Patches should not mix code changes with code formatting changes (except, perhaps, in very trivial cases.)
Code patches should follow Mesa coding conventions.
Whenever possible, patches should only affect individual Mesa/Gallium components.
Patches should never introduce build breaks and should be bisectable (see
Patches should be properly formatted.
Patches should be sufficiently tested before submitting.
Lines should be limited to 75 characters or less so that Git logs displayed in 80-column terminals avoid line wrapping. Note that
git loguses 4 spaces of indentation (4 + 75 < 80).
The first line should be a short, concise summary of the change prefixed with a module name. Examples:
mesa: Add support for querying GL_VERTEX_ATTRIB_ARRAY_LONG gallium: add PIPE_CAP_DEVICE_RESET_STATUS_QUERY i965: Fix missing type in local variable declaration.
Subsequent patch comments should describe the change in more detail, if needed. For example:
i965: Remove end-of-thread SEND alignment code. This was present in Eric's initial implementation of the compaction code for Sandybridge (commit 077d01b6). There is no documentation saying this is necessary, and removing it causes no regressions in piglit on any platform.
A “Signed-off-by:” line is not required, but not discouraged either.
If a patch addresses an issue in GitLab, use the Closes: tag For example:
Prefer the full URL to just
Closes: #1, since the URL makes it easier to get to the bug page from
Do not use the ``Fixes:`` tag for this! Mesa already uses
Fixes:for something else. See below.
If there have been several revisions to a patch during the review process, they should be noted such as in this example:
st/mesa: add ARB_texture_stencil8 support (v4) if we support stencil texturing, enable texture_stencil8 there is no requirement to support native S8 for this, the texture can be converted to x24s8 fine. v2: fold fixes from Marek in: a) put S8 last in the list b) fix renderable to always test for d/s renderable fixup the texture case to use a stencil only format for picking the format for the texture view. v3: hit fallback for getteximage v4: put s8 back in front, it shouldn't get picked now (Ilia)
If someone tested your patch, document it with a line like this:
Tested-by: Joe Hacker <email@example.com>
If the patch was reviewed (usually the case) or acked by someone, that should be documented with:
Reviewed-by: Joe Hacker <firstname.lastname@example.org> Acked-by: Joe Hacker <email@example.com>
When updating a merge request add all the tags (
Cc: mesa-stableand/or other) to the commit messages. This provides reviewers with quick feedback if the patch has already been reviewed.
If a patch addresses a issue introduced with earlier commit, that should be noted in the commit message. For example:
Fixes: d7b3707c612 ("util/disk_cache: use stat() to check if entry is a directory")
You can produce those fixes lines by running this command once:
git config --global alias.fixes "show -s --pretty='format:Fixes: %h (\"%s\")'"
After that, using
git fixes <sha1> will print the full line for you.
The stable tag¶
If you want a commit to be applied to a stable branch, you should add an appropriate note to the commit message.
Fixes: tag as described in Patch formatting
is the preferred way to nominate a commit that should be backported.
There are scripts that will figure out which releases to apply the patch
to automatically, so you don’t need to figure it out.
Alternatively, you may use a “CC:” tag. Here are some examples of such a note:
Cc: mesa-stable Cc: 20.0 <mesa-stable> CC: 20.0 19.3 <mesa-stable>
Using the CC tag should include the stable branches you want to nominate the patch to. If you do not provide any version it is nominated to all active stable branches.
It should go without saying that patches must be tested. In general, do whatever testing is prudent.
You should always run the Mesa test suite before submitting patches. The test suite can be run using the ‘meson test’ command. All tests must pass before patches will be accepted, this may mean you have to update the tests themselves.
As mentioned at the beginning, patches should be bisectable. A good way
to test this is to make use of the `git rebase` command, to run your
tests on each commit. Assuming your branch is based off
origin/master, you can run:
$ git rebase --interactive --exec "meson test -C build/" origin/master
"meson test" with whatever other test you want to run.
Patches are submitted to the Mesa project via a GitLab Merge Request.
Add labels to your MR to help reviewers find it. For example:
Mesa changes affecting all drivers: mesa
Hardware vendor specific code: amd, intel, nvidia, …
Driver specific code: anvil, freedreno, i965, iris, radeonsi, radv, vc4, …
Other tag examples: gallium, util
Tick the following when creating the MR. It allows developers to rebase your work on top of master.
Allow commits from members who can merge to the target branch
If you revise your patches based on code review and push an update to your branch, you should maintain a clean history in your patches. There should not be “fixup” patches in the history. The series should be buildable and functional after every commit whenever you push the branch.
It is your responsibility to keep the MR alive and making progress, as there are no guarantees that a Mesa dev will independently take interest in it.
Some other notes:
Make changes and update your branch based on feedback
After an update, for the feedback you handled, close the feedback discussion with the “Resolve Discussion” button. This way the reviewers know which feedback got handled and which didn’t.
Old, stale MR may be closed, but you can reopen it if you still want to pursue the changes
You should periodically check to see if your MR needs to be rebased
Make sure your MR is closed if your patches get pushed outside of GitLab
Please send MRs from a personal fork rather than from the main Mesa repository, as it clutters it unnecessarily.
To participate in code review, you can monitor the GitLab Mesa Merge Requests page, and/or register for notifications in your GitLab settings.
When you’ve reviewed a patch, please be unambiguous about your review. That is, state either
Reviewed-by: Joe Hacker <firstname.lastname@example.org>
Acked-by: Joe Hacker <email@example.com>
Rather than saying just “LGTM” or “Seems OK”.
If small changes are suggested, it’s OK to say something like:
With the above fixes, Reviewed-by: Joe Hacker <firstname.lastname@example.org>
which tells the patch author that the patch can be committed, as long as the issues are resolved first.
These Reviewed-by, Acked-by, and Tested-by tags should also be amended into commits in a MR before it is merged.
When providing a Reviewed-by, Acked-by, or Tested-by tag in a GitLab MR, enclose the tag in backticks:
`Reviewed-by: Joe Hacker <email@example.com>`
This is the markdown format for literal, and will prevent GitLab from hiding the < and > symbols.
Review by non-experts is encouraged. Understanding how someone else goes about solving a problem is a great way to learn your way around the project. The submitter is expected to evaluate whether they have an appropriate amount of review feedback from people who also understand the code before merging their patches.
Nominating a commit for a stable branch¶
There are several ways to nominate a patch for inclusion in the stable branch and release. In order or preference:
By adding the
Fixes:tag in the commit message as described above, if you are fixing a specific commit.
By adding the
Cc: mesa-stabletag in the commit message as described above.
By submitting a merge request against the
staging/year.quarterbranch on GitLab.
Please DO NOT send patches to firstname.lastname@example.org, it is not monitored actively and is a historical artifact.
If you are not the author of the original patch, please Cc: them in your nomination request.
The current patch status can be observed in the staging branch.
Criteria for accepting patches to the stable branch¶
Mesa has a designated release manager for each stable branch, and the release manager is the only developer that should be pushing changes to these branches. Everyone else should nominate patches using the mechanism described above. The following rules define which patches are accepted and which are not. The stable-release manager is also given broad discretion in rejecting patches that have been nominated.
Patch must conform with the Basic guidelines
Patch must have landed in master first. In case where the original patch is too large and/or otherwise contradicts with the rules set within, a backport is appropriate.
It must not introduce a regression - be that build or runtime wise.
If the regression is due to faulty piglit/dEQP/CTS/other test the latter must be fixed first. A reference to the offending test(s) and respective fix(es) should be provided in the nominated patch.
Patch cannot be larger than 100 lines.
Patches that move code around with no functional change should be rejected.
Patch must be a bug fix and not a new feature.
An exception to this rule, are hardware-enabling “features”. For example, backports of new code to support a newly-developed hardware product can be accepted if they can be reasonably determined not to have effects on other hardware.
Patch must be reviewed, For example, the commit message has Reviewed-by, Signed-off-by, or Tested-by tags from someone but the author.
Performance patches are considered only if they provide information about the hardware, program in question and observed improvement. Use numbers to represent your measurements.
If the patch complies with the rules it will be cherry-picked. Alternatively the release manager will reply to the patch in question stating why the patch has been rejected or would request a backport. The stable-release manager may at times need to force-push changes to the stable branches, for example, to drop a previously-picked patch that was later identified as causing a regression). These force-pushes may cause changes to be lost from the stable branch if developers push things directly. Consider yourself warned.
Sending backports for the stable branch¶
By default merge conflicts are resolved by the stable-release manager. The release maintainer should resolve trivial conflicts, but for complex conflicts they should ask the original author to provide a backport or de-nominate the patch.
For patches that either need to be nominated after they’ve landed in master, or that are known ahead of time to not not apply cleanly to a stable branch (such as due to a rename), using a GitLab MR is most appropriate. The MR should be based on and target the staging/year.quarter branch, not on the year.quarter branch, per the stable branch policy. Assigning the MR to release maintainer for said branch or mentioning them is helpful, but not required.
git rebase -i ...is your friend. Don’t be afraid to use it.
Apply a fixup to commit FOO.
git add ... git commit --fixup=FOO git rebase -i --autosquash ...
Test for build breakage between patches e.g last 8 commits.
git rebase -i --exec="ninja -C build/" HEAD~8