Thank you for your interest in contributing to RIOT! There are many ways to contribute, and we appreciate all of them. You can jump to the major sections of this document using the following links:
If you have questions, please send an email to [email protected] or [email protected] mailing list or chat on #riot-os.
As a reminder, all contributors are expected to follow our Code of Conduct.
Before opening a new feature request, check the existing feature requests if there's one already open on the same topic.
To request new features or enhancements, just open a new feature request issue. Describe your use case, why you need this feature and why this feature is important for RIOT.
While bugs are unfortunate, they're a reality in software. We can't fix what we don't know about, so please report liberally. If you're not sure if something is a bug or not, feel free to file a bug report anyway.
If you believe reporting your bug publicly represents a security risk to RIOT users, please send an email describing the bug to [email protected]. We would appreciate waiting for a 6 months grace period before reporting it on public channels, to allow us adequate time to release the fix.
Before reporting a bug, have a look at open bugs. Maybe someone has already reported your error.
Once you have verified that they haven't, opening an issue is as easy as clicking on this link and filling out the fields.
Each bug report issue uses a template with 5 sections that are there to help other contributors understand your issue and eventually reproduce it:
#### Description
#### Steps to reproduce the issue
#### Expected results
#### Actual results
#### Versions
To fill the Versions
section, you can use the script provided in the RIOT git
repository:
./dist/tools/ci/print_toolchain_versions.sh
In summary, try to include as much information as possible, to help maintainers or other developers to fix the bug quickly.
GitHub's Pull Request (PR) feature is the primary mechanism used to make contributions to the RIOT codebase. GitHub itself has some great documentation on using the Pull Request feature. We use the fork and pull model, where contributors push changes to their personal fork and create pull requests to bring those changes into the source repository.
-
Before opening a new Pull Request, have a look at existing ones. Maybe someone has already opened one about the same thing. If it's the case, you might be able to help with the contribution. Just comment on the PR and ask. Old and stalled PRs are sometimes archived with the "State: archived" label, maybe one of them is also about the same topic.
-
Each Pull Request form uses a template with 3 sections that are here to help maintainers understand your contribution and help them testing it:
#### Contribution description #### Testing procedure #### Issues/PRs references
Please fill each section with as much information as possible.
-
The Pull Request title should reflect what it is about and be in the form:
area of change: description of changes
-
Remember that smaller PRs tend to be merged faster, so keep your changes as concise as possible. They should be confined to a single explainable change, and be runnable on their own. So don't hesitate to split your PRs into smaller ones when possible.
-
In the Pull Request form, we recommend that you leave the "Allow edits from maintainers" check box ticked. This will allow maintainer finalizing your PR by pushing in your branch. In general, this speeds up the PR merge in the main repository. Note that this is not an obligation.
-
Check if your code follows the coding conventions. If it doesn't, you can uncrustify a file:
$ uncrustify -c $RIOTBASE/uncrustify-riot.cfg <your file>
-
RIOT provides static test tools to verify the quality of changes (cppcheck, trailing whitespaces, documentation, etc). These tools are wrapped in a single
make
target:static-test
.Watch out: the command below will rebase your branch on your master branch, so make sure they can be rebased (e.g. there's no potential conflict).
$ make static-test
Use it before opening a PR to perform last time checks.
-
Each commit should target changes of specific parts/modules of RIOT. The commits use the following pattern:
area of code: description of changes
.You can use multi-line commit messages if you want to detail more the changes.
-
Try to answer reviews as quickly as possible to speed up the review process and avoid stalled PRs.
-
Maintainers try their best to review every PR as fast as possible, but they are also only human and it can happen that they miss a few PRs or might be preoccupied with other PRs. If it happens that your PR receive no review for a long time, don't hesitate to gently solicit a review by commenting or by explicitly mentioning a maintainer that you know is knowledgeable in the area of the PR. You can also advertise the PR on [email protected] mailing list and ask for a review there.
You can find more information about RIOT development procedure on this wiki page.
Using git is a bit difficult for newcomers. If you are completely new to git, we recommend that you start by learning it a bit. You can also read the official getting started documentation.
In this section, we give the bare minimum for a better experience with our development workflow on GitHub.
Before you start modifying code, you need to fork the RIOT upstream repository from the RIOT main GitHub page.
If it's your first time with git, configure your name and emails:
$ git config --global user.name = "<your name here>"
$ git config --global user.email = "<your email address here>"
Then clone locally your fork of RIOT (replace account name
with your actual
login on GitHub):
$ git clone [email protected]:<account name>/RIOT.git
You can keep any branch of your local repository up-to-date with the upstream master branch with the following commands:
$ git checkout <branch name>
$ git pull --rebase https://github.com/RIOT-OS/RIOT.git
Use it before opening a PR. This will at least ensure the PR is mergeable but also that it is up-to-date with the upstream repository.
Avoid opening PR from the master
branch of your fork to the master branch of
the RIOT upstream repository: update your master branch and start a new branch
from it.
$ git checkout master
$ git pull --rebase https://github.com/RIOT-OS/RIOT.git
$ git checkout -b <new branch>
# Do your changes, commit, update with latest upstream master
$ git push
To keep the history of changes easier to track for reviewers, it is recommended to push your review request updates in fixup commits.
Let's say your PR contains 3 commits with comments: prefix1: change 1
,
prefix2: change 2
and prefix3: change 3
.
Instead of committing changes in prefix2
in a 4th commit prefix2: change 4
,
you can use the --fixup
option:
$ git add /path/of/prefix2
$ git commit --fixup <prefix2 commit hash>
Squashing a commit is done using the rebase subcommand of git in interactive mode:
$ git rebase master -i
You can find information on rebasing in GitHub rebase documentation.
If you used fixup commits during the review phase, squashing commits can be performed in a single command:
$ git rebase -i --autosquash
Watch out: Don't squash your commit until a maintainer asks you to do it.
Otherwise the history of review changes is lost and for large PRs, it makes it difficult for the reviewer to follow them. It might also happen that you introduce regression and won't be able to recover them from previous commits.
Once squashing is done, you will have to force push your branch to update the PR:
$ git push --force-with-lease
Documentation improvements are always welcome and a good starting point for new contributors. This kind of contribution is merged quite quickly in general.
RIOT documentation is built with doxygen. Doxygen is configured to
parse header (.h) and doc.txt
files in the RIOT source code to generate
the modules, cpus, boards and packages documentation.
General documentation pages are written in Markdown and located in
doc/doxygen/src
.
To generate the documentation, simply run:
$ make doc
from the base directory of the RIOT source code.
The generated documentation is located in doc/doxygen/html