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 a bugzilla issue, that should be noted in the
  79 patch comment.  For example:
  80 <pre>
  81    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89689
  82 </pre>
  83 <li>If a patch addresses a issue introduced with earlier commit, that should be
  84 noted in the patch comment.  For example:
  85 <pre>
  86    Fixes: d7b3707c612 "util/disk_cache: use stat() to check if entry is a directory"
  87 </pre>
  88 <li>You can produce those fixes lines by running
  89 <pre>git config --global alias.fixes "show -s --pretty='format:Fixes: %h (\"%s\")'"</pre>
  90 once and then using <pre>git fixes &lt;sha1&gt;</pre>
  91 <li>If there have been several revisions to a patch during the review
  92 process, they should be noted such as in this example:
  93 <pre>
  94     st/mesa: add ARB_texture_stencil8 support (v4)
  95
  96     if we support stencil texturing, enable texture_stencil8
  97     there is no requirement to support native S8 for this,
  98     the texture can be converted to x24s8 fine.
  99
 100     v2: fold fixes from Marek in:
 101        a) put S8 last in the list
 102        b) fix renderable to always test for d/s renderable
 103         fixup the texture case to use a stencil only format
 104         for picking the format for the texture view.
 105     v3: hit fallback for getteximage
 106     v4: put s8 back in front, it shouldn't get picked now (Ilia)
 107 </pre>
 108 <li>If someone tested your patch, document it with a line like this:
 109 <pre>
 110     Tested-by: Joe Hacker &lt;jhacker@foo.com&gt;
 111 </pre>
 112 <li>If the patch was reviewed (usually the case) or acked by someone,
 113 that should be documented with:
 114 <pre>
 115     Reviewed-by: Joe Hacker &lt;jhacker@foo.com&gt;
 116     Acked-by: Joe Hacker &lt;jhacker@foo.com&gt;
 117 </pre>
 118 <li>If sending later revision of a patch, add all the tags - ack, r-b,
 119 Cc: mesa-stable and/or other. This provides reviewers with quick feedback if the
 120 patch has already been reviewed.
 121 <li>In order for your patch to reach the prospective reviewer easier/faster,
 122 use the script scripts/get_reviewer.pl to get a list of individuals and include
 123 them in the CC list.
 124 <p>
 125 Please use common sense and do <strong>not</strong> blindly add everyone.
 126 </p>
 127 <pre>
 128     $ scripts/get_reviewer.pl --help # to get the help screen
 129     $ scripts/get_reviewer.pl -f src/egl/drivers/dri2/platform_android.c
 130     Rob Herring &lt;robh@kernel.org&gt; (reviewer:ANDROID EGL SUPPORT,added_lines:188/700=27%,removed_lines:58/283=20%)
 131     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%)
 132     Emil Velikov &lt;emil.l.velikov@gmail.com&gt; (authored:13/41=32%,removed_lines:76/283=27%)
 133 </pre>
 134 </ul>
 135
 136
 137
 138 <h2 id="testing">Testing Patches</h2>
 139
 140 <p>
 141 It should go without saying that patches must be tested.  In general,
 142 do whatever testing is prudent.
 143 </p>
 144
 145 <p>
 146 You should always run the Mesa test suite before submitting patches.
 147 The test suite can be run using the 'meson test' command. All tests
 148 must pass before patches will be accepted, this may mean you have
 149 to update the tests themselves.
 150 </p>
 151
 152 <p>
 153 Whenever possible and applicable, test the patch with
 154 <a href="https://piglit.freedesktop.org">Piglit</a> and/or
 155 <a href="https://android.googlesource.com/platform/external/deqp/">dEQP</a>
 156 to check for regressions.
 157 </p>
 158
 159 <p>
 160 As mentioned at the begining, patches should be bisectable.
 161 A good way to test this is to make use of the `git rebase` command,
 162 to run your tests on each commit. Assuming your branch is based off
 163 <code>origin/master</code>, you can run:
 164 </p>
 165 <pre>
 166 $ git rebase --interactive --exec "meson test -C build/" origin/master
 167 </pre>
 168 <p>
 169 replacing <code>"meson test"</code> with whatever other test you want to
 170 run.
 171 </p>
 172
 173
 174 <h2 id="submit">Submitting Patches</h2>
 175
 176 <p>
 177 Patches may be submitted to the Mesa project by
 178 <a href="#mailing">email</a> or with a
 179 GitLab <a href="#merge-request">merge request</a>. To prevent
 180 duplicate code review, only use one method to submit your changes.
 181 </p>
 182
 183 <h3 id="mailing">Mailing Patches</h3>
 184
 185 <p>
 186 Patches may be sent to the mesa-dev mailing list for review:
 187 <a href="https://lists.freedesktop.org/mailman/listinfo/mesa-dev">
 188 mesa-dev@lists.freedesktop.org</a>.
 189 When submitting a patch make sure to use
 190 <a href="https://git-scm.com/docs/git-send-email">git send-email</a>
 191 rather than attaching patches to emails. Sending patches as
 192 attachments prevents people from being able to provide in-line review
 193 comments.
 194 </p>
 195
 196 <p>
 197 When submitting follow-up patches you can use --in-reply-to to make v2, v3,
 198 etc patches show up as replies to the originals. This usually works well
 199 when you're sending out updates to individual patches (as opposed to
 200 re-sending the whole series). Using --in-reply-to makes
 201 it harder for reviewers to accidentally review old patches.
 202 </p>
 203
 204 <p>
 205 When submitting follow-up patches you should also login to
 206 <a href="https://patchwork.freedesktop.org">patchwork</a> and change the
 207 state of your old patches to Superseded.
 208 </p>
 209
 210 <p>
 211 Some companies' mail server automatically append a legal disclaimer,
 212 usually containing something along the lines of "The information in this
 213 email is confidential" and "distribution is strictly prohibited".
 214 </p>
 215 <p>
 216 These legal notices prevent us from being able to accept your patch,
 217 rendering the whole process pointless. Please make sure these are
 218 disabled before sending your patches. (Note that you may need to contact
 219 your email administrator for this.)
 220 </p>
 221
 222 <h3 id="merge-request">GitLab Merge Requests</h3>
 223
 224 <p>
 225   <a href="https://gitlab.freedesktop.org/mesa/mesa">GitLab</a> Merge
 226   Requests (MR) can also be used to submit patches for Mesa.
 227 </p>
 228
 229 <p>
 230   If the MR may have interest for most of the Mesa community, you can
 231   send an email to the mesa-dev email list including a link to the MR.
 232   Don't send the patch to mesa-dev, just the MR link.
 233 </p>
 234 <p>
 235   Add labels to your MR to help reviewers find it. For example:
 236 </p>
 237 <ul>
 238   <li>Mesa changes affecting all drivers: mesa
 239   <li>Hardware vendor specific code: amd, intel, nvidia, ...
 240   <li>Driver specific code: anvil, freedreno, i965, iris, radeonsi,
 241     radv, vc4, ...
 242   <li>Other tag examples: gallium, util
 243 </ul>
 244 <p>
 245   Tick the following when creating the MR. It allows developers to
 246   rebase your work on top of master.
 247 </p>
 248 <pre>Allow commits from members who can merge to the target branch</pre>
 249 <p>
 250   If you revise your patches based on code review and push an update
 251   to your branch, you should maintain a <strong>clean</strong> history
 252   in your patches. There should not be "fixup" patches in the history.
 253   The series should be buildable and functional after every commit
 254   whenever you push the branch.
 255 </p>
 256 <p>
 257   It is your responsibility to keep the MR alive and making progress,
 258   as there are no guarantees that a Mesa dev will independently take
 259   interest in it.
 260 </p>
 261 <p>
 262   Some other notes:
 263 </p>
 264 <ul>
 265   <li>Make changes and update your branch based on feedback
 266   <li>After an update, for the feedback you handled, close the
 267     feedback discussion with the "Resolve Discussion" button. This way
 268     the reviewers know which feedback got handled and which didn't.
 269   <li>Old, stale MR may be closed, but you can reopen it if you
 270     still want to pursue the changes
 271   <li>You should periodically check to see if your MR needs to be
 272     rebased
 273   <li>Make sure your MR is closed if your patches get pushed outside
 274     of GitLab
 275   <li>Please send MRs from a personal fork rather than from the main
 276     Mesa repository, as it clutters it unnecessarily.
 277 </ul>
 278
 279 <h2 id="reviewing">Reviewing Patches</h2>
 280
 281 <p>
 282   To participate in code review, you should monitor the
 283   <a href="https://lists.freedesktop.org/mailman/listinfo/mesa-dev">
 284   mesa-dev</a> email list and the GitLab
 285   Mesa <a href="https://gitlab.freedesktop.org/mesa/mesa/merge_requests">Merge
 286   Requests</a> page.
 287 </p>
 288
 289 <p>
 290 When you've reviewed a patch on the mailing list, please be unambiguous
 291 about your review.  That is, state either
 292 </p>
 293 <pre>
 294     Reviewed-by: Joe Hacker &lt;jhacker@foo.com&gt;
 295 </pre>
 296 or
 297 <pre>
 298     Acked-by: Joe Hacker &lt;jhacker@foo.com&gt;
 299 </pre>
 300 <p>
 301 Rather than saying just "LGTM" or "Seems OK".
 302 </p>
 303
 304 <p>
 305 If small changes are suggested, it's OK to say something like:
 306 </p>
 307 <pre>
 308    With the above fixes, Reviewed-by: Joe Hacker &lt;jhacker@foo.com&gt;
 309 </pre>
 310 <p>
 311 which tells the patch author that the patch can be committed, as long
 312 as the issues are resolved first.
 313 </p>
 314
 315 <p>
 316 These Reviewed-by, Acked-by, and Tested-by tags should also be amended
 317 into commits in a MR before it is merged.
 318 </p>
 319
 320 <p>
 321 When providing a Reviewed-by, Acked-by, or Tested-by tag in a gitlab MR,
 322 enclose the tag in backticks:
 323 </p>
 324 <pre>
 325   `Reviewed-by: Joe Hacker &lt;jhacker@example.com&gt;`</pre>
 326 <p>
 327 This is the markdown format for literal, and will prevent gitlab from hiding
 328 the &lt; and &gt; symbols.
 329 </p>
 330
 331 <p>
 332 Review by non-experts is encouraged.  Understanding how someone else
 333 goes about solving a problem is a great way to learn your way around
 334 the project.  The submitter is expected to evaluate whether they have
 335 an appropriate amount of review feedback from people who also
 336 understand the code before merging their patches.
 337 </p>
 338
 339 <h2 id="nominations">Nominating a commit for a stable branch</h2>
 340
 341 <p>
 342 There are three ways to nominate a patch for inclusion in the stable branch and
 343 release.
 344 </p>
 345 <ul>
 346 <li> By adding the Cc: mesa-stable@ tag as described below.
 347 <li> Sending the commit ID (as seen in master branch) to the mesa-stable@ mailing list.
 348 <li> Forwarding the patch from the mesa-dev@ mailing list.
 349 </li>
 350 </ul>
 351 <p>
 352 Note: resending patch identical to one on mesa-dev@ or one that differs only
 353 by the extra mesa-stable@ tag is <strong>not</strong> recommended.
 354 </p>
 355 <p>
 356 If you are not the author of the original patch, please Cc: them in your
 357 nomination request.
 358 </p>
 359
 360 <p>
 361 The current patch status can be observed in the <a href="releasing.html#stagingbranch">staging branch</a>.
 362 </p>
 363
 364 <h3 id="thetag">The stable tag</h3>
 365
 366 <p>
 367 If you want a commit to be applied to a stable branch,
 368 you should add an appropriate note to the commit message.
 369 </p>
 370
 371 <p>
 372 Here are some examples of such a note:
 373 </p>
 374 <pre>
 375 CC: &lt;mesa-stable@lists.freedesktop.org&gt;
 376 </pre>
 377
 378 Simply adding the CC to the mesa-stable list address is adequate to nominate
 379 the commit for all the active stable branches. If the commit is not applicable
 380 for said branch the stable-release manager will reply stating so.
 381
 382 This "CC" syntax for patch nomination will cause patches to automatically be
 383 copied to the mesa-stable@ mailing list when you use "git send-email" to send
 384 patches to the mesa-dev@ mailing list. If you prefer using --suppress-cc that
 385 won't have any negative effect on the patch nomination.
 386
 387 <p>
 388 Note: by removing the tag [as the commit is pushed] the patch is
 389 <strong>explicitly</strong> rejected from inclusion in the stable branch(es).
 390 Thus, drop the line <strong>only</strong> if you want to cancel the nomination.
 391 </p>
 392
 393 Alternatively, if one uses the "Fixes" tag as described in the "Patch formatting"
 394 section, it nominates a commit for all active stable branches that include the
 395 commit that is referred to.
 396
 397 <h2 id="criteria">Criteria for accepting patches to the stable branch</h2>
 398
 399 Mesa has a designated release manager for each stable branch, and the release
 400 manager is the only developer that should be pushing changes to these branches.
 401 Everyone else should nominate patches using the mechanism described above.
 402
 403 The following rules define which patches are accepted and which are not. The
 404 stable-release manager is also given broad discretion in rejecting patches
 405 that have been nominated.
 406
 407 <ul>
 408   <li>Patch must conform with the <a href="#guidelines">Basic guidelines</a></li>
 409
 410   <li>Patch must have landed in master first. In case where the original
 411   patch is too large and/or otherwise contradicts with the rules set within, a
 412   backport is appropriate.</li>
 413
 414   <li>It must not introduce a regression - be that build or runtime wise.
 415
 416   Note:  If the regression is due to faulty piglit/dEQP/CTS/other test the
 417   latter must be fixed first. A reference to the offending test(s) and
 418   respective fix(es) should be provided in the nominated patch.</li>
 419
 420   <li>Patch cannot be larger than 100 lines.</li>
 421
 422   <li>Patches that move code around with no functional change should be
 423   rejected.</li>
 424
 425   <li>Patch must be a bug fix and not a new feature.
 426
 427   Note: An exception to this rule, are hardware-enabling "features". For
 428   example, <a href="#backports">backports</a> of new code to support a
 429   newly-developed hardware product can be accepted if they can be reasonably
 430   determined not to have effects on other hardware.</li>
 431
 432   <li>Patch must be reviewed, For example, the commit message has Reviewed-by,
 433   Signed-off-by, or Tested-by tags from someone but the author.</li>
 434
 435   <li>Performance patches are considered only if they provide information
 436   about the hardware, program in question and observed improvement. Use numbers
 437   to represent your measurements.</li>
 438 </ul>
 439
 440 If the patch complies with the rules it will be
 441 <a href="releasing.html#pickntest">cherry-picked</a>. Alternatively the release
 442 manager will reply to the patch in question stating why the patch has been
 443 rejected or would request a backport.
 444
 445 A summary of all the picked/rejected patches will be presented in the
 446 <a href="releasing.html#prerelease">pre-release</a> announcement.
 447
 448 The stable-release manager may at times need to force-push changes to the
 449 stable branches, for example, to drop a previously-picked patch that was later
 450 identified as causing a regression). These force-pushes may cause changes to
 451 be lost from the stable branch if developers push things directly. Consider
 452 yourself warned.
 453
 454 <h2 id="backports">Sending backports for the stable branch</h2>
 455 <p>
 456 By default merge conflicts are resolved by the stable-release manager. In which
 457 case he/she should provide a comment about the changes required, alongside the
 458 <code>Conflicts</code> section. Summary of which will be provided in the
 459 <a href="releasing.html#prerelease">pre-release</a> announcement.
 460 </p>
 461
 462 <p>
 463 Developers are interested in sending backports are recommended to use either a
 464 <code>[BACKPORT #branch]</code> subject prefix or provides similar information
 465 within the commit summary.
 466 </p>
 467
 468 <h2 id="gittips">Git tips</h2>
 469
 470 <ul>
 471 <li><code>git rebase -i ...</code> is your friend. Don't be afraid to use it.
 472 <li>Apply a fixup to commit FOO.
 473 <pre>
 474     git add ...
 475     git commit --fixup=FOO
 476     git rebase -i --autosquash ...
 477 </pre>
 478 <li>Test for build breakage between patches e.g last 8 commits.
 479 <pre>
 480     git rebase -i --exec="ninja -C build/" HEAD~8
 481 </pre>
 482 <li>Sets the default mailing address for your repo.
 483 <pre>
 484     git config --local sendemail.to mesa-dev@lists.freedesktop.org
 485 </pre>
 486 <li> Add version to subject line of patch series in this case for the last 8
 487 commits before sending.
 488 <pre>
 489     git send-email --subject-prefix="PATCH v4" HEAD~8
 490     git send-email -v4 @~8 # shorter version, inherited from git format-patch
 491 </pre>
 492 <li> Configure git to use the get_reviewer.pl script interactively. Thus you
 493 can avoid adding the world to the CC list.
 494 <pre>
 495     git config sendemail.cccmd "./scripts/get_reviewer.pl -i"
 496 </pre>
 497 </ul>
 498
 499
 500 </div>
 501 </body>
 502 </html>