Gitiles
Code Review Sign In
gerrit-public.fairphone.software / platform / external / mesa3d / 9e4248b20695672bd0fe8668cea290d2aecd36bd / . / docs / submittingpatches.html
blob: 564ad731b24ca590311ed0b223e1a9c797f6f71c [file] [log] [blame]
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]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 <h1>The Mesa 3D Graphics Library</h1>
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="#mailing">Mailing 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>
Emil Velikovf9cdfa32017-02-13 19:23:39 +0000[diff] [blame]28<li><a href="#backports">Sending backports for the stable branch</a>
Timothy Arcerie260bfe2016-11-21 16:30:12 +0000[diff] [blame]29<li><a href="#gittips">Git tips</a>
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]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,
36perhaps, 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 effect individual Mesa/Gallium
40components.
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.
Andres Gomez28158c32016-11-28 18:47:42 +0200[diff] [blame]45<li>Patches should be submitted to <a href="#mailing">mesa-dev</a>
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]46for <a href="#reviewing">review</a> using <code>git send-email</code>.
47
48</ul>
49
50<h2 id="formatting">Patch formatting</h2>
51
52<ul>
53<li>Lines should be limited to 75 characters or less so that git logs
54displayed in 80-column terminals avoid line wrapping. Note that git
55log uses 4 spaces of indentation (4 + 75 &lt; 80).
56<li>The first line should be a short, concise summary of the change prefixed
57with a module name. Examples:
58<pre>
59 mesa: Add support for querying GL_VERTEX_ATTRIB_ARRAY_LONG
60
61 gallium: add PIPE_CAP_DEVICE_RESET_STATUS_QUERY
62
63 i965: Fix missing type in local variable declaration.
64</pre>
65<li>Subsequent patch comments should describe the change in more detail,
66if needed. For example:
67<pre>
68 i965: Remove end-of-thread SEND alignment code.
69
70 This was present in Eric's initial implementation of the compaction code
71 for Sandybridge (commit 077d01b6). There is no documentation saying this
72 is necessary, and removing it causes no regressions in piglit on any
73 platform.
74</pre>
75<li>A "Signed-off-by:" line is not required, but not discouraged either.
Emil Velikov99266ec2017-02-11 12:08:34 +0000[diff] [blame]76<li>If a patch addresses a bugzilla issue, that should be noted in the
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]77patch comment. For example:
78<pre>
79 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89689
80</pre>
81<li>If there have been several revisions to a patch during the review
82process, they should be noted such as in this example:
83<pre>
84 st/mesa: add ARB_texture_stencil8 support (v4)
85
86 if we support stencil texturing, enable texture_stencil8
87 there is no requirement to support native S8 for this,
88 the texture can be converted to x24s8 fine.
89
90 v2: fold fixes from Marek in:
91 a) put S8 last in the list
92 b) fix renderable to always test for d/s renderable
93 fixup the texture case to use a stencil only format
94 for picking the format for the texture view.
95 v3: hit fallback for getteximage
96 v4: put s8 back in front, it shouldn't get picked now (Ilia)
97</pre>
98<li>If someone tested your patch, document it with a line like this:
99<pre>
100 Tested-by: Joe Hacker &lt;jhacker@foo.com&gt;
101</pre>
102<li>If the patch was reviewed (usually the case) or acked by someone,
103that should be documented with:
104<pre>
105 Reviewed-by: Joe Hacker &lt;jhacker@foo.com&gt;
106 Acked-by: Joe Hacker &lt;jhacker@foo.com&gt;
107</pre>
George Kyriazisba28f212016-11-28 17:35:26 +0000[diff] [blame]108<li>If sending later revision of a patch, add all the tags - ack, r-b,
Eric Engestromfae5e1d2016-11-29 11:40:55 +0000[diff] [blame]109Cc: mesa-stable and/or other. This provides reviewers with quick feedback if the
George Kyriazisba28f212016-11-28 17:35:26 +0000[diff] [blame]110patch has already been reviewed.
Emil Velikova2283b52016-11-21 17:53:33 +0000[diff] [blame]111<li>In order for your patch to reach the prospective reviewer easier/faster,
112use the script scripts/get_reviewer.pl to get a list of individuals and include
113them in the CC list.
114<br>
115Please use common sense and do <strong>not</strong> blindly add everyone.
116<br>
117<pre>
118 $ scripts/get_reviewer.pl --help # to get the the help screen
119 $ scripts/get_reviewer.pl -f src/egl/drivers/dri2/platform_android.c
120 Rob Herring <robh@kernel.org> (reviewer:ANDROID EGL SUPPORT,added_lines:188/700=27%,removed_lines:58/283=20%)
121 Tomasz Figa <tfiga@chromium.org> (reviewer:ANDROID EGL SUPPORT,authored:12/41=29%,added_lines:308/700=44%,removed_lines:115/283=41%)
122 Emil Velikov <emil.l.velikov@gmail.com> (authored:13/41=32%,removed_lines:76/283=27%)
123</pre>
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]124</ul>
125
126
127
128<h2 id="testing">Testing Patches</h2>
129
130<p>
131It should go without saying that patches must be tested. In general,
132do whatever testing is prudent.
133</p>
134
135<p>
136You should always run the Mesa test suite before submitting patches.
137The test suite can be run using the 'make check' command. All tests
138must pass before patches will be accepted, this may mean you have
139to update the tests themselves.
140</p>
141
142<p>
143Whenever possible and applicable, test the patch with
Eric Engestrom30cf9ff2017-02-09 02:10:17 +0000[diff] [blame]144<a href="https://piglit.freedesktop.org">Piglit</a> and/or
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]145<a href="https://android.googlesource.com/platform/external/deqp/">dEQP</a>
146to check for regressions.
147</p>
148
149
150<h2 id="mailing">Mailing Patches</h2>
151
152<p>
153Patches should be sent to the mesa-dev mailing list for review:
154<a href="https://lists.freedesktop.org/mailman/listinfo/mesa-dev">
Emil Velikovb571c072016-11-16 11:54:54 +0000[diff] [blame]155mesa-dev@lists.freedesktop.org</a>.
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]156When submitting a patch make sure to use
157<a href="https://git-scm.com/docs/git-send-email">git send-email</a>
158rather than attaching patches to emails. Sending patches as
159attachments prevents people from being able to provide in-line review
160comments.
161</p>
162
163<p>
164When submitting follow-up patches you can use --in-reply-to to make v2, v3,
165etc patches show up as replies to the originals. This usually works well
166when you're sending out updates to individual patches (as opposed to
167re-sending the whole series). Using --in-reply-to makes
168it harder for reviewers to accidentally review old patches.
169</p>
170
171<p>
172When submitting follow-up patches you should also login to
173<a href="https://patchwork.freedesktop.org">patchwork</a> and change the
174state of your old patches to Superseded.
175</p>
176
177<h2 id="reviewing">Reviewing Patches</h2>
178
179<p>
180When you've reviewed a patch on the mailing list, please be unambiguous
181about your review. That is, state either
Emil Velikovb571c072016-11-16 11:54:54 +0000[diff] [blame]182</p>
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]183<pre>
184 Reviewed-by: Joe Hacker &lt;jhacker@foo.com&gt;
185</pre>
186or
187<pre>
188 Acked-by: Joe Hacker &lt;jhacker@foo.com&gt;
189</pre>
Emil Velikovb571c072016-11-16 11:54:54 +0000[diff] [blame]190<p>
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]191Rather than saying just "LGTM" or "Seems OK".
192</p>
193
194<p>
195If small changes are suggested, it's OK to say something like:
Emil Velikovb571c072016-11-16 11:54:54 +0000[diff] [blame]196</p>
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]197<pre>
198 With the above fixes, Reviewed-by: Joe Hacker &lt;jhacker@foo.com&gt;
199</pre>
Emil Velikovb571c072016-11-16 11:54:54 +0000[diff] [blame]200<p>
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]201which tells the patch author that the patch can be committed, as long
202as the issues are resolved first.
203</p>
204
205
206<h2 id="nominations">Nominating a commit for a stable branch</h2>
207
208<p>
Emil Velikov99266ec2017-02-11 12:08:34 +0000[diff] [blame]209There are three ways to nominate a patch for inclusion in the stable branch and
Emil Velikov07384462016-11-16 11:51:50 +0000[diff] [blame]210release.
211</p>
212<ul>
213<li> By adding the Cc: mesa-stable@ tag as described below.
214<li> Sending the commit ID (as seen in master branch) to the mesa-stable@ mailing list.
215<li> Forwarding the patch from the mesa-dev@ mailing list.
216</li>
217</ul>
218<p>
219Note: resending patch identical to one on mesa-dev@ or one that differs only
220by the extra mesa-stable@ tag is <strong>not</strong> recommended.
221</p>
222
223
224<h3 id="thetag">The stable tag</h3>
225
226<p>
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]227If you want a commit to be applied to a stable branch,
228you should add an appropriate note to the commit message.
229</p>
230
231<p>
232Here are some examples of such a note:
233</p>
234<ul>
235 <li>CC: &lt;mesa-stable@lists.freedesktop.org&gt;</li>
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]236</ul>
237
238Simply adding the CC to the mesa-stable list address is adequate to nominate
Emil Velikov9e4248b2017-02-13 19:23:40 +0000[diff] [blame^]239the commit for all the active stable branches. If the commit is not applicable
240for said branch the stable-release manager will reply stating so.
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]241
242This "CC" syntax for patch nomination will cause patches to automatically be
243copied to the mesa-stable@ mailing list when you use "git send-email" to send
Emil Velikov07384462016-11-16 11:51:50 +0000[diff] [blame]244patches to the mesa-dev@ mailing list. If you prefer using --suppress-cc that
Emil Velikov99266ec2017-02-11 12:08:34 +0000[diff] [blame]245won't have any negative effect on the patch nomination.
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]246
Emil Velikov0715ba42016-12-05 16:11:21 +0000[diff] [blame]247<p>
248Note: by removing the tag [as the commit is pushed] the patch is
249<strong>explicitly</strong> rejected from inclusion in the stable branch(es).
250<br>
251Thus, drop the line <strong>only</strong> if you want to cancel the nomination.
252</p>
253
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]254<h2 id="criteria">Criteria for accepting patches to the stable branch</h2>
255
256Mesa has a designated release manager for each stable branch, and the release
Emil Velikovd7e0ff02017-02-13 19:23:38 +0000[diff] [blame]257manager is the only developer that should be pushing changes to these branches.
258Everyone else should nominate patches using the mechanism described above.
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]259
Emil Velikovd7e0ff02017-02-13 19:23:38 +0000[diff] [blame]260The following rules define which patches are accepted and which are not. The
261stable-release manager is also given broad discretion in rejecting patches
262that have been nominated.
263
264<ul>
265 <li>Patch must conform with the <a href="#guidelines">Basic guidelines</a></li>
266
267 <li>Patch must have landed in master first. In case where the original
268 patch is too large and/or otherwise contradicts with the rules set within, a
269 backport is appropriate.</li>
270
271 <li>It must not introduce a regression - be that build or runtime wise.
272
273 Note: If the regression is due to faulty piglit/dEQP/CTS/other test the
274 latter must be fixed first. A reference to the offending test(s) and
275 respective fix(es) should be provided in the nominated patch.</li>
276
277 <li>Patch cannot be larger than 100 lines.</li>
278
279 <li>Patches that move code around with no functional change should be
280 rejected.</li>
281
282 <li>Patch must be a bug fix and not a new feature.
283
284 Note: An exception to this rule, are hardware-enabling "features". For
Emil Velikovf9cdfa32017-02-13 19:23:39 +0000[diff] [blame]285 example, <a href="#backports">backports</a> of new code to support a
286 newly-developed hardware product can be accepted if they can be reasonably
287 determined not to have effects on other hardware.</li>
Emil Velikovd7e0ff02017-02-13 19:23:38 +0000[diff] [blame]288
289 <li>Patch must be reviewed, For example, the commit message has Reviewed-by,
290 Signed-off-by, or Tested-by tags from someone but the author.</li>
291
292 <li>Performance patches are considered only if they provide information
293 about the hardware, program in question and observed improvement. Use numbers
294 to represent your measurements.</li>
295</ul>
296
297If the patch complies with the rules it will be
298<a href="releasing.html#pickntest">cherry-picked</a>. Alternatively the release
299manager will reply to the patch in question stating why the patch has been
300rejected or would request a backport.
301
302A summary of all the picked/rejected patches will be presented in the
303<a href="releasing.html#prerelease">pre-release</a> announcement.
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]304
305The stable-release manager may at times need to force-push changes to the
306stable branches, for example, to drop a previously-picked patch that was later
307identified as causing a regression). These force-pushes may cause changes to
308be lost from the stable branch if developers push things directly. Consider
309yourself warned.
310
Emil Velikovf9cdfa32017-02-13 19:23:39 +0000[diff] [blame]311<h2 id="backports">Sending backports for the stable branch</h2>
312By default merge conflicts are resolved by the stable-release manager. In which
313case he/she should provide a comment about the changes required, alongside the
314<code>Conflicts</code> section. Summary of which will be provided in the
315<a href="releasing.html#prerelease">pre-release</a> announcement.
316<br>
317Developers are interested in sending backports are recommended to use either a
318<code>[BACKPORT #branch]</code> subject prefix or provides similar information
319within the commit summary.
320
Timothy Arcerie260bfe2016-11-21 16:30:12 +0000[diff] [blame]321<h2 id="gittips">Git tips</h2>
322
323<ul>
Emil Velikov6dae5be2016-11-28 17:40:04 +0000[diff] [blame]324<li><code>git rebase -i ...</code> is your friend. Don't be afraid to use it.
325<li>Apply a fixup to commit FOO.
326<pre>
327 git add ...
328 git commit --fixup=FOO
329 git rebase -i --autosquash ...
330</pre>
Timothy Arcerie260bfe2016-11-21 16:30:12 +0000[diff] [blame]331<li>Test for build breakage between patches e.g last 8 commits.
332<pre>
333 git rebase -i --exec="make -j4" HEAD~8
334</pre>
335<li>Sets the default mailing address for your repo.
336<pre>
337 git config --local sendemail.to mesa-dev@lists.freedesktop.org
338</pre>
339<li> Add version to subject line of patch series in this case for the last 8
340commits before sending.
341<pre>
342 git send-email --subject-prefix="PATCH v4" HEAD~8
343 git send-email -v4 @~8 # shorter version, inherited from git format-patch
344</pre>
Emil Velikova2283b52016-11-21 17:53:33 +0000[diff] [blame]345<li> Configure git to use the get_reviewer.pl script interactively. Thus you
346can avoid adding the world to the CC list.
347<pre>
348 git config sendemail.cccmd "./scripts/get_reviewer.pl -i"
349</pre>
Timothy Arcerie260bfe2016-11-21 16:30:12 +0000[diff] [blame]350</ul>
351
Emil Velikov259e65c2016-11-16 00:20:56 +0000[diff] [blame]352
353</div>
354</body>
355</html>