html>`_
185 unless the feature or regression being tested requires another language
186 (e.g. the bug being fixed or feature being implemented is in the llvm-gcc C++
187 front-end, in which case it must be written in C++).
189 * Test cases, especially for regressions, should be reduced as much as possible,
190 by `bugpoint <Bugpoint.html>`_ or manually. It is unacceptable to place an
191 entire failing program into ``llvm/test`` as this creates a *time-to-test*
192 burden on all developers. Please keep them short.
194 Note that llvm/test and clang/test are designed for regression and small feature
195 tests only. More extensive test cases (e.g., entire applications, benchmarks,
196 etc) should be added to the ``llvm-test`` test suite. The llvm-test suite is
197 for coverage (correctness, performance, etc) testing, not feature or regression
203 The minimum quality standards that any change must satisfy before being
204 committed to the main development branch are:
206 #. Code must adhere to the `LLVM Coding Standards <CodingStandards.html>`_.
208 #. Code must compile cleanly (no errors, no warnings) on at least one platform.
210 #. Bug fixes and new features should `include a testcase`_ so we know if the
211 fix/feature ever regresses in the future.
213 #. Code must pass the ``llvm/test`` test suite.
215 #. The code must not cause regressions on a reasonable subset of llvm-test,
216 where "reasonable" depends on the contributor's judgement and the scope of
217 the change (more invasive changes require more testing). A reasonable subset
218 might be something like "``llvm-test/MultiSource/Benchmarks``".
220 Additionally, the committer is responsible for addressing any problems found in
221 the future that the change is responsible for. For example:
223 * The code should compile cleanly on all supported platforms.
225 * The changes should not cause any correctness regressions in the ``llvm-test``
226 suite and must not cause any major performance regressions.
228 * The change set should not cause performance or correctness regressions for the
231 * The changes should not cause performance or correctness regressions in code
232 compiled by LLVM on all applicable targets.
234 * You are expected to address any `Bugzilla bugs <http://llvm.org/bugs/>`_ that
235 result from your change.
237 We prefer for this to be handled before submission but understand that it isn't
238 possible to test all of this for every submission. Our build bots and nightly
239 testing infrastructure normally finds these problems. A good rule of thumb is
240 to check the nightly testers for regressions the day after your change. Build
241 bots will directly email you if a group of commits that included yours caused a
242 failure. You are expected to check the build bot messages to see if they are
243 your fault and, if so, fix the breakage.
245 Commits that violate these quality standards (e.g. are very broken) may be
246 reverted. This is necessary when the change blocks other developers from making
247 progress. The developer is welcome to re-commit the change after the problem has
250 Obtaining Commit Access
251 -----------------------
253 We grant commit access to contributors with a track record of submitting high
254 quality patches. If you would like commit access, please send an email to
255 `Chris <mailto:sabre@nondot.org>`_ with the following information:
257 #. The user name you want to commit with, e.g. "hacker".
259 #. The full name and email address you want message to llvm-commits to come
260 from, e.g. "J. Random Hacker <hacker@yoyodyne.com>".
262 #. A "password hash" of the password you want to use, e.g. "``2ACR96qjUqsyM``".
263 Note that you don't ever tell us what your password is, you just give it to
264 us in an encrypted form. To get this, run "``htpasswd``" (a utility that
265 comes with apache) in crypt mode (often enabled with "``-d``"), or find a web
266 page that will do it for you.
268 Once you've been granted commit access, you should be able to check out an LLVM
269 tree with an SVN URL of "https://username@llvm.org/..." instead of the normal
270 anonymous URL of "http://llvm.org/...". The first time you commit you'll have
271 to type in your password. Note that you may get a warning from SVN about an
272 untrusted key, you can ignore this. To verify that your commit access works,
273 please do a test commit (e.g. change a comment or add a blank line). Your first
274 commit to a repository may require the autogenerated email to be approved by a
275 mailing list. This is normal, and will be done when the mailing list owner has
278 If you have recently been granted commit access, these policies apply:
280 #. You are granted *commit-after-approval* to all parts of LLVM. To get
281 approval, submit a `patch`_ to `llvm-commits
282 <http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits>`_. When approved
283 you may commit it yourself.
285 #. You are allowed to commit patches without approval which you think are
286 obvious. This is clearly a subjective decision --- we simply expect you to
287 use good judgement. Examples include: fixing build breakage, reverting
288 obviously broken patches, documentation/comment changes, any other minor
291 #. You are allowed to commit patches without approval to those portions of LLVM
292 that you have contributed or maintain (i.e., have been assigned
293 responsibility for), with the proviso that such commits must not break the
294 build. This is a "trust but verify" policy and commits of this nature are
295 reviewed after they are committed.
297 #. Multiple violations of these policies or a single egregious violation may
298 cause commit access to be revoked.
300 In any case, your changes are still subject to `code review`_ (either before or
301 after they are committed, depending on the nature of the change). You are
302 encouraged to review other peoples' patches as well, but you aren't required
305 .. _discuss the change/gather consensus:
307 Making a Major Change
308 ---------------------
310 When a developer begins a major new project with the aim of contributing it back
311 to LLVM, s/he should inform the community with an email to the `llvmdev
312 <http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev>`_ email list, to the extent
313 possible. The reason for this is to:
315 #. keep the community informed about future changes to LLVM,
317 #. avoid duplication of effort by preventing multiple parties working on the
318 same thing and not knowing about it, and
320 #. ensure that any technical issues around the proposed work are discussed and
321 resolved before any significant work is done.
323 The design of LLVM is carefully controlled to ensure that all the pieces fit
324 together well and are as consistent as possible. If you plan to make a major
325 change to the way LLVM works or want to add a major new extension, it is a good
326 idea to get consensus with the development community before you start working on
329 Once the design of the new feature is finalized, the work itself should be done
330 as a series of `incremental changes`_, not as a long-term development branch.
332 .. _incremental changes:
334 Incremental Development
335 -----------------------
337 In the LLVM project, we do all significant changes as a series of incremental
338 patches. We have a strong dislike for huge changes or long-term development
339 branches. Long-term development branches have a number of drawbacks:
341 #. Branches must have mainline merged into them periodically. If the branch
342 development and mainline development occur in the same pieces of code,
343 resolving merge conflicts can take a lot of time.
345 #. Other people in the community tend to ignore work on branches.
347 #. Huge changes (produced when a branch is merged back onto mainline) are
348 extremely difficult to `code review`_.
350 #. Branches are not routinely tested by our nightly tester infrastructure.
352 #. Changes developed as monolithic large changes often don't work until the
353 entire set of changes is done. Breaking it down into a set of smaller
354 changes increases the odds that any of the work will be committed to the main
357 To address these problems, LLVM uses an incremental development style and we
358 require contributors to follow this practice when making a large/invasive
361 * Large/invasive changes usually have a number of secondary changes that are
362 required before the big change can be made (e.g. API cleanup, etc). These
363 sorts of changes can often be done before the major change is done,
364 independently of that work.
366 * The remaining inter-related work should be decomposed into unrelated sets of
367 changes if possible. Once this is done, define the first increment and get
368 consensus on what the end goal of the change is.
370 * Each change in the set can be stand alone (e.g. to fix a bug), or part of a
371 planned series of changes that works towards the development goal.
373 * Each change should be kept as small as possible. This simplifies your work
374 (into a logical progression), simplifies code review and reduces the chance
375 that you will get negative feedback on the change. Small increments also
376 facilitate the maintenance of a high quality code base.
378 * Often, an independent precursor to a big change is to add a new API and slowly
379 migrate clients to use the new API. Each change to use the new API is often
380 "obvious" and can be committed without review. Once the new API is in place
381 and used, it is much easier to replace the underlying implementation of the
382 API. This implementation change is logically separate from the API
385 If you are interested in making a large change, and this scares you, please make
386 sure to first `discuss the change/gather consensus`_ then ask about the best way
387 to go about making the change.
389 Attribution of Changes
390 ----------------------
392 We believe in correct attribution of contributions to their contributors.
393 However, we do not want the source code to be littered with random attributions
394 "this code written by J. Random Hacker" (this is noisy and distracting). In
395 practice, the revision control system keeps a perfect history of who changed
396 what, and the CREDITS.txt file describes higher-level contributions. If you
397 commit a patch for someone else, please say "patch contributed by J. Random
398 Hacker!" in the commit message.
400 Overall, please do not add contributor names to the source code.
402 .. _copyright-license-patents:
404 Copyright, License, and Patents
405 ===============================
409 This section deals with legal matters but does not provide legal advice. We
410 are not lawyers --- please seek legal counsel from an attorney.
412 This section addresses the issues of copyright, license and patents for the LLVM
413 project. The copyright for the code is held by the individual contributors of
414 the code and the terms of its license to LLVM users and developers is the
415 `University of Illinois/NCSA Open Source License
416 <http://www.opensource.org/licenses/UoI-NCSA.php>`_ (with portions dual licensed
417 under the `MIT License <http://www.opensource.org/licenses/mit-license.php>`_,
418 see below). As contributor to the LLVM project, you agree to allow any
419 contributions to the project to licensed under these terms.
424 The LLVM project does not require copyright assignments, which means that the
425 copyright for the code in the project is held by its respective contributors who
426 have each agreed to release their contributed code under the terms of the `LLVM
429 An implication of this is that the LLVM license is unlikely to ever change:
430 changing it would require tracking down all the contributors to LLVM and getting
431 them to agree that a license change is acceptable for their contribution. Since
432 there are no plans to change the license, this is not a cause for concern.
434 As a contributor to the project, this means that you (or your company) retain
435 ownership of the code you contribute, that it cannot be used in a way that
436 contradicts the license (which is a liberal BSD-style license), and that the
437 license for your contributions won't change without your approval in the
445 We intend to keep LLVM perpetually open source and to use a liberal open source
446 license. **As a contributor to the project, you agree that any contributions be
447 licensed under the terms of the corresponding subproject.** All of the code in
448 LLVM is available under the `University of Illinois/NCSA Open Source License
449 <http://www.opensource.org/licenses/UoI-NCSA.php>`_, which boils down to
452 * You can freely distribute LLVM.
453 * You must retain the copyright notice if you redistribute LLVM.
454 * Binaries derived from LLVM must reproduce the copyright notice (e.g. in an
455 included readme file).
456 * You can't use our names to promote your LLVM derived products.
457 * There's no warranty on LLVM at all.
459 We believe this fosters the widest adoption of LLVM because it **allows
460 commercial products to be derived from LLVM** with few restrictions and without
461 a requirement for making any derived works also open source (i.e. LLVM's
462 license is not a "copyleft" license like the GPL). We suggest that you read the
463 `License <http://www.opensource.org/licenses/UoI-NCSA.php>`_ if further
464 clarification is needed.
466 In addition to the UIUC license, the runtime library components of LLVM
467 (**compiler_rt, libc++, and libclc**) are also licensed under the `MIT License
468 <http://www.opensource.org/licenses/mit-license.php>`_, which does not contain
469 the binary redistribution clause. As a user of these runtime libraries, it
470 means that you can choose to use the code under either license (and thus don't
471 need the binary redistribution clause), and as a contributor to the code that
472 you agree that any contributions to these libraries be licensed under both
473 licenses. We feel that this is important for runtime libraries, because they
474 are implicitly linked into applications and therefore should not subject those
475 applications to the binary redistribution clause. This also means that it is ok
476 to move code from (e.g.) libc++ to the LLVM core without concern, but that code
477 cannot be moved from the LLVM core to libc++ without the copyright owner's
480 Note that the LLVM Project does distribute llvm-gcc and dragonegg, **which are
481 GPL.** This means that anything "linked" into llvm-gcc must itself be compatible
482 with the GPL, and must be releasable under the terms of the GPL. This implies
483 that **any code linked into llvm-gcc and distributed to others may be subject to
484 the viral aspects of the GPL** (for example, a proprietary code generator linked
485 into llvm-gcc must be made available under the GPL). This is not a problem for
486 code already distributed under a more liberal license (like the UIUC license),
487 and GPL-containing subprojects are kept in separate SVN repositories whose
488 LICENSE.txt files specifically indicate that they contain GPL code.
490 We have no plans to change the license of LLVM. If you have questions or
491 comments about the license, please contact the `LLVM Developer's Mailing
492 List <mailto:llvmdev@cs.uiuc.edu>`_.
497 To the best of our knowledge, LLVM does not infringe on any patents (we have
498 actually removed code from LLVM in the past that was found to infringe). Having
499 code in LLVM that infringes on patents would violate an important goal of the
500 project by making it hard or impossible to reuse the code for arbitrary purposes
501 (including commercial use).
503 When contributing code, we expect contributors to notify us of any potential for
504 patent-related trouble with their changes (including from third parties). If
505 you or your employer own the rights to a patent and would like to contribute
506 code to LLVM that relies on it, we require that the copyright owner sign an
507 agreement that allows any other user of LLVM to freely use your patent. Please
508 contact the `oversight group <mailto:llvm-oversight@cs.uiuc.edu>`_ for more