docs: Add use of Closes: tag for closing gitlab issues
[mesa.git] / docs / submittingpatches.html
1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
2 <html lang="en">
3 <head>
4 <meta http-equiv="content-type" content="text/html; charset=utf-8">
5 <title>Submitting Patches</title>
6 <link rel="stylesheet" type="text/css" href="mesa.css">
7 </head>
8 <body>
9
10 <div class="header">
11 The Mesa 3D Graphics Library
12 </div>
13
14 <iframe src="contents.html"></iframe>
15 <div class="content">
16
17 <h1>Submitting Patches</h1>
18
19
20 <ul>
21 <li><a href="#guidelines">Basic guidelines</a>
22 <li><a href="#formatting">Patch formatting</a>
23 <li><a href="#testing">Testing Patches</a>
24 <li><a href="#submit">Submitting Patches</a>
25 <li><a href="#reviewing">Reviewing Patches</a>
26 <li><a href="#nominations">Nominating a commit for a stable branch</a>
27 <li><a href="#criteria">Criteria for accepting patches to the stable branch</a>
28 <li><a href="#backports">Sending backports for the stable branch</a>
29 <li><a href="#gittips">Git tips</a>
30 </ul>
31
32 <h2 id="guidelines">Basic guidelines</h2>
33
34 <ul>
35 <li>Patches should not mix code changes with code formatting changes (except,
36 perhaps, in very trivial cases.)
37 <li>Code patches should follow Mesa
38 <a href="codingstyle.html" target="_parent">coding conventions</a>.
39 <li>Whenever possible, patches should only affect individual Mesa/Gallium
40 components.
41 <li>Patches should never introduce build breaks and should be bisectable (see
42 <code>git bisect</code>.)
43 <li>Patches should be properly <a href="#formatting">formatted</a>.
44 <li>Patches should be sufficiently <a href="#testing">tested</a> before submitting.
45 <li>Patches should be <a href="#submit">submitted</a>
46 to <a href="#mailing">mesa-dev</a> or with
47 a <a href="#merge-request">merge request</a>
48 for <a href="#reviewing">review</a>.
49
50 </ul>
51
52 <h2 id="formatting">Patch formatting</h2>
53
54 <ul>
55 <li>Lines should be limited to 75 characters or less so that git logs
56 displayed in 80-column terminals avoid line wrapping. Note that git
57 log uses 4 spaces of indentation (4 + 75 &lt; 80).
58 <li>The first line should be a short, concise summary of the change prefixed
59 with a module name. Examples:
60 <pre>
61 mesa: Add support for querying GL_VERTEX_ATTRIB_ARRAY_LONG
62
63 gallium: add PIPE_CAP_DEVICE_RESET_STATUS_QUERY
64
65 i965: Fix missing type in local variable declaration.
66 </pre>
67 <li>Subsequent patch comments should describe the change in more detail,
68 if needed. For example:
69 <pre>
70 i965: Remove end-of-thread SEND alignment code.
71
72 This was present in Eric's initial implementation of the compaction code
73 for Sandybridge (commit 077d01b6). There is no documentation saying this
74 is necessary, and removing it causes no regressions in piglit on any
75 platform.
76 </pre>
77 <li>A "Signed-off-by:" line is not required, but not discouraged either.
78 <li>If a patch addresses an issue in gitlab, use the Closes: tag
79 For example:
80 <pre>
81 Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/1
82 </pre>
83 <p>Prefer the full url to just <pre>Closes: #1</pre>, since the url makes it
84 easier to get to the bug page from <pre>git log</pre></p>
85 <b>Do not use the Fixes: tag for this!</b> Mesa already uses Fixes for something else.
86
87 <li>If a patch addresses a issue introduced with earlier commit, that should be
88 noted in the patch comment. For example:
89 <pre>
90 Fixes: d7b3707c612 "util/disk_cache: use stat() to check if entry is a directory"
91 </pre>
92 <li>You can produce those fixes lines by running
93 <pre>git config --global alias.fixes "show -s --pretty='format:Fixes: %h (\"%s\")'"</pre>
94 once and then using <pre>git fixes &lt;sha1&gt;</pre>
95 <li>If there have been several revisions to a patch during the review
96 process, they should be noted such as in this example:
97 <pre>
98 st/mesa: add ARB_texture_stencil8 support (v4)
99
100 if we support stencil texturing, enable texture_stencil8
101 there is no requirement to support native S8 for this,
102 the texture can be converted to x24s8 fine.
103
104 v2: fold fixes from Marek in:
105 a) put S8 last in the list
106 b) fix renderable to always test for d/s renderable
107 fixup the texture case to use a stencil only format
108 for picking the format for the texture view.
109 v3: hit fallback for getteximage
110 v4: put s8 back in front, it shouldn't get picked now (Ilia)
111 </pre>
112 <li>If someone tested your patch, document it with a line like this:
113 <pre>
114 Tested-by: Joe Hacker &lt;jhacker@foo.com&gt;
115 </pre>
116 <li>If the patch was reviewed (usually the case) or acked by someone,
117 that should be documented with:
118 <pre>
119 Reviewed-by: Joe Hacker &lt;jhacker@foo.com&gt;
120 Acked-by: Joe Hacker &lt;jhacker@foo.com&gt;
121 </pre>
122 <li>If sending later revision of a patch, add all the tags - ack, r-b,
123 Cc: mesa-stable and/or other. This provides reviewers with quick feedback if the
124 patch has already been reviewed.
125 <li>In order for your patch to reach the prospective reviewer easier/faster,
126 use the script scripts/get_reviewer.pl to get a list of individuals and include
127 them in the CC list.
128 <p>
129 Please use common sense and do <strong>not</strong> blindly add everyone.
130 </p>
131 <pre>
132 $ scripts/get_reviewer.pl --help # to get the help screen
133 $ scripts/get_reviewer.pl -f src/egl/drivers/dri2/platform_android.c
134 Rob Herring &lt;robh@kernel.org&gt; (reviewer:ANDROID EGL SUPPORT,added_lines:188/700=27%,removed_lines:58/283=20%)
135 Tomasz Figa &lt;tfiga@chromium.org&gt; (reviewer:ANDROID EGL SUPPORT,authored:12/41=29%,added_lines:308/700=44%,removed_lines:115/283=41%)
136 Emil Velikov &lt;emil.l.velikov@gmail.com&gt; (authored:13/41=32%,removed_lines:76/283=27%)
137 </pre>
138 </ul>
139
140
141
142 <h2 id="testing">Testing Patches</h2>
143
144 <p>
145 It should go without saying that patches must be tested. In general,
146 do whatever testing is prudent.
147 </p>
148
149 <p>
150 You should always run the Mesa test suite before submitting patches.
151 The test suite can be run using the 'meson test' command. All tests
152 must pass before patches will be accepted, this may mean you have
153 to update the tests themselves.
154 </p>
155
156 <p>
157 Whenever possible and applicable, test the patch with
158 <a href="https://piglit.freedesktop.org">Piglit</a> and/or
159 <a href="https://android.googlesource.com/platform/external/deqp/">dEQP</a>
160 to check for regressions.
161 </p>
162
163 <p>
164 As mentioned at the begining, patches should be bisectable.
165 A good way to test this is to make use of the `git rebase` command,
166 to run your tests on each commit. Assuming your branch is based off
167 <code>origin/master</code>, you can run:
168 </p>
169 <pre>
170 $ git rebase --interactive --exec "meson test -C build/" origin/master
171 </pre>
172 <p>
173 replacing <code>"meson test"</code> with whatever other test you want to
174 run.
175 </p>
176
177
178 <h2 id="submit">Submitting Patches</h2>
179
180 <p>
181 Patches may be submitted to the Mesa project by
182 <a href="#mailing">email</a> or with a
183 GitLab <a href="#merge-request">merge request</a>. To prevent
184 duplicate code review, only use one method to submit your changes.
185 </p>
186
187 <h3 id="mailing">Mailing Patches</h3>
188
189 <p>
190 Patches may be sent to the mesa-dev mailing list for review:
191 <a href="https://lists.freedesktop.org/mailman/listinfo/mesa-dev">
192 mesa-dev@lists.freedesktop.org</a>.
193 When submitting a patch make sure to use
194 <a href="https://git-scm.com/docs/git-send-email">git send-email</a>
195 rather than attaching patches to emails. Sending patches as
196 attachments prevents people from being able to provide in-line review
197 comments.
198 </p>
199
200 <p>
201 When submitting follow-up patches you can use --in-reply-to to make v2, v3,
202 etc patches show up as replies to the originals. This usually works well
203 when you're sending out updates to individual patches (as opposed to
204 re-sending the whole series). Using --in-reply-to makes
205 it harder for reviewers to accidentally review old patches.
206 </p>
207
208 <p>
209 When submitting follow-up patches you should also login to
210 <a href="https://patchwork.freedesktop.org">patchwork</a> and change the
211 state of your old patches to Superseded.
212 </p>
213
214 <p>
215 Some companies' mail server automatically append a legal disclaimer,
216 usually containing something along the lines of "The information in this
217 email is confidential" and "distribution is strictly prohibited".
218 </p>
219 <p>
220 These legal notices prevent us from being able to accept your patch,
221 rendering the whole process pointless. Please make sure these are
222 disabled before sending your patches. (Note that you may need to contact
223 your email administrator for this.)
224 </p>
225
226 <h3 id="merge-request">GitLab Merge Requests</h3>
227
228 <p>
229 <a href="https://gitlab.freedesktop.org/mesa/mesa">GitLab</a> Merge
230 Requests (MR) can also be used to submit patches for Mesa.
231 </p>
232
233 <p>
234 If the MR may have interest for most of the Mesa community, you can
235 send an email to the mesa-dev email list including a link to the MR.
236 Don't send the patch to mesa-dev, just the MR link.
237 </p>
238 <p>
239 Add labels to your MR to help reviewers find it. For example:
240 </p>
241 <ul>
242 <li>Mesa changes affecting all drivers: mesa
243 <li>Hardware vendor specific code: amd, intel, nvidia, ...
244 <li>Driver specific code: anvil, freedreno, i965, iris, radeonsi,
245 radv, vc4, ...
246 <li>Other tag examples: gallium, util
247 </ul>
248 <p>
249 Tick the following when creating the MR. It allows developers to
250 rebase your work on top of master.
251 </p>
252 <pre>Allow commits from members who can merge to the target branch</pre>
253 <p>
254 If you revise your patches based on code review and push an update
255 to your branch, you should maintain a <strong>clean</strong> history
256 in your patches. There should not be "fixup" patches in the history.
257 The series should be buildable and functional after every commit
258 whenever you push the branch.
259 </p>
260 <p>
261 It is your responsibility to keep the MR alive and making progress,
262 as there are no guarantees that a Mesa dev will independently take
263 interest in it.
264 </p>
265 <p>
266 Some other notes:
267 </p>
268 <ul>
269 <li>Make changes and update your branch based on feedback
270 <li>After an update, for the feedback you handled, close the
271 feedback discussion with the "Resolve Discussion" button. This way
272 the reviewers know which feedback got handled and which didn't.
273 <li>Old, stale MR may be closed, but you can reopen it if you
274 still want to pursue the changes
275 <li>You should periodically check to see if your MR needs to be
276 rebased
277 <li>Make sure your MR is closed if your patches get pushed outside
278 of GitLab
279 <li>Please send MRs from a personal fork rather than from the main
280 Mesa repository, as it clutters it unnecessarily.
281 </ul>
282
283 <h2 id="reviewing">Reviewing Patches</h2>
284
285 <p>
286 To participate in code review, you should monitor the
287 <a href="https://lists.freedesktop.org/mailman/listinfo/mesa-dev">
288 mesa-dev</a> email list and the GitLab
289 Mesa <a href="https://gitlab.freedesktop.org/mesa/mesa/merge_requests">Merge
290 Requests</a> page.
291 </p>
292
293 <p>
294 When you've reviewed a patch on the mailing list, please be unambiguous
295 about your review. That is, state either
296 </p>
297 <pre>
298 Reviewed-by: Joe Hacker &lt;jhacker@foo.com&gt;
299 </pre>
300 or
301 <pre>
302 Acked-by: Joe Hacker &lt;jhacker@foo.com&gt;
303 </pre>
304 <p>
305 Rather than saying just "LGTM" or "Seems OK".
306 </p>
307
308 <p>
309 If small changes are suggested, it's OK to say something like:
310 </p>
311 <pre>
312 With the above fixes, Reviewed-by: Joe Hacker &lt;jhacker@foo.com&gt;
313 </pre>
314 <p>
315 which tells the patch author that the patch can be committed, as long
316 as the issues are resolved first.
317 </p>
318
319 <p>
320 These Reviewed-by, Acked-by, and Tested-by tags should also be amended
321 into commits in a MR before it is merged.
322 </p>
323
324 <p>
325 When providing a Reviewed-by, Acked-by, or Tested-by tag in a gitlab MR,
326 enclose the tag in backticks:
327 </p>
328 <pre>
329 `Reviewed-by: Joe Hacker &lt;jhacker@example.com&gt;`</pre>
330 <p>
331 This is the markdown format for literal, and will prevent gitlab from hiding
332 the &lt; and &gt; symbols.
333 </p>
334
335 <p>
336 Review by non-experts is encouraged. Understanding how someone else
337 goes about solving a problem is a great way to learn your way around
338 the project. The submitter is expected to evaluate whether they have
339 an appropriate amount of review feedback from people who also
340 understand the code before merging their patches.
341 </p>
342
343 <h2 id="nominations">Nominating a commit for a stable branch</h2>
344
345 <p>
346 There are three ways to nominate a patch for inclusion in the stable branch and
347 release.
348 </p>
349 <ul>
350 <li> By adding the Cc: mesa-stable@ tag as described below.
351 <li> Sending the commit ID (as seen in master branch) to the mesa-stable@ mailing list.
352 <li> Forwarding the patch from the mesa-dev@ mailing list.
353 </li>
354 </ul>
355 <p>
356 Note: resending patch identical to one on mesa-dev@ or one that differs only
357 by the extra mesa-stable@ tag is <strong>not</strong> recommended.
358 </p>
359 <p>
360 If you are not the author of the original patch, please Cc: them in your
361 nomination request.
362 </p>
363
364 <p>
365 The current patch status can be observed in the <a href="releasing.html#stagingbranch">staging branch</a>.
366 </p>
367
368 <h3 id="thetag">The stable tag</h3>
369
370 <p>
371 If you want a commit to be applied to a stable branch,
372 you should add an appropriate note to the commit message.
373 </p>
374
375 <p>
376 Here are some examples of such a note:
377 </p>
378 <pre>
379 CC: &lt;mesa-stable@lists.freedesktop.org&gt;
380 </pre>
381
382 Simply adding the CC to the mesa-stable list address is adequate to nominate
383 the commit for all the active stable branches. If the commit is not applicable
384 for said branch the stable-release manager will reply stating so.
385
386 This "CC" syntax for patch nomination will cause patches to automatically be
387 copied to the mesa-stable@ mailing list when you use "git send-email" to send
388 patches to the mesa-dev@ mailing list. If you prefer using --suppress-cc that
389 won't have any negative effect on the patch nomination.
390
391 <p>
392 Note: by removing the tag [as the commit is pushed] the patch is
393 <strong>explicitly</strong> rejected from inclusion in the stable branch(es).
394 Thus, drop the line <strong>only</strong> if you want to cancel the nomination.
395 </p>
396
397 Alternatively, if one uses the "Fixes" tag as described in the "Patch formatting"
398 section, it nominates a commit for all active stable branches that include the
399 commit that is referred to.
400
401 <h2 id="criteria">Criteria for accepting patches to the stable branch</h2>
402
403 Mesa has a designated release manager for each stable branch, and the release
404 manager is the only developer that should be pushing changes to these branches.
405 Everyone else should nominate patches using the mechanism described above.
406
407 The following rules define which patches are accepted and which are not. The
408 stable-release manager is also given broad discretion in rejecting patches
409 that have been nominated.
410
411 <ul>
412 <li>Patch must conform with the <a href="#guidelines">Basic guidelines</a></li>
413
414 <li>Patch must have landed in master first. In case where the original
415 patch is too large and/or otherwise contradicts with the rules set within, a
416 backport is appropriate.</li>
417
418 <li>It must not introduce a regression - be that build or runtime wise.
419
420 Note: If the regression is due to faulty piglit/dEQP/CTS/other test the
421 latter must be fixed first. A reference to the offending test(s) and
422 respective fix(es) should be provided in the nominated patch.</li>
423
424 <li>Patch cannot be larger than 100 lines.</li>
425
426 <li>Patches that move code around with no functional change should be
427 rejected.</li>
428
429 <li>Patch must be a bug fix and not a new feature.
430
431 Note: An exception to this rule, are hardware-enabling "features". For
432 example, <a href="#backports">backports</a> of new code to support a
433 newly-developed hardware product can be accepted if they can be reasonably
434 determined not to have effects on other hardware.</li>
435
436 <li>Patch must be reviewed, For example, the commit message has Reviewed-by,
437 Signed-off-by, or Tested-by tags from someone but the author.</li>
438
439 <li>Performance patches are considered only if they provide information
440 about the hardware, program in question and observed improvement. Use numbers
441 to represent your measurements.</li>
442 </ul>
443
444 If the patch complies with the rules it will be
445 <a href="releasing.html#pickntest">cherry-picked</a>. Alternatively the release
446 manager will reply to the patch in question stating why the patch has been
447 rejected or would request a backport.
448
449 A summary of all the picked/rejected patches will be presented in the
450 <a href="releasing.html#prerelease">pre-release</a> announcement.
451
452 The stable-release manager may at times need to force-push changes to the
453 stable branches, for example, to drop a previously-picked patch that was later
454 identified as causing a regression). These force-pushes may cause changes to
455 be lost from the stable branch if developers push things directly. Consider
456 yourself warned.
457
458 <h2 id="backports">Sending backports for the stable branch</h2>
459 <p>
460 By default merge conflicts are resolved by the stable-release manager. In which
461 case he/she should provide a comment about the changes required, alongside the
462 <code>Conflicts</code> section. Summary of which will be provided in the
463 <a href="releasing.html#prerelease">pre-release</a> announcement.
464 </p>
465
466 <p>
467 Developers are interested in sending backports are recommended to use either a
468 <code>[BACKPORT #branch]</code> subject prefix or provides similar information
469 within the commit summary.
470 </p>
471
472 <h2 id="gittips">Git tips</h2>
473
474 <ul>
475 <li><code>git rebase -i ...</code> is your friend. Don't be afraid to use it.
476 <li>Apply a fixup to commit FOO.
477 <pre>
478 git add ...
479 git commit --fixup=FOO
480 git rebase -i --autosquash ...
481 </pre>
482 <li>Test for build breakage between patches e.g last 8 commits.
483 <pre>
484 git rebase -i --exec="ninja -C build/" HEAD~8
485 </pre>
486 <li>Sets the default mailing address for your repo.
487 <pre>
488 git config --local sendemail.to mesa-dev@lists.freedesktop.org
489 </pre>
490 <li> Add version to subject line of patch series in this case for the last 8
491 commits before sending.
492 <pre>
493 git send-email --subject-prefix="PATCH v4" HEAD~8
494 git send-email -v4 @~8 # shorter version, inherited from git format-patch
495 </pre>
496 <li> Configure git to use the get_reviewer.pl script interactively. Thus you
497 can avoid adding the world to the CC list.
498 <pre>
499 git config sendemail.cccmd "./scripts/get_reviewer.pl -i"
500 </pre>
501 </ul>
502
503
504 </div>
505 </body>
506 </html>