Commit | Line | Data |
---|---|---|
1da177e4 | 1 | |
5903019b MCC |
2 | How to Get Your Change Into the Linux Kernel or Care And Operation Of Your Linus Torvalds |
3 | ========================================================================================= | |
1da177e4 LT |
4 | |
5 | For a person or company who wishes to submit a change to the Linux | |
6 | kernel, the process can sometimes be daunting if you're not familiar | |
7 | with "the system." This text is a collection of suggestions which | |
8 | can greatly increase the chances of your change being accepted. | |
9 | ||
d00c4559 JC |
10 | This document contains a large number of suggestions in a relatively terse |
11 | format. For detailed information on how the kernel development process | |
12 | works, see Documentation/development-process. Also, read | |
13 | Documentation/SubmitChecklist for a list of items to check before | |
14 | submitting code. If you are submitting a driver, also read | |
082bd1ca JC |
15 | Documentation/SubmittingDrivers; for device tree binding patches, read |
16 | Documentation/devicetree/bindings/submitting-patches.txt. | |
1da177e4 | 17 | |
9b2c7677 MCC |
18 | Many of these steps describe the default behavior of the ``git`` version |
19 | control system; if you use ``git`` to prepare your patches, you'll find much | |
8e3072a2 | 20 | of the mechanical work done for you, though you'll still need to prepare |
9b2c7677 | 21 | and document a sensible set of patches. In general, use of ``git`` will make |
d00c4559 | 22 | your life as a kernel developer easier. |
1da177e4 | 23 | |
5903019b MCC |
24 | Creating and Sending your Change |
25 | ******************************** | |
1da177e4 LT |
26 | |
27 | ||
7994cc15 JC |
28 | 0) Obtain a current source tree |
29 | ------------------------------- | |
30 | ||
31 | If you do not have a repository with the current kernel source handy, use | |
9b2c7677 | 32 | ``git`` to obtain one. You'll want to start with the mainline repository, |
5903019b | 33 | which can be grabbed with:: |
7994cc15 | 34 | |
5903019b | 35 | git clone git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git |
7994cc15 JC |
36 | |
37 | Note, however, that you may not want to develop against the mainline tree | |
38 | directly. Most subsystem maintainers run their own trees and want to see | |
5903019b | 39 | patches prepared against those trees. See the **T:** entry for the subsystem |
7994cc15 JC |
40 | in the MAINTAINERS file to find that tree, or simply ask the maintainer if |
41 | the tree is not listed there. | |
42 | ||
43 | It is still possible to download kernel releases via tarballs (as described | |
44 | in the next section), but that is the hard way to do kernel development. | |
1da177e4 | 45 | |
5903019b MCC |
46 | 1) ``diff -up`` |
47 | --------------- | |
1da177e4 | 48 | |
5903019b | 49 | If you must generate your patches by hand, use ``diff -up`` or ``diff -uprN`` |
7994cc15 | 50 | to create patches. Git generates patches in this form by default; if |
9b2c7677 | 51 | you're using ``git``, you can skip this section entirely. |
1da177e4 LT |
52 | |
53 | All changes to the Linux kernel occur in the form of patches, as | |
9b2c7677 MCC |
54 | generated by :manpage:`diff(1)`. When creating your patch, make sure to |
55 | create it in "unified diff" format, as supplied by the ``-u`` argument | |
56 | to :manpage:`diff(1)`. | |
5903019b | 57 | Also, please use the ``-p`` argument which shows which C function each |
9b2c7677 | 58 | change is in - that makes the resultant ``diff`` a lot easier to read. |
1da177e4 LT |
59 | Patches should be based in the root kernel source directory, |
60 | not in any lower subdirectory. | |
61 | ||
5903019b | 62 | To create a patch for a single file, it is often sufficient to do:: |
1da177e4 | 63 | |
d00c4559 | 64 | SRCTREE= linux |
1da177e4 LT |
65 | MYFILE= drivers/net/mydriver.c |
66 | ||
67 | cd $SRCTREE | |
68 | cp $MYFILE $MYFILE.orig | |
69 | vi $MYFILE # make your change | |
70 | cd .. | |
71 | diff -up $SRCTREE/$MYFILE{.orig,} > /tmp/patch | |
72 | ||
73 | To create a patch for multiple files, you should unpack a "vanilla", | |
9b2c7677 | 74 | or unmodified kernel source tree, and generate a ``diff`` against your |
5903019b | 75 | own source tree. For example:: |
1da177e4 | 76 | |
d00c4559 | 77 | MYSRC= /devel/linux |
1da177e4 | 78 | |
d00c4559 JC |
79 | tar xvfz linux-3.19.tar.gz |
80 | mv linux-3.19 linux-3.19-vanilla | |
81 | diff -uprN -X linux-3.19-vanilla/Documentation/dontdiff \ | |
82 | linux-3.19-vanilla $MYSRC > /tmp/patch | |
1da177e4 | 83 | |
5903019b | 84 | ``dontdiff`` is a list of files which are generated by the kernel during |
9b2c7677 | 85 | the build process, and should be ignored in any :manpage:`diff(1)`-generated |
d00c4559 | 86 | patch. |
1da177e4 LT |
87 | |
88 | Make sure your patch does not include any extra files which do not | |
89 | belong in a patch submission. Make sure to review your patch -after- | |
9b2c7677 | 90 | generating it with :manpage:`diff(1)`, to ensure accuracy. |
1da177e4 | 91 | |
8e3072a2 | 92 | If your changes produce a lot of deltas, you need to split them into |
5903019b MCC |
93 | individual patches which modify things in logical stages; see |
94 | :ref:`split_changes`. This will facilitate review by other kernel developers, | |
8e3072a2 | 95 | very important if you want your patch accepted. |
1da177e4 | 96 | |
9b2c7677 MCC |
97 | If you're using ``git``, ``git rebase -i`` can help you with this process. If |
98 | you're not using ``git``, ``quilt`` <http://savannah.nongnu.org/projects/quilt> | |
8e3072a2 | 99 | is another popular alternative. |
84da7c08 | 100 | |
5903019b | 101 | .. _describe_changes: |
84da7c08 | 102 | |
5903019b MCC |
103 | 2) Describe your changes |
104 | ------------------------ | |
1da177e4 | 105 | |
7b9828d4 JW |
106 | Describe your problem. Whether your patch is a one-line bug fix or |
107 | 5000 lines of a new feature, there must be an underlying problem that | |
108 | motivated you to do this work. Convince the reviewer that there is a | |
109 | problem worth fixing and that it makes sense for them to read past the | |
110 | first paragraph. | |
111 | ||
112 | Describe user-visible impact. Straight up crashes and lockups are | |
113 | pretty convincing, but not all bugs are that blatant. Even if the | |
114 | problem was spotted during code review, describe the impact you think | |
115 | it can have on users. Keep in mind that the majority of Linux | |
116 | installations run kernels from secondary stable trees or | |
117 | vendor/product-specific trees that cherry-pick only specific patches | |
118 | from upstream, so include anything that could help route your change | |
119 | downstream: provoking circumstances, excerpts from dmesg, crash | |
120 | descriptions, performance regressions, latency spikes, lockups, etc. | |
121 | ||
122 | Quantify optimizations and trade-offs. If you claim improvements in | |
123 | performance, memory consumption, stack footprint, or binary size, | |
124 | include numbers that back them up. But also describe non-obvious | |
125 | costs. Optimizations usually aren't free but trade-offs between CPU, | |
126 | memory, and readability; or, when it comes to heuristics, between | |
127 | different workloads. Describe the expected downsides of your | |
128 | optimization so that the reviewer can weigh costs against benefits. | |
129 | ||
130 | Once the problem is established, describe what you are actually doing | |
131 | about it in technical detail. It's important to describe the change | |
132 | in plain English for the reviewer to verify that the code is behaving | |
133 | as you intend it to. | |
1da177e4 | 134 | |
2ae19aca TT |
135 | The maintainer will thank you if you write your patch description in a |
136 | form which can be easily pulled into Linux's source code management | |
9b2c7677 | 137 | system, ``git``, as a "commit log". See :ref:`explicit_in_reply_to`. |
2ae19aca | 138 | |
7b9828d4 JW |
139 | Solve only one problem per patch. If your description starts to get |
140 | long, that's a sign that you probably need to split up your patch. | |
5903019b | 141 | See :ref:`split_changes`. |
1da177e4 | 142 | |
d89b1945 RD |
143 | When you submit or resubmit a patch or patch series, include the |
144 | complete patch description and justification for it. Don't just | |
145 | say that this is version N of the patch (series). Don't expect the | |
d00c4559 | 146 | subsystem maintainer to refer back to earlier patch versions or referenced |
d89b1945 RD |
147 | URLs to find the patch description and put that into the patch. |
148 | I.e., the patch (series) and its description should be self-contained. | |
d00c4559 | 149 | This benefits both the maintainers and reviewers. Some reviewers |
d89b1945 RD |
150 | probably didn't even receive earlier versions of the patch. |
151 | ||
74a475ac JT |
152 | Describe your changes in imperative mood, e.g. "make xyzzy do frotz" |
153 | instead of "[This patch] makes xyzzy do frotz" or "[I] changed xyzzy | |
154 | to do frotz", as if you are giving orders to the codebase to change | |
155 | its behaviour. | |
156 | ||
d89b1945 | 157 | If the patch fixes a logged bug entry, refer to that bug entry by |
9547c706 JT |
158 | number and URL. If the patch follows from a mailing list discussion, |
159 | give a URL to the mailing list archive; use the https://lkml.kernel.org/ | |
9b2c7677 | 160 | redirector with a ``Message-Id``, to ensure that the links cannot become |
9547c706 JT |
161 | stale. |
162 | ||
163 | However, try to make your explanation understandable without external | |
164 | resources. In addition to giving a URL to a mailing list archive or | |
165 | bug, summarize the relevant points of the discussion that led to the | |
166 | patch as submitted. | |
1da177e4 | 167 | |
0af52703 GU |
168 | If you want to refer to a specific commit, don't just refer to the |
169 | SHA-1 ID of the commit. Please also include the oneline summary of | |
170 | the commit, to make it easier for reviewers to know what it is about. | |
5903019b | 171 | Example:: |
0af52703 GU |
172 | |
173 | Commit e21d2170f36602ae2708 ("video: remove unnecessary | |
174 | platform_set_drvdata()") removed the unnecessary | |
175 | platform_set_drvdata(), but left the variable "dev" unused, | |
176 | delete it. | |
177 | ||
7994cc15 JC |
178 | You should also be sure to use at least the first twelve characters of the |
179 | SHA-1 ID. The kernel repository holds a *lot* of objects, making | |
180 | collisions with shorter IDs a real possibility. Bear in mind that, even if | |
181 | there is no collision with your six-character ID now, that condition may | |
182 | change five years from now. | |
183 | ||
8401aa1f | 184 | If your patch fixes a bug in a specific commit, e.g. you found an issue using |
9b2c7677 MCC |
185 | ``git bisect``, please use the 'Fixes:' tag with the first 12 characters of |
186 | the SHA-1 ID, and the one line summary. For example:: | |
8401aa1f JK |
187 | |
188 | Fixes: e21d2170f366 ("video: remove unnecessary platform_set_drvdata()") | |
189 | ||
9b2c7677 MCC |
190 | The following ``git config`` settings can be used to add a pretty format for |
191 | outputting the above style in the ``git log`` or ``git show`` commands:: | |
8401aa1f JK |
192 | |
193 | [core] | |
194 | abbrev = 12 | |
195 | [pretty] | |
196 | fixes = Fixes: %h (\"%s\") | |
1da177e4 | 197 | |
5903019b MCC |
198 | .. _split_changes: |
199 | ||
200 | 3) Separate your changes | |
201 | ------------------------ | |
1da177e4 | 202 | |
5903019b | 203 | Separate each **logical change** into a separate patch. |
1da177e4 LT |
204 | |
205 | For example, if your changes include both bug fixes and performance | |
206 | enhancements for a single driver, separate those changes into two | |
207 | or more patches. If your changes include an API update, and a new | |
208 | driver which uses that new API, separate those into two patches. | |
209 | ||
210 | On the other hand, if you make a single change to numerous files, | |
211 | group those changes into a single patch. Thus a single logical change | |
212 | is contained within a single patch. | |
213 | ||
d00c4559 JC |
214 | The point to remember is that each patch should make an easily understood |
215 | change that can be verified by reviewers. Each patch should be justifiable | |
216 | on its own merits. | |
217 | ||
1da177e4 | 218 | If one patch depends on another patch in order for a change to be |
5903019b | 219 | complete, that is OK. Simply note **"this patch depends on patch X"** |
1da177e4 LT |
220 | in your patch description. |
221 | ||
7994cc15 JC |
222 | When dividing your change into a series of patches, take special care to |
223 | ensure that the kernel builds and runs properly after each patch in the | |
5903019b | 224 | series. Developers using ``git bisect`` to track down a problem can end up |
7994cc15 JC |
225 | splitting your patch series at any point; they will not thank you if you |
226 | introduce bugs in the middle. | |
227 | ||
5b0ed2c6 XVP |
228 | If you cannot condense your patch set into a smaller set of patches, |
229 | then only post say 15 or so at a time and wait for review and integration. | |
230 | ||
231 | ||
1da177e4 | 232 | |
5903019b MCC |
233 | 4) Style-check your changes |
234 | --------------------------- | |
0a920b5b AW |
235 | |
236 | Check your patch for basic style violations, details of which can be | |
237 | found in Documentation/CodingStyle. Failure to do so simply wastes | |
f56d35e7 | 238 | the reviewers time and will get your patch rejected, probably |
0a920b5b AW |
239 | without even being read. |
240 | ||
6de16eba JC |
241 | One significant exception is when moving code from one file to |
242 | another -- in this case you should not modify the moved code at all in | |
243 | the same patch which moves it. This clearly delineates the act of | |
244 | moving the code and your changes. This greatly aids review of the | |
245 | actual differences and allows tools to better track the history of | |
246 | the code itself. | |
247 | ||
248 | Check your patches with the patch style checker prior to submission | |
249 | (scripts/checkpatch.pl). Note, though, that the style checker should be | |
250 | viewed as a guide, not as a replacement for human judgment. If your code | |
251 | looks better with a violation then its probably best left alone. | |
0a920b5b | 252 | |
6de16eba JC |
253 | The checker reports at three levels: |
254 | - ERROR: things that are very likely to be wrong | |
255 | - WARNING: things requiring careful review | |
256 | - CHECK: things requiring thought | |
0a920b5b | 257 | |
6de16eba JC |
258 | You should be able to justify all violations that remain in your |
259 | patch. | |
0a920b5b AW |
260 | |
261 | ||
5903019b MCC |
262 | 5) Select the recipients for your patch |
263 | --------------------------------------- | |
1da177e4 | 264 | |
ccae8616 JC |
265 | You should always copy the appropriate subsystem maintainer(s) on any patch |
266 | to code that they maintain; look through the MAINTAINERS file and the | |
267 | source code revision history to see who those maintainers are. The | |
268 | script scripts/get_maintainer.pl can be very useful at this step. If you | |
d6eff078 | 269 | cannot find a maintainer for the subsystem you are working on, Andrew |
ccae8616 | 270 | Morton (akpm@linux-foundation.org) serves as a maintainer of last resort. |
1da177e4 | 271 | |
ccae8616 JC |
272 | You should also normally choose at least one mailing list to receive a copy |
273 | of your patch set. linux-kernel@vger.kernel.org functions as a list of | |
274 | last resort, but the volume on that list has caused a number of developers | |
275 | to tune it out. Look in the MAINTAINERS file for a subsystem-specific | |
276 | list; your patch will probably get more attention there. Please do not | |
277 | spam unrelated lists, though. | |
1da177e4 | 278 | |
ccae8616 JC |
279 | Many kernel-related lists are hosted on vger.kernel.org; you can find a |
280 | list of them at http://vger.kernel.org/vger-lists.html. There are | |
281 | kernel-related lists hosted elsewhere as well, though. | |
5b0ed2c6 XVP |
282 | |
283 | Do not send more than 15 patches at once to the vger mailing lists!!! | |
284 | ||
1da177e4 | 285 | Linus Torvalds is the final arbiter of all changes accepted into the |
e00bfcbf | 286 | Linux kernel. His e-mail address is <torvalds@linux-foundation.org>. |
ccae8616 JC |
287 | He gets a lot of e-mail, and, at this point, very few patches go through |
288 | Linus directly, so typically you should do your best to -avoid- | |
e00bfcbf | 289 | sending him e-mail. |
1da177e4 | 290 | |
ccae8616 JC |
291 | If you have a patch that fixes an exploitable security bug, send that patch |
292 | to security@kernel.org. For severe bugs, a short embargo may be considered | |
253508ca | 293 | to allow distributors to get the patch out to users; in such cases, |
ccae8616 | 294 | obviously, the patch should not be sent to any public lists. |
1da177e4 | 295 | |
ccae8616 | 296 | Patches that fix a severe bug in a released kernel should be directed |
5903019b | 297 | toward the stable maintainers by putting a line like this:: |
1da177e4 | 298 | |
ccae8616 | 299 | Cc: stable@vger.kernel.org |
1da177e4 | 300 | |
8cda4c3a LD |
301 | into the sign-off area of your patch (note, NOT an email recipient). You |
302 | should also read Documentation/stable_kernel_rules.txt in addition to this | |
303 | file. | |
1da177e4 | 304 | |
ccae8616 JC |
305 | Note, however, that some subsystem maintainers want to come to their own |
306 | conclusions on which patches should go to the stable trees. The networking | |
307 | maintainer, in particular, would rather not see individual developers | |
308 | adding lines like the above to their patches. | |
5b0ed2c6 | 309 | |
ccae8616 JC |
310 | If changes affect userland-kernel interfaces, please send the MAN-PAGES |
311 | maintainer (as listed in the MAINTAINERS file) a man-pages patch, or at | |
312 | least a notification of the change, so that some information makes its way | |
313 | into the manual pages. User-space API changes should also be copied to | |
5903019b | 314 | linux-api@vger.kernel.org. |
1da177e4 LT |
315 | |
316 | For small patches you may want to CC the Trivial Patch Monkey | |
82d27b2b MH |
317 | trivial@kernel.org which collects "trivial" patches. Have a look |
318 | into the MAINTAINERS file for its current manager. | |
5903019b | 319 | |
82d27b2b | 320 | Trivial patches must qualify for one of the following rules: |
5903019b | 321 | |
9b2c7677 MCC |
322 | - Spelling fixes in documentation |
323 | - Spelling fixes for errors which could break :manpage:`grep(1)` | |
324 | - Warning fixes (cluttering with useless warnings is bad) | |
325 | - Compilation fixes (only if they are actually correct) | |
326 | - Runtime fixes (only if they actually fix things) | |
327 | - Removing use of deprecated functions/macros | |
328 | - Contact detail and documentation fixes | |
329 | - Non-portable code replaced by portable code (even in arch-specific, | |
330 | since people copy, as long as it's trivial) | |
331 | - Any fix by the author/maintainer of the file (ie. patch monkey | |
332 | in re-transmission mode) | |
84da7c08 | 333 | |
1da177e4 LT |
334 | |
335 | ||
5903019b MCC |
336 | 6) No MIME, no links, no compression, no attachments. Just plain text |
337 | ---------------------------------------------------------------------- | |
1da177e4 LT |
338 | |
339 | Linus and other kernel developers need to be able to read and comment | |
340 | on the changes you are submitting. It is important for a kernel | |
341 | developer to be able to "quote" your changes, using standard e-mail | |
342 | tools, so that they may comment on specific portions of your code. | |
343 | ||
bdc89213 | 344 | For this reason, all patches should be submitted by e-mail "inline". |
9b2c7677 MCC |
345 | |
346 | .. warning:: | |
347 | ||
348 | Be wary of your editor's word-wrap corrupting your patch, | |
349 | if you choose to cut-n-paste your patch. | |
1da177e4 LT |
350 | |
351 | Do not attach the patch as a MIME attachment, compressed or not. | |
352 | Many popular e-mail applications will not always transmit a MIME | |
353 | attachment as plain text, making it impossible to comment on your | |
354 | code. A MIME attachment also takes Linus a bit more time to process, | |
355 | decreasing the likelihood of your MIME-attached change being accepted. | |
356 | ||
357 | Exception: If your mailer is mangling patches then someone may ask | |
358 | you to re-send them using MIME. | |
359 | ||
097091c0 MO |
360 | See Documentation/email-clients.txt for hints about configuring |
361 | your e-mail client so that it sends your patches untouched. | |
1da177e4 | 362 | |
5903019b MCC |
363 | 7) E-mail size |
364 | -------------- | |
1da177e4 LT |
365 | |
366 | Large changes are not appropriate for mailing lists, and some | |
4932be77 | 367 | maintainers. If your patch, uncompressed, exceeds 300 kB in size, |
1da177e4 | 368 | it is preferred that you store your patch on an Internet-accessible |
d00c4559 JC |
369 | server, and provide instead a URL (link) pointing to your patch. But note |
370 | that if your patch exceeds 300 kB, it almost certainly needs to be broken up | |
371 | anyway. | |
1da177e4 | 372 | |
5903019b MCC |
373 | 8) Respond to review comments |
374 | ----------------------------- | |
1da177e4 | 375 | |
0eea2314 JC |
376 | Your patch will almost certainly get comments from reviewers on ways in |
377 | which the patch can be improved. You must respond to those comments; | |
378 | ignoring reviewers is a good way to get ignored in return. Review comments | |
379 | or questions that do not lead to a code change should almost certainly | |
380 | bring about a comment or changelog entry so that the next reviewer better | |
381 | understands what is going on. | |
1da177e4 | 382 | |
0eea2314 JC |
383 | Be sure to tell the reviewers what changes you are making and to thank them |
384 | for their time. Code review is a tiring and time-consuming process, and | |
385 | reviewers sometimes get grumpy. Even in that case, though, respond | |
386 | politely and address the problems they have pointed out. | |
1da177e4 | 387 | |
1da177e4 | 388 | |
5903019b MCC |
389 | 9) Don't get discouraged - or impatient |
390 | --------------------------------------- | |
1da177e4 | 391 | |
0eea2314 JC |
392 | After you have submitted your change, be patient and wait. Reviewers are |
393 | busy people and may not get to your patch right away. | |
1da177e4 | 394 | |
0eea2314 JC |
395 | Once upon a time, patches used to disappear into the void without comment, |
396 | but the development process works more smoothly than that now. You should | |
397 | receive comments within a week or so; if that does not happen, make sure | |
398 | that you have sent your patches to the right place. Wait for a minimum of | |
399 | one week before resubmitting or pinging reviewers - possibly longer during | |
400 | busy times like merge windows. | |
1da177e4 | 401 | |
1da177e4 | 402 | |
ccae8616 | 403 | 10) Include PATCH in the subject |
d00c4559 | 404 | -------------------------------- |
1da177e4 LT |
405 | |
406 | Due to high e-mail traffic to Linus, and to linux-kernel, it is common | |
407 | convention to prefix your subject line with [PATCH]. This lets Linus | |
408 | and other kernel developers more easily distinguish patches from other | |
409 | e-mail discussions. | |
410 | ||
411 | ||
412 | ||
ccae8616 | 413 | 11) Sign your work |
d00c4559 | 414 | ------------------ |
1da177e4 LT |
415 | |
416 | To improve tracking of who did what, especially with patches that can | |
417 | percolate to their final resting place in the kernel through several | |
418 | layers of maintainers, we've introduced a "sign-off" procedure on | |
419 | patches that are being emailed around. | |
420 | ||
421 | The sign-off is a simple line at the end of the explanation for the | |
422 | patch, which certifies that you wrote it or otherwise have the right to | |
db12fb83 | 423 | pass it on as an open-source patch. The rules are pretty simple: if you |
1da177e4 LT |
424 | can certify the below: |
425 | ||
5903019b MCC |
426 | Developer's Certificate of Origin 1.1 |
427 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ | |
1da177e4 | 428 | |
5903019b | 429 | By making a contribution to this project, I certify that: |
1da177e4 LT |
430 | |
431 | (a) The contribution was created in whole or in part by me and I | |
432 | have the right to submit it under the open source license | |
433 | indicated in the file; or | |
434 | ||
435 | (b) The contribution is based upon previous work that, to the best | |
436 | of my knowledge, is covered under an appropriate open source | |
437 | license and I have the right under that license to submit that | |
438 | work with modifications, whether created in whole or in part | |
439 | by me, under the same open source license (unless I am | |
440 | permitted to submit under a different license), as indicated | |
441 | in the file; or | |
442 | ||
443 | (c) The contribution was provided directly to me by some other | |
444 | person who certified (a), (b) or (c) and I have not modified | |
445 | it. | |
446 | ||
e00bfcbf SB |
447 | (d) I understand and agree that this project and the contribution |
448 | are public and that a record of the contribution (including all | |
449 | personal information I submit with it, including my sign-off) is | |
450 | maintained indefinitely and may be redistributed consistent with | |
451 | this project or the open source license(s) involved. | |
cbd83da8 | 452 | |
5903019b | 453 | then you just add a line saying:: |
1da177e4 | 454 | |
9fd5559c | 455 | Signed-off-by: Random J Developer <random@developer.example.org> |
1da177e4 | 456 | |
af45f32d GK |
457 | using your real name (sorry, no pseudonyms or anonymous contributions.) |
458 | ||
1da177e4 LT |
459 | Some people also put extra tags at the end. They'll just be ignored for |
460 | now, but you can do this to mark internal company procedures or just | |
e00bfcbf | 461 | point out some special detail about the sign-off. |
1da177e4 | 462 | |
adbd5886 WT |
463 | If you are a subsystem or branch maintainer, sometimes you need to slightly |
464 | modify patches you receive in order to merge them, because the code is not | |
465 | exactly the same in your tree and the submitters'. If you stick strictly to | |
466 | rule (c), you should ask the submitter to rediff, but this is a totally | |
467 | counter-productive waste of time and energy. Rule (b) allows you to adjust | |
468 | the code, but then it is very impolite to change one submitter's code and | |
469 | make him endorse your bugs. To solve this problem, it is recommended that | |
470 | you add a line between the last Signed-off-by header and yours, indicating | |
471 | the nature of your changes. While there is nothing mandatory about this, it | |
472 | seems like prepending the description with your mail and/or name, all | |
473 | enclosed in square brackets, is noticeable enough to make it obvious that | |
5903019b | 474 | you are responsible for last-minute changes. Example:: |
adbd5886 WT |
475 | |
476 | Signed-off-by: Random J Developer <random@developer.example.org> | |
477 | [lucky@maintainer.example.org: struct foo moved from foo.c to foo.h] | |
478 | Signed-off-by: Lucky K Maintainer <lucky@maintainer.example.org> | |
479 | ||
305af08c | 480 | This practice is particularly helpful if you maintain a stable branch and |
adbd5886 WT |
481 | want at the same time to credit the author, track changes, merge the fix, |
482 | and protect the submitter from complaints. Note that under no circumstances | |
483 | can you change the author's identity (the From header), as it is the one | |
484 | which appears in the changelog. | |
485 | ||
305af08c | 486 | Special note to back-porters: It seems to be a common and useful practice |
adbd5886 WT |
487 | to insert an indication of the origin of a patch at the top of the commit |
488 | message (just after the subject line) to facilitate tracking. For instance, | |
5903019b | 489 | here's what we see in a 3.x-stable release:: |
adbd5886 | 490 | |
5903019b | 491 | Date: Tue Oct 7 07:26:38 2014 -0400 |
adbd5886 | 492 | |
7994cc15 | 493 | libata: Un-break ATA blacklist |
adbd5886 | 494 | |
7994cc15 | 495 | commit 1c40279960bcd7d52dbdf1d466b20d24b99176c8 upstream. |
adbd5886 | 496 | |
5903019b | 497 | And here's what might appear in an older kernel once a patch is backported:: |
adbd5886 WT |
498 | |
499 | Date: Tue May 13 22:12:27 2008 +0200 | |
500 | ||
501 | wireless, airo: waitbusy() won't delay | |
502 | ||
503 | [backport of 2.6 commit b7acbdfbd1f277c1eb23f344f899cfa4cd0bf36a] | |
504 | ||
505 | Whatever the format, this information provides a valuable help to people | |
7994cc15 | 506 | tracking your trees, and to people trying to troubleshoot bugs in your |
adbd5886 WT |
507 | tree. |
508 | ||
1da177e4 | 509 | |
ccae8616 | 510 | 12) When to use Acked-by: and Cc: |
d00c4559 | 511 | --------------------------------- |
0a920b5b | 512 | |
0f44cd23 AM |
513 | The Signed-off-by: tag indicates that the signer was involved in the |
514 | development of the patch, or that he/she was in the patch's delivery path. | |
515 | ||
516 | If a person was not directly involved in the preparation or handling of a | |
517 | patch but wishes to signify and record their approval of it then they can | |
d00c4559 | 518 | ask to have an Acked-by: line added to the patch's changelog. |
0f44cd23 AM |
519 | |
520 | Acked-by: is often used by the maintainer of the affected code when that | |
521 | maintainer neither contributed to nor forwarded the patch. | |
522 | ||
523 | Acked-by: is not as formal as Signed-off-by:. It is a record that the acker | |
524 | has at least reviewed the patch and has indicated acceptance. Hence patch | |
525 | mergers will sometimes manually convert an acker's "yep, looks good to me" | |
d00c4559 JC |
526 | into an Acked-by: (but note that it is usually better to ask for an |
527 | explicit ack). | |
0f44cd23 AM |
528 | |
529 | Acked-by: does not necessarily indicate acknowledgement of the entire patch. | |
530 | For example, if a patch affects multiple subsystems and has an Acked-by: from | |
531 | one subsystem maintainer then this usually indicates acknowledgement of just | |
532 | the part which affects that maintainer's code. Judgement should be used here. | |
ef40203a | 533 | When in doubt people should refer to the original discussion in the mailing |
0f44cd23 AM |
534 | list archives. |
535 | ||
ef40203a | 536 | If a person has had the opportunity to comment on a patch, but has not |
5903019b | 537 | provided such comments, you may optionally add a ``Cc:`` tag to the patch. |
ef40203a | 538 | This is the only tag which might be added without an explicit action by the |
d00c4559 JC |
539 | person it names - but it should indicate that this person was copied on the |
540 | patch. This tag documents that potentially interested parties | |
541 | have been included in the discussion. | |
0f44cd23 | 542 | |
ef40203a | 543 | |
ccae8616 | 544 | 13) Using Reported-by:, Tested-by:, Reviewed-by:, Suggested-by: and Fixes: |
d00c4559 | 545 | -------------------------------------------------------------------------- |
bbb0a424 | 546 | |
d75ef707 DC |
547 | The Reported-by tag gives credit to people who find bugs and report them and it |
548 | hopefully inspires them to help us again in the future. Please note that if | |
549 | the bug was reported in private, then ask for permission first before using the | |
550 | Reported-by tag. | |
ef40203a JC |
551 | |
552 | A Tested-by: tag indicates that the patch has been successfully tested (in | |
553 | some environment) by the person named. This tag informs maintainers that | |
554 | some testing has been performed, provides a means to locate testers for | |
555 | future patches, and ensures credit for the testers. | |
556 | ||
557 | Reviewed-by:, instead, indicates that the patch has been reviewed and found | |
558 | acceptable according to the Reviewer's Statement: | |
559 | ||
5903019b MCC |
560 | Reviewer's statement of oversight |
561 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ | |
ef40203a | 562 | |
5903019b | 563 | By offering my Reviewed-by: tag, I state that: |
ef40203a | 564 | |
5903019b | 565 | (a) I have carried out a technical review of this patch to |
ef40203a JC |
566 | evaluate its appropriateness and readiness for inclusion into |
567 | the mainline kernel. | |
568 | ||
569 | (b) Any problems, concerns, or questions relating to the patch | |
570 | have been communicated back to the submitter. I am satisfied | |
571 | with the submitter's response to my comments. | |
572 | ||
573 | (c) While there may be things that could be improved with this | |
574 | submission, I believe that it is, at this time, (1) a | |
575 | worthwhile modification to the kernel, and (2) free of known | |
576 | issues which would argue against its inclusion. | |
577 | ||
578 | (d) While I have reviewed the patch and believe it to be sound, I | |
579 | do not (unless explicitly stated elsewhere) make any | |
580 | warranties or guarantees that it will achieve its stated | |
581 | purpose or function properly in any given situation. | |
582 | ||
583 | A Reviewed-by tag is a statement of opinion that the patch is an | |
584 | appropriate modification of the kernel without any remaining serious | |
585 | technical issues. Any interested reviewer (who has done the work) can | |
586 | offer a Reviewed-by tag for a patch. This tag serves to give credit to | |
587 | reviewers and to inform maintainers of the degree of review which has been | |
588 | done on the patch. Reviewed-by: tags, when supplied by reviewers known to | |
589 | understand the subject area and to perform thorough reviews, will normally | |
5801da1b | 590 | increase the likelihood of your patch getting into the kernel. |
ef40203a | 591 | |
8543ae12 M |
592 | A Suggested-by: tag indicates that the patch idea is suggested by the person |
593 | named and ensures credit to the person for the idea. Please note that this | |
594 | tag should not be added without the reporter's permission, especially if the | |
595 | idea was not posted in a public forum. That said, if we diligently credit our | |
596 | idea reporters, they will, hopefully, be inspired to help us again in the | |
597 | future. | |
598 | ||
8401aa1f JK |
599 | A Fixes: tag indicates that the patch fixes an issue in a previous commit. It |
600 | is used to make it easy to determine where a bug originated, which can help | |
601 | review a bug fix. This tag also assists the stable kernel team in determining | |
602 | which stable kernel versions should receive your fix. This is the preferred | |
5903019b MCC |
603 | method for indicating a bug fixed by the patch. See :ref:`describe_changes` |
604 | for more details. | |
8401aa1f | 605 | |
ef40203a | 606 | |
ccae8616 | 607 | 14) The canonical patch format |
7994cc15 JC |
608 | ------------------------------ |
609 | ||
610 | This section describes how the patch itself should be formatted. Note | |
9b2c7677 | 611 | that, if you have your patches stored in a ``git`` repository, proper patch |
5903019b | 612 | formatting can be had with ``git format-patch``. The tools cannot create |
7994cc15 | 613 | the necessary text, though, so read the instructions below anyway. |
84da7c08 | 614 | |
5903019b | 615 | The canonical patch subject line is:: |
75f8426c | 616 | |
d6b9acc0 | 617 | Subject: [PATCH 001/123] subsystem: summary phrase |
75f8426c PJ |
618 | |
619 | The canonical patch message body contains the following: | |
620 | ||
5903019b | 621 | - A ``from`` line specifying the patch author (only needed if the person |
ccae8616 | 622 | sending the patch is not the author). |
75f8426c PJ |
623 | |
624 | - An empty line. | |
625 | ||
2a076f40 JP |
626 | - The body of the explanation, line wrapped at 75 columns, which will |
627 | be copied to the permanent changelog to describe this patch. | |
75f8426c | 628 | |
5903019b | 629 | - The ``Signed-off-by:`` lines, described above, which will |
75f8426c PJ |
630 | also go in the changelog. |
631 | ||
5903019b | 632 | - A marker line containing simply ``---``. |
75f8426c PJ |
633 | |
634 | - Any additional comments not suitable for the changelog. | |
635 | ||
9b2c7677 | 636 | - The actual patch (``diff`` output). |
75f8426c PJ |
637 | |
638 | The Subject line format makes it very easy to sort the emails | |
639 | alphabetically by subject line - pretty much any email reader will | |
640 | support that - since because the sequence number is zero-padded, | |
641 | the numerical and alphabetic sort is the same. | |
642 | ||
5903019b | 643 | The ``subsystem`` in the email's Subject should identify which |
d6b9acc0 PJ |
644 | area or subsystem of the kernel is being patched. |
645 | ||
5903019b MCC |
646 | The ``summary phrase`` in the email's Subject should concisely |
647 | describe the patch which that email contains. The ``summary | |
648 | phrase`` should not be a filename. Do not use the same ``summary | |
649 | phrase`` for every patch in a whole patch series (where a ``patch | |
650 | series`` is an ordered sequence of multiple, related patches). | |
d6b9acc0 | 651 | |
5903019b | 652 | Bear in mind that the ``summary phrase`` of your email becomes a |
2ae19aca | 653 | globally-unique identifier for that patch. It propagates all the way |
9b2c7677 | 654 | into the ``git`` changelog. The ``summary phrase`` may later be used in |
2ae19aca | 655 | developer discussions which refer to the patch. People will want to |
5903019b | 656 | google for the ``summary phrase`` to read discussion regarding that |
2ae19aca TT |
657 | patch. It will also be the only thing that people may quickly see |
658 | when, two or three months later, they are going through perhaps | |
9b2c7677 MCC |
659 | thousands of patches using tools such as ``gitk`` or ``git log |
660 | --oneline``. | |
2ae19aca | 661 | |
5903019b | 662 | For these reasons, the ``summary`` must be no more than 70-75 |
2ae19aca TT |
663 | characters, and it must describe both what the patch changes, as well |
664 | as why the patch might be necessary. It is challenging to be both | |
665 | succinct and descriptive, but that is what a well-written summary | |
666 | should do. | |
667 | ||
5903019b | 668 | The ``summary phrase`` may be prefixed by tags enclosed in square |
e12d7462 AH |
669 | brackets: "Subject: [PATCH <tag>...] <summary phrase>". The tags are |
670 | not considered part of the summary phrase, but describe how the patch | |
2ae19aca TT |
671 | should be treated. Common tags might include a version descriptor if |
672 | the multiple versions of the patch have been sent out in response to | |
673 | comments (i.e., "v1, v2, v3"), or "RFC" to indicate a request for | |
674 | comments. If there are four patches in a patch series the individual | |
675 | patches may be numbered like this: 1/4, 2/4, 3/4, 4/4. This assures | |
676 | that developers understand the order in which the patches should be | |
677 | applied and that they have reviewed or applied all of the patches in | |
678 | the patch series. | |
d6b9acc0 | 679 | |
5903019b | 680 | A couple of example Subjects:: |
d6b9acc0 | 681 | |
e12d7462 AH |
682 | Subject: [PATCH 2/5] ext2: improve scalability of bitmap searching |
683 | Subject: [PATCH v2 01/27] x86: fix eflags tracking | |
75f8426c | 684 | |
5903019b | 685 | The ``from`` line must be the very first line in the message body, |
75f8426c PJ |
686 | and has the form: |
687 | ||
688 | From: Original Author <author@example.com> | |
689 | ||
5903019b MCC |
690 | The ``from`` line specifies who will be credited as the author of the |
691 | patch in the permanent changelog. If the ``from`` line is missing, | |
692 | then the ``From:`` line from the email header will be used to determine | |
75f8426c PJ |
693 | the patch author in the changelog. |
694 | ||
695 | The explanation body will be committed to the permanent source | |
696 | changelog, so should make sense to a competent reader who has long | |
697 | since forgotten the immediate details of the discussion that might | |
2ae19aca TT |
698 | have led to this patch. Including symptoms of the failure which the |
699 | patch addresses (kernel log messages, oops messages, etc.) is | |
700 | especially useful for people who might be searching the commit logs | |
701 | looking for the applicable patch. If a patch fixes a compile failure, | |
702 | it may not be necessary to include _all_ of the compile failures; just | |
703 | enough that it is likely that someone searching for the patch can find | |
5903019b | 704 | it. As in the ``summary phrase``, it is important to be both succinct as |
2ae19aca | 705 | well as descriptive. |
75f8426c | 706 | |
5903019b | 707 | The ``---`` marker line serves the essential purpose of marking for patch |
75f8426c PJ |
708 | handling tools where the changelog message ends. |
709 | ||
5903019b | 710 | One good use for the additional comments after the ``---`` marker is for |
9b2c7677 MCC |
711 | a ``diffstat``, to show what files have changed, and the number of |
712 | inserted and deleted lines per file. A ``diffstat`` is especially useful | |
2ae19aca TT |
713 | on bigger patches. Other comments relevant only to the moment or the |
714 | maintainer, not suitable for the permanent changelog, should also go | |
5903019b | 715 | here. A good example of such comments might be ``patch changelogs`` |
2ae19aca TT |
716 | which describe what has changed between the v1 and v2 version of the |
717 | patch. | |
718 | ||
9b2c7677 MCC |
719 | If you are going to include a ``diffstat`` after the ``---`` marker, please |
720 | use ``diffstat`` options ``-p 1 -w 70`` so that filenames are listed from | |
2ae19aca | 721 | the top of the kernel source tree and don't use too much horizontal |
9b2c7677 | 722 | space (easily fit in 80 columns, maybe with some indentation). (``git`` |
8e3072a2 | 723 | generates appropriate diffstats by default.) |
75f8426c PJ |
724 | |
725 | See more details on the proper patch format in the following | |
726 | references. | |
727 | ||
5903019b MCC |
728 | .. _explicit_in_reply_to: |
729 | ||
d7ac8d85 CM |
730 | 15) Explicit In-Reply-To headers |
731 | -------------------------------- | |
732 | ||
733 | It can be helpful to manually add In-Reply-To: headers to a patch | |
5903019b | 734 | (e.g., when using ``git send-email``) to associate the patch with |
d7ac8d85 CM |
735 | previous relevant discussion, e.g. to link a bug fix to the email with |
736 | the bug report. However, for a multi-patch series, it is generally | |
737 | best to avoid using In-Reply-To: to link to older versions of the | |
738 | series. This way multiple versions of the patch don't become an | |
739 | unmanageable forest of references in email clients. If a link is | |
740 | helpful, you can use the https://lkml.kernel.org/ redirector (e.g., in | |
741 | the cover email text) to link to an earlier version of the patch series. | |
742 | ||
75f8426c | 743 | |
5903019b MCC |
744 | 16) Sending ``git pull`` requests |
745 | --------------------------------- | |
1da177e4 | 746 | |
7994cc15 JC |
747 | If you have a series of patches, it may be most convenient to have the |
748 | maintainer pull them directly into the subsystem repository with a | |
5903019b | 749 | ``git pull`` operation. Note, however, that pulling patches from a developer |
7994cc15 JC |
750 | requires a higher degree of trust than taking patches from a mailing list. |
751 | As a result, many subsystem maintainers are reluctant to take pull | |
b792ffe4 JC |
752 | requests, especially from new, unknown developers. If in doubt you can use |
753 | the pull request as the cover letter for a normal posting of the patch | |
754 | series, giving the maintainer the option of using either. | |
1da177e4 | 755 | |
7994cc15 JC |
756 | A pull request should have [GIT] or [PULL] in the subject line. The |
757 | request itself should include the repository name and the branch of | |
5903019b | 758 | interest on a single line; it should look something like:: |
1da177e4 | 759 | |
7994cc15 | 760 | Please pull from |
1da177e4 | 761 | |
7994cc15 | 762 | git://jdelvare.pck.nerim.net/jdelvare-2.6 i2c-for-linus |
1da177e4 | 763 | |
64e32895 | 764 | to get these changes: |
1da177e4 | 765 | |
7994cc15 | 766 | A pull request should also include an overall message saying what will be |
5903019b | 767 | included in the request, a ``git shortlog`` listing of the patches |
9b2c7677 | 768 | themselves, and a ``diffstat`` showing the overall effect of the patch series. |
7994cc15 | 769 | The easiest way to get all this information together is, of course, to let |
9b2c7677 | 770 | ``git`` do it for you with the ``git request-pull`` command. |
1da177e4 | 771 | |
7994cc15 JC |
772 | Some maintainers (including Linus) want to see pull requests from signed |
773 | commits; that increases their confidence that the request actually came | |
774 | from you. Linus, in particular, will not pull from public hosting sites | |
775 | like GitHub in the absence of a signed tag. | |
1da177e4 | 776 | |
7994cc15 JC |
777 | The first step toward creating such tags is to make a GNUPG key and get it |
778 | signed by one or more core kernel developers. This step can be hard for | |
779 | new developers, but there is no way around it. Attending conferences can | |
780 | be a good way to find developers who can sign your key. | |
1da177e4 | 781 | |
9b2c7677 | 782 | Once you have prepared a patch series in ``git`` that you wish to have somebody |
5903019b | 783 | pull, create a signed tag with ``git tag -s``. This will create a new tag |
7994cc15 JC |
784 | identifying the last commit in the series and containing a signature |
785 | created with your private key. You will also have the opportunity to add a | |
786 | changelog-style message to the tag; this is an ideal place to describe the | |
787 | effects of the pull request as a whole. | |
1da177e4 | 788 | |
7994cc15 JC |
789 | If the tree the maintainer will be pulling from is not the repository you |
790 | are working from, don't forget to push the signed tag explicitly to the | |
791 | public tree. | |
1da177e4 | 792 | |
7994cc15 | 793 | When generating your pull request, use the signed tag as the target. A |
5903019b | 794 | command like this will do the trick:: |
1da177e4 | 795 | |
7994cc15 | 796 | git request-pull master git://my.public.tree/linux.git my-signed-tag |
5b0ed2c6 XVP |
797 | |
798 | ||
5903019b MCC |
799 | REFERENCES |
800 | ********** | |
5b0ed2c6 XVP |
801 | |
802 | Andrew Morton, "The perfect patch" (tpp). | |
37c703f4 | 803 | <http://www.ozlabs.org/~akpm/stuff/tpp.txt> |
5b0ed2c6 | 804 | |
8e9cb8fd | 805 | Jeff Garzik, "Linux kernel patch submission format". |
5b0ed2c6 XVP |
806 | <http://linux.yyz.us/patch-format.html> |
807 | ||
8e9cb8fd | 808 | Greg Kroah-Hartman, "How to piss off a kernel subsystem maintainer". |
f5039935 | 809 | <http://www.kroah.com/log/linux/maintainer.html> |
9b2c7677 | 810 | |
f5039935 | 811 | <http://www.kroah.com/log/linux/maintainer-02.html> |
9b2c7677 | 812 | |
f5039935 | 813 | <http://www.kroah.com/log/linux/maintainer-03.html> |
9b2c7677 | 814 | |
f5039935 | 815 | <http://www.kroah.com/log/linux/maintainer-04.html> |
9b2c7677 | 816 | |
f5039935 | 817 | <http://www.kroah.com/log/linux/maintainer-05.html> |
9b2c7677 | 818 | |
7e0dae61 | 819 | <http://www.kroah.com/log/linux/maintainer-06.html> |
5b0ed2c6 | 820 | |
bc7455fa | 821 | NO!!!! No more huge patch bombs to linux-kernel@vger.kernel.org people! |
37c703f4 | 822 | <https://lkml.org/lkml/2005/7/11/336> |
5b0ed2c6 | 823 | |
8e9cb8fd | 824 | Kernel Documentation/CodingStyle: |
60498bb5 | 825 | <Documentation/CodingStyle> |
5b0ed2c6 | 826 | |
8e9cb8fd | 827 | Linus Torvalds's mail on the canonical patch format: |
5b0ed2c6 | 828 | <http://lkml.org/lkml/2005/4/7/183> |
9536727e AK |
829 | |
830 | Andi Kleen, "On submitting kernel patches" | |
25985edc | 831 | Some strategies to get difficult or controversial changes in. |
9b2c7677 | 832 | |
9536727e AK |
833 | http://halobates.de/on-submitting-patches.pdf |
834 |