blob: 5c555a8b39e553cf0c3117026ebf37d481c588b6 [file] [log] [blame]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001
2 How to Get Your Change Into the Linux Kernel
3 or
4 Care And Operation Of Your Linus Torvalds
5
6
7
8For a person or company who wishes to submit a change to the Linux
9kernel, the process can sometimes be daunting if you're not familiar
10with "the system." This text is a collection of suggestions which
11can greatly increase the chances of your change being accepted.
12
Randy Dunlapbc7455f2006-07-30 03:03:45 -070013Read Documentation/SubmitChecklist for a list of items to check
14before submitting code. If you are submitting a driver, also read
15Documentation/SubmittingDrivers.
Linus Torvalds1da177e2005-04-16 15:20:36 -070016
17
18
19--------------------------------------------
20SECTION 1 - CREATING AND SENDING YOUR CHANGE
21--------------------------------------------
22
23
24
251) "diff -up"
26------------
27
28Use "diff -up" or "diff -uprN" to create patches.
29
30All changes to the Linux kernel occur in the form of patches, as
31generated by diff(1). When creating your patch, make sure to create it
32in "unified diff" format, as supplied by the '-u' argument to diff(1).
33Also, please use the '-p' argument which shows which C function each
34change is in - that makes the resultant diff a lot easier to read.
35Patches should be based in the root kernel source directory,
36not in any lower subdirectory.
37
38To create a patch for a single file, it is often sufficient to do:
39
Randy Dunlap84da7c02005-06-28 20:45:30 -070040 SRCTREE= linux-2.6
Linus Torvalds1da177e2005-04-16 15:20:36 -070041 MYFILE= drivers/net/mydriver.c
42
43 cd $SRCTREE
44 cp $MYFILE $MYFILE.orig
45 vi $MYFILE # make your change
46 cd ..
47 diff -up $SRCTREE/$MYFILE{.orig,} > /tmp/patch
48
49To create a patch for multiple files, you should unpack a "vanilla",
50or unmodified kernel source tree, and generate a diff against your
51own source tree. For example:
52
Randy Dunlap84da7c02005-06-28 20:45:30 -070053 MYSRC= /devel/linux-2.6
Linus Torvalds1da177e2005-04-16 15:20:36 -070054
Randy Dunlap84da7c02005-06-28 20:45:30 -070055 tar xvfz linux-2.6.12.tar.gz
56 mv linux-2.6.12 linux-2.6.12-vanilla
57 diff -uprN -X linux-2.6.12-vanilla/Documentation/dontdiff \
58 linux-2.6.12-vanilla $MYSRC > /tmp/patch
Linus Torvalds1da177e2005-04-16 15:20:36 -070059
60"dontdiff" is a list of files which are generated by the kernel during
61the build process, and should be ignored in any diff(1)-generated
Randy Dunlap84da7c02005-06-28 20:45:30 -070062patch. The "dontdiff" file is included in the kernel tree in
632.6.12 and later. For earlier kernel versions, you can get it
64from <http://www.xenotime.net/linux/doc/dontdiff>.
Linus Torvalds1da177e2005-04-16 15:20:36 -070065
66Make sure your patch does not include any extra files which do not
67belong in a patch submission. Make sure to review your patch -after-
68generated it with diff(1), to ensure accuracy.
69
70If your changes produce a lot of deltas, you may want to look into
71splitting them into individual patches which modify things in
Randy Dunlap84da7c02005-06-28 20:45:30 -070072logical stages. This will facilitate easier reviewing by other
Linus Torvalds1da177e2005-04-16 15:20:36 -070073kernel developers, very important if you want your patch accepted.
Randy Dunlap84da7c02005-06-28 20:45:30 -070074There are a number of scripts which can aid in this:
Linus Torvalds1da177e2005-04-16 15:20:36 -070075
76Quilt:
77http://savannah.nongnu.org/projects/quilt
78
Linus Torvalds1da177e2005-04-16 15:20:36 -070079Andrew Morton's patch scripts:
FD Cami2223c652008-10-15 22:02:00 -070080http://userweb.kernel.org/~akpm/stuff/patch-scripts.tar.gz
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -080081Instead of these scripts, quilt is the recommended patch management
82tool (see above).
Randy Dunlap84da7c02005-06-28 20:45:30 -070083
84
Linus Torvalds1da177e2005-04-16 15:20:36 -070085
862) Describe your changes.
87
88Describe the technical detail of the change(s) your patch includes.
89
90Be as specific as possible. The WORST descriptions possible include
91things like "update driver X", "bug fix for driver X", or "this patch
92includes updates for subsystem X. Please apply."
93
Theodore Ts'o2ae19ac2009-04-16 07:44:45 -040094The maintainer will thank you if you write your patch description in a
95form which can be easily pulled into Linux's source code management
96system, git, as a "commit log". See #15, below.
97
Linus Torvalds1da177e2005-04-16 15:20:36 -070098If your description starts to get long, that's a sign that you probably
99need to split up your patch. See #3, next.
100
101
102
1033) Separate your changes.
104
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800105Separate _logical changes_ into a single patch file.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700106
107For example, if your changes include both bug fixes and performance
108enhancements for a single driver, separate those changes into two
109or more patches. If your changes include an API update, and a new
110driver which uses that new API, separate those into two patches.
111
112On the other hand, if you make a single change to numerous files,
113group those changes into a single patch. Thus a single logical change
114is contained within a single patch.
115
116If one patch depends on another patch in order for a change to be
117complete, that is OK. Simply note "this patch depends on patch X"
118in your patch description.
119
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800120If you cannot condense your patch set into a smaller set of patches,
121then only post say 15 or so at a time and wait for review and integration.
122
123
Linus Torvalds1da177e2005-04-16 15:20:36 -0700124
Andy Whitcroft0a920b52007-06-01 00:46:48 -07001254) Style check your changes.
126
127Check your patch for basic style violations, details of which can be
128found in Documentation/CodingStyle. Failure to do so simply wastes
Linus Nilssonf56d35e2007-07-21 17:49:06 +0200129the reviewers time and will get your patch rejected, probably
Andy Whitcroft0a920b52007-06-01 00:46:48 -0700130without even being read.
131
132At a minimum you should check your patches with the patch style
Andre Haupta570ab62007-09-11 15:23:47 -0700133checker prior to submission (scripts/checkpatch.pl). You should
Andy Whitcroft0a920b52007-06-01 00:46:48 -0700134be able to justify all violations that remain in your patch.
135
136
137
1385) Select e-mail destination.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700139
140Look through the MAINTAINERS file and the source code, and determine
141if your change applies to a specific subsystem of the kernel, with
142an assigned maintainer. If so, e-mail that person.
143
144If no maintainer is listed, or the maintainer does not respond, send
145your patch to the primary Linux kernel developer's mailing list,
146linux-kernel@vger.kernel.org. Most kernel developers monitor this
147e-mail list, and can comment on your changes.
148
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800149
150Do not send more than 15 patches at once to the vger mailing lists!!!
151
152
Linus Torvalds1da177e2005-04-16 15:20:36 -0700153Linus Torvalds is the final arbiter of all changes accepted into the
Linus Torvalds99ddcc72007-01-23 14:22:35 -0800154Linux kernel. His e-mail address is <torvalds@linux-foundation.org>.
155He gets a lot of e-mail, so typically you should do your best to -avoid-
156sending him e-mail.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700157
158Patches which are bug fixes, are "obvious" changes, or similarly
159require little discussion should be sent or CC'd to Linus. Patches
160which require discussion or do not have a clear advantage should
161usually be sent first to linux-kernel. Only after the patch is
162discussed should the patch then be submitted to Linus.
163
Linus Torvalds1da177e2005-04-16 15:20:36 -0700164
165
Andy Whitcroft0a920b52007-06-01 00:46:48 -07001666) Select your CC (e-mail carbon copy) list.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700167
168Unless you have a reason NOT to do so, CC linux-kernel@vger.kernel.org.
169
170Other kernel developers besides Linus need to be aware of your change,
171so that they may comment on it and offer code review and suggestions.
172linux-kernel is the primary Linux kernel developer mailing list.
173Other mailing lists are available for specific subsystems, such as
174USB, framebuffer devices, the VFS, the SCSI subsystem, etc. See the
175MAINTAINERS file for a mailing list that relates specifically to
176your change.
177
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800178Majordomo lists of VGER.KERNEL.ORG at:
179 <http://vger.kernel.org/vger-lists.html>
180
Paul Jackson1caf1f02005-07-31 22:34:48 -0700181If changes affect userland-kernel interfaces, please send
182the MAN-PAGES maintainer (as listed in the MAINTAINERS file)
183a man-pages patch, or at least a notification of the change,
184so that some information makes its way into the manual pages.
185
Linus Torvalds1da177e2005-04-16 15:20:36 -0700186Even if the maintainer did not respond in step #4, make sure to ALWAYS
187copy the maintainer when you change their code.
188
189For small patches you may want to CC the Trivial Patch Monkey
Markus Heidelberg82d27b22009-06-12 01:02:34 +0200190trivial@kernel.org which collects "trivial" patches. Have a look
191into the MAINTAINERS file for its current manager.
192Trivial patches must qualify for one of the following rules:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700193 Spelling fixes in documentation
Pavel Machek8e9cb8f2006-09-29 02:01:29 -0700194 Spelling fixes which could break grep(1)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700195 Warning fixes (cluttering with useless warnings is bad)
196 Compilation fixes (only if they are actually correct)
197 Runtime fixes (only if they actually fix things)
Pavel Machek8e9cb8f2006-09-29 02:01:29 -0700198 Removing use of deprecated functions/macros (eg. check_region)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700199 Contact detail and documentation fixes
200 Non-portable code replaced by portable code (even in arch-specific,
201 since people copy, as long as it's trivial)
Pavel Machek8e9cb8f2006-09-29 02:01:29 -0700202 Any fix by the author/maintainer of the file (ie. patch monkey
Linus Torvalds1da177e2005-04-16 15:20:36 -0700203 in re-transmission mode)
Randy Dunlap84da7c02005-06-28 20:45:30 -0700204
Linus Torvalds1da177e2005-04-16 15:20:36 -0700205
206
Andy Whitcroft0a920b52007-06-01 00:46:48 -07002077) No MIME, no links, no compression, no attachments. Just plain text.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700208
209Linus and other kernel developers need to be able to read and comment
210on the changes you are submitting. It is important for a kernel
211developer to be able to "quote" your changes, using standard e-mail
212tools, so that they may comment on specific portions of your code.
213
214For this reason, all patches should be submitting e-mail "inline".
215WARNING: Be wary of your editor's word-wrap corrupting your patch,
216if you choose to cut-n-paste your patch.
217
218Do not attach the patch as a MIME attachment, compressed or not.
219Many popular e-mail applications will not always transmit a MIME
220attachment as plain text, making it impossible to comment on your
221code. A MIME attachment also takes Linus a bit more time to process,
222decreasing the likelihood of your MIME-attached change being accepted.
223
224Exception: If your mailer is mangling patches then someone may ask
225you to re-send them using MIME.
226
Michael Opdenacker097091c2008-02-03 18:06:58 +0200227See Documentation/email-clients.txt for hints about configuring
228your e-mail client so that it sends your patches untouched.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700229
Andy Whitcroft0a920b52007-06-01 00:46:48 -07002308) E-mail size.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700231
Andy Whitcroft0a920b52007-06-01 00:46:48 -0700232When sending patches to Linus, always follow step #7.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700233
234Large changes are not appropriate for mailing lists, and some
235maintainers. If your patch, uncompressed, exceeds 40 kB in size,
236it is preferred that you store your patch on an Internet-accessible
237server, and provide instead a URL (link) pointing to your patch.
238
239
240
Andy Whitcroft0a920b52007-06-01 00:46:48 -07002419) Name your kernel version.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700242
243It is important to note, either in the subject line or in the patch
244description, the kernel version to which this patch applies.
245
246If the patch does not apply cleanly to the latest kernel version,
247Linus will not apply it.
248
249
250
Andy Whitcroft0a920b52007-06-01 00:46:48 -070025110) Don't get discouraged. Re-submit.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700252
253After you have submitted your change, be patient and wait. If Linus
254likes your change and applies it, it will appear in the next version
255of the kernel that he releases.
256
257However, if your change doesn't appear in the next version of the
258kernel, there could be any number of reasons. It's YOUR job to
259narrow down those reasons, correct what was wrong, and submit your
260updated change.
261
262It is quite common for Linus to "drop" your patch without comment.
263That's the nature of the system. If he drops your patch, it could be
264due to
Pavel Machek8e9cb8f2006-09-29 02:01:29 -0700265* Your patch did not apply cleanly to the latest kernel version.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700266* Your patch was not sufficiently discussed on linux-kernel.
Pavel Machek8e9cb8f2006-09-29 02:01:29 -0700267* A style issue (see section 2).
268* An e-mail formatting issue (re-read this section).
269* A technical problem with your change.
270* He gets tons of e-mail, and yours got lost in the shuffle.
271* You are being annoying.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700272
273When in doubt, solicit comments on linux-kernel mailing list.
274
275
276
Andy Whitcroft0a920b52007-06-01 00:46:48 -070027711) Include PATCH in the subject
Linus Torvalds1da177e2005-04-16 15:20:36 -0700278
279Due to high e-mail traffic to Linus, and to linux-kernel, it is common
280convention to prefix your subject line with [PATCH]. This lets Linus
281and other kernel developers more easily distinguish patches from other
282e-mail discussions.
283
284
285
Andy Whitcroft0a920b52007-06-01 00:46:48 -070028612) Sign your work
Linus Torvalds1da177e2005-04-16 15:20:36 -0700287
288To improve tracking of who did what, especially with patches that can
289percolate to their final resting place in the kernel through several
290layers of maintainers, we've introduced a "sign-off" procedure on
291patches that are being emailed around.
292
293The sign-off is a simple line at the end of the explanation for the
294patch, which certifies that you wrote it or otherwise have the right to
295pass it on as a open-source patch. The rules are pretty simple: if you
296can certify the below:
297
Linus Torvaldscbd83da2005-06-13 17:51:55 -0700298 Developer's Certificate of Origin 1.1
Linus Torvalds1da177e2005-04-16 15:20:36 -0700299
300 By making a contribution to this project, I certify that:
301
302 (a) The contribution was created in whole or in part by me and I
303 have the right to submit it under the open source license
304 indicated in the file; or
305
306 (b) The contribution is based upon previous work that, to the best
307 of my knowledge, is covered under an appropriate open source
308 license and I have the right under that license to submit that
309 work with modifications, whether created in whole or in part
310 by me, under the same open source license (unless I am
311 permitted to submit under a different license), as indicated
312 in the file; or
313
314 (c) The contribution was provided directly to me by some other
315 person who certified (a), (b) or (c) and I have not modified
316 it.
317
Linus Torvaldscbd83da2005-06-13 17:51:55 -0700318 (d) I understand and agree that this project and the contribution
319 are public and that a record of the contribution (including all
320 personal information I submit with it, including my sign-off) is
321 maintained indefinitely and may be redistributed consistent with
322 this project or the open source license(s) involved.
323
Linus Torvalds1da177e2005-04-16 15:20:36 -0700324then you just add a line saying
325
Alexey Dobriyan9fd55592005-06-25 14:59:34 -0700326 Signed-off-by: Random J Developer <random@developer.example.org>
Linus Torvalds1da177e2005-04-16 15:20:36 -0700327
Greg KHaf45f322006-09-12 20:35:52 -0700328using your real name (sorry, no pseudonyms or anonymous contributions.)
329
Linus Torvalds1da177e2005-04-16 15:20:36 -0700330Some people also put extra tags at the end. They'll just be ignored for
331now, but you can do this to mark internal company procedures or just
332point out some special detail about the sign-off.
333
Willy Tarreauadbd5882008-06-03 00:20:28 +0200334If you are a subsystem or branch maintainer, sometimes you need to slightly
335modify patches you receive in order to merge them, because the code is not
336exactly the same in your tree and the submitters'. If you stick strictly to
337rule (c), you should ask the submitter to rediff, but this is a totally
338counter-productive waste of time and energy. Rule (b) allows you to adjust
339the code, but then it is very impolite to change one submitter's code and
340make him endorse your bugs. To solve this problem, it is recommended that
341you add a line between the last Signed-off-by header and yours, indicating
342the nature of your changes. While there is nothing mandatory about this, it
343seems like prepending the description with your mail and/or name, all
344enclosed in square brackets, is noticeable enough to make it obvious that
345you are responsible for last-minute changes. Example :
346
347 Signed-off-by: Random J Developer <random@developer.example.org>
348 [lucky@maintainer.example.org: struct foo moved from foo.c to foo.h]
349 Signed-off-by: Lucky K Maintainer <lucky@maintainer.example.org>
350
351This practise is particularly helpful if you maintain a stable branch and
352want at the same time to credit the author, track changes, merge the fix,
353and protect the submitter from complaints. Note that under no circumstances
354can you change the author's identity (the From header), as it is the one
355which appears in the changelog.
356
357Special note to back-porters: It seems to be a common and useful practise
358to insert an indication of the origin of a patch at the top of the commit
359message (just after the subject line) to facilitate tracking. For instance,
360here's what we see in 2.6-stable :
361
362 Date: Tue May 13 19:10:30 2008 +0000
363
364 SCSI: libiscsi regression in 2.6.25: fix nop timer handling
365
366 commit 4cf1043593db6a337f10e006c23c69e5fc93e722 upstream
367
368And here's what appears in 2.4 :
369
370 Date: Tue May 13 22:12:27 2008 +0200
371
372 wireless, airo: waitbusy() won't delay
373
374 [backport of 2.6 commit b7acbdfbd1f277c1eb23f344f899cfa4cd0bf36a]
375
376Whatever the format, this information provides a valuable help to people
377tracking your trees, and to people trying to trouble-shoot bugs in your
378tree.
379
Linus Torvalds1da177e2005-04-16 15:20:36 -0700380
Jonathan Corbetef402032008-03-28 11:22:38 -060038113) When to use Acked-by: and Cc:
Andy Whitcroft0a920b52007-06-01 00:46:48 -0700382
Andrew Morton0f44cd22007-06-08 13:46:45 -0700383The Signed-off-by: tag indicates that the signer was involved in the
384development of the patch, or that he/she was in the patch's delivery path.
385
386If a person was not directly involved in the preparation or handling of a
387patch but wishes to signify and record their approval of it then they can
388arrange to have an Acked-by: line added to the patch's changelog.
389
390Acked-by: is often used by the maintainer of the affected code when that
391maintainer neither contributed to nor forwarded the patch.
392
393Acked-by: is not as formal as Signed-off-by:. It is a record that the acker
394has at least reviewed the patch and has indicated acceptance. Hence patch
395mergers will sometimes manually convert an acker's "yep, looks good to me"
396into an Acked-by:.
397
398Acked-by: does not necessarily indicate acknowledgement of the entire patch.
399For example, if a patch affects multiple subsystems and has an Acked-by: from
400one subsystem maintainer then this usually indicates acknowledgement of just
401the part which affects that maintainer's code. Judgement should be used here.
Jonathan Corbetef402032008-03-28 11:22:38 -0600402When in doubt people should refer to the original discussion in the mailing
Andrew Morton0f44cd22007-06-08 13:46:45 -0700403list archives.
404
Jonathan Corbetef402032008-03-28 11:22:38 -0600405If a person has had the opportunity to comment on a patch, but has not
406provided such comments, you may optionally add a "Cc:" tag to the patch.
407This is the only tag which might be added without an explicit action by the
408person it names. This tag documents that potentially interested parties
409have been included in the discussion
Andrew Morton0f44cd22007-06-08 13:46:45 -0700410
Jonathan Corbetef402032008-03-28 11:22:38 -0600411
Jonathan Corbetbbb0a422009-01-16 09:49:50 -070041214) Using Reported-by:, Tested-by: and Reviewed-by:
413
414If this patch fixes a problem reported by somebody else, consider adding a
415Reported-by: tag to credit the reporter for their contribution. Please
416note that this tag should not be added without the reporter's permission,
417especially if the problem was not reported in a public forum. That said,
418if we diligently credit our bug reporters, they will, hopefully, be
419inspired to help us again in the future.
Jonathan Corbetef402032008-03-28 11:22:38 -0600420
421A Tested-by: tag indicates that the patch has been successfully tested (in
422some environment) by the person named. This tag informs maintainers that
423some testing has been performed, provides a means to locate testers for
424future patches, and ensures credit for the testers.
425
426Reviewed-by:, instead, indicates that the patch has been reviewed and found
427acceptable according to the Reviewer's Statement:
428
429 Reviewer's statement of oversight
430
431 By offering my Reviewed-by: tag, I state that:
432
433 (a) I have carried out a technical review of this patch to
434 evaluate its appropriateness and readiness for inclusion into
435 the mainline kernel.
436
437 (b) Any problems, concerns, or questions relating to the patch
438 have been communicated back to the submitter. I am satisfied
439 with the submitter's response to my comments.
440
441 (c) While there may be things that could be improved with this
442 submission, I believe that it is, at this time, (1) a
443 worthwhile modification to the kernel, and (2) free of known
444 issues which would argue against its inclusion.
445
446 (d) While I have reviewed the patch and believe it to be sound, I
447 do not (unless explicitly stated elsewhere) make any
448 warranties or guarantees that it will achieve its stated
449 purpose or function properly in any given situation.
450
451A Reviewed-by tag is a statement of opinion that the patch is an
452appropriate modification of the kernel without any remaining serious
453technical issues. Any interested reviewer (who has done the work) can
454offer a Reviewed-by tag for a patch. This tag serves to give credit to
455reviewers and to inform maintainers of the degree of review which has been
456done on the patch. Reviewed-by: tags, when supplied by reviewers known to
457understand the subject area and to perform thorough reviews, will normally
Pavel Machek5801da12009-06-04 16:26:50 +0200458increase the likelihood of your patch getting into the kernel.
Jonathan Corbetef402032008-03-28 11:22:38 -0600459
460
46115) The canonical patch format
Randy Dunlap84da7c02005-06-28 20:45:30 -0700462
Paul Jackson75f84262005-10-02 18:01:42 -0700463The canonical patch subject line is:
464
Paul Jacksond6b9acc2005-10-03 00:29:10 -0700465 Subject: [PATCH 001/123] subsystem: summary phrase
Paul Jackson75f84262005-10-02 18:01:42 -0700466
467The canonical patch message body contains the following:
468
469 - A "from" line specifying the patch author.
470
471 - An empty line.
472
473 - The body of the explanation, which will be copied to the
474 permanent changelog to describe this patch.
475
476 - The "Signed-off-by:" lines, described above, which will
477 also go in the changelog.
478
479 - A marker line containing simply "---".
480
481 - Any additional comments not suitable for the changelog.
482
483 - The actual patch (diff output).
484
485The Subject line format makes it very easy to sort the emails
486alphabetically by subject line - pretty much any email reader will
487support that - since because the sequence number is zero-padded,
488the numerical and alphabetic sort is the same.
489
Paul Jacksond6b9acc2005-10-03 00:29:10 -0700490The "subsystem" in the email's Subject should identify which
491area or subsystem of the kernel is being patched.
492
493The "summary phrase" in the email's Subject should concisely
494describe the patch which that email contains. The "summary
495phrase" should not be a filename. Do not use the same "summary
Randy Dunlap66effdc2007-05-09 02:33:42 -0700496phrase" for every patch in a whole patch series (where a "patch
497series" is an ordered sequence of multiple, related patches).
Paul Jacksond6b9acc2005-10-03 00:29:10 -0700498
Theodore Ts'o2ae19ac2009-04-16 07:44:45 -0400499Bear in mind that the "summary phrase" of your email becomes a
500globally-unique identifier for that patch. It propagates all the way
501into the git changelog. The "summary phrase" may later be used in
502developer discussions which refer to the patch. People will want to
503google for the "summary phrase" to read discussion regarding that
504patch. It will also be the only thing that people may quickly see
505when, two or three months later, they are going through perhaps
506thousands of patches using tools such as "gitk" or "git log
507--oneline".
508
509For these reasons, the "summary" must be no more than 70-75
510characters, and it must describe both what the patch changes, as well
511as why the patch might be necessary. It is challenging to be both
512succinct and descriptive, but that is what a well-written summary
513should do.
514
515The "summary phrase" may be prefixed by tags enclosed in square
516brackets: "Subject: [PATCH tag] <summary phrase>". The tags are not
517considered part of the summary phrase, but describe how the patch
518should be treated. Common tags might include a version descriptor if
519the multiple versions of the patch have been sent out in response to
520comments (i.e., "v1, v2, v3"), or "RFC" to indicate a request for
521comments. If there are four patches in a patch series the individual
522patches may be numbered like this: 1/4, 2/4, 3/4, 4/4. This assures
523that developers understand the order in which the patches should be
524applied and that they have reviewed or applied all of the patches in
525the patch series.
Paul Jacksond6b9acc2005-10-03 00:29:10 -0700526
527A couple of example Subjects:
528
529 Subject: [patch 2/5] ext2: improve scalability of bitmap searching
530 Subject: [PATCHv2 001/207] x86: fix eflags tracking
Paul Jackson75f84262005-10-02 18:01:42 -0700531
532The "from" line must be the very first line in the message body,
533and has the form:
534
535 From: Original Author <author@example.com>
536
537The "from" line specifies who will be credited as the author of the
538patch in the permanent changelog. If the "from" line is missing,
539then the "From:" line from the email header will be used to determine
540the patch author in the changelog.
541
542The explanation body will be committed to the permanent source
543changelog, so should make sense to a competent reader who has long
544since forgotten the immediate details of the discussion that might
Theodore Ts'o2ae19ac2009-04-16 07:44:45 -0400545have led to this patch. Including symptoms of the failure which the
546patch addresses (kernel log messages, oops messages, etc.) is
547especially useful for people who might be searching the commit logs
548looking for the applicable patch. If a patch fixes a compile failure,
549it may not be necessary to include _all_ of the compile failures; just
550enough that it is likely that someone searching for the patch can find
551it. As in the "summary phrase", it is important to be both succinct as
552well as descriptive.
Paul Jackson75f84262005-10-02 18:01:42 -0700553
554The "---" marker line serves the essential purpose of marking for patch
555handling tools where the changelog message ends.
556
557One good use for the additional comments after the "---" marker is for
Theodore Ts'o2ae19ac2009-04-16 07:44:45 -0400558a diffstat, to show what files have changed, and the number of
559inserted and deleted lines per file. A diffstat is especially useful
560on bigger patches. Other comments relevant only to the moment or the
561maintainer, not suitable for the permanent changelog, should also go
562here. A good example of such comments might be "patch changelogs"
563which describe what has changed between the v1 and v2 version of the
564patch.
565
566If you are going to include a diffstat after the "---" marker, please
567use diffstat options "-p 1 -w 70" so that filenames are listed from
568the top of the kernel source tree and don't use too much horizontal
569space (easily fit in 80 columns, maybe with some indentation).
Paul Jackson75f84262005-10-02 18:01:42 -0700570
571See more details on the proper patch format in the following
572references.
573
574
Randy Dunlap14863612008-07-27 20:44:24 -070057516) Sending "git pull" requests (from Linus emails)
Randy Dunlap84da7c02005-06-28 20:45:30 -0700576
Randy Dunlap14863612008-07-27 20:44:24 -0700577Please write the git repo address and branch name alone on the same line
578so that I can't even by mistake pull from the wrong branch, and so
579that a triple-click just selects the whole thing.
580
581So the proper format is something along the lines of:
582
583 "Please pull from
584
585 git://jdelvare.pck.nerim.net/jdelvare-2.6 i2c-for-linus
586
587 to get these changes:"
588
589so that I don't have to hunt-and-peck for the address and inevitably
590get it wrong (actually, I've only gotten it wrong a few times, and
591checking against the diffstat tells me when I get it wrong, but I'm
592just a lot more comfortable when I don't have to "look for" the right
593thing to pull, and double-check that I have the right branch-name).
594
595
596Please use "git diff -M --stat --summary" to generate the diffstat:
597the -M enables rename detection, and the summary enables a summary of
598new/deleted or renamed files.
599
600With rename detection, the statistics are rather different [...]
601because git will notice that a fair number of the changes are renames.
Randy Dunlap84da7c02005-06-28 20:45:30 -0700602
Linus Torvalds1da177e2005-04-16 15:20:36 -0700603-----------------------------------
604SECTION 2 - HINTS, TIPS, AND TRICKS
605-----------------------------------
606
607This section lists many of the common "rules" associated with code
608submitted to the kernel. There are always exceptions... but you must
609have a really good reason for doing so. You could probably call this
610section Linus Computer Science 101.
611
612
613
6141) Read Documentation/CodingStyle
615
616Nuff said. If your code deviates too much from this, it is likely
617to be rejected without further review, and without comment.
618
Keiichi Kii5ab3bd52007-10-26 15:51:44 +0900619One significant exception is when moving code from one file to
620another -- in this case you should not modify the moved code at all in
Andy Whitcroftde7d4f02007-07-15 23:37:22 -0700621the same patch which moves it. This clearly delineates the act of
622moving the code and your changes. This greatly aids review of the
623actual differences and allows tools to better track the history of
624the code itself.
625
Andy Whitcroft0a920b52007-06-01 00:46:48 -0700626Check your patches with the patch style checker prior to submission
Andy Whitcroftde7d4f02007-07-15 23:37:22 -0700627(scripts/checkpatch.pl). The style checker should be viewed as
628a guide not as the final word. If your code looks better with
629a violation then its probably best left alone.
630
631The checker reports at three levels:
632 - ERROR: things that are very likely to be wrong
633 - WARNING: things requiring careful review
634 - CHECK: things requiring thought
635
636You should be able to justify all violations that remain in your
637patch.
Andy Whitcroft0a920b52007-06-01 00:46:48 -0700638
Linus Torvalds1da177e2005-04-16 15:20:36 -0700639
640
6412) #ifdefs are ugly
642
643Code cluttered with ifdefs is difficult to read and maintain. Don't do
644it. Instead, put your ifdefs in a header, and conditionally define
645'static inline' functions, or macros, which are used in the code.
646Let the compiler optimize away the "no-op" case.
647
648Simple example, of poor code:
649
650 dev = alloc_etherdev (sizeof(struct funky_private));
651 if (!dev)
652 return -ENODEV;
653 #ifdef CONFIG_NET_FUNKINESS
654 init_funky_net(dev);
655 #endif
656
657Cleaned-up example:
658
659(in header)
660 #ifndef CONFIG_NET_FUNKINESS
661 static inline void init_funky_net (struct net_device *d) {}
662 #endif
663
664(in the code itself)
665 dev = alloc_etherdev (sizeof(struct funky_private));
666 if (!dev)
667 return -ENODEV;
668 init_funky_net(dev);
669
670
671
6723) 'static inline' is better than a macro
673
674Static inline functions are greatly preferred over macros.
675They provide type safety, have no length limitations, no formatting
676limitations, and under gcc they are as cheap as macros.
677
678Macros should only be used for cases where a static inline is clearly
Jim Meyeringf2b2ea62008-04-02 13:04:46 -0700679suboptimal [there are a few, isolated cases of this in fast paths],
Linus Torvalds1da177e2005-04-16 15:20:36 -0700680or where it is impossible to use a static inline function [such as
681string-izing].
682
683'static inline' is preferred over 'static __inline__', 'extern inline',
684and 'extern __inline__'.
685
686
687
6884) Don't over-design.
689
690Don't try to anticipate nebulous future cases which may or may not
Randy Dunlap84da7c02005-06-28 20:45:30 -0700691be useful: "Make it as simple as you can, and no simpler."
Linus Torvalds1da177e2005-04-16 15:20:36 -0700692
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800693
694
695----------------------
696SECTION 3 - REFERENCES
697----------------------
698
699Andrew Morton, "The perfect patch" (tpp).
FD Cami2223c652008-10-15 22:02:00 -0700700 <http://userweb.kernel.org/~akpm/stuff/tpp.txt>
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800701
Pavel Machek8e9cb8f2006-09-29 02:01:29 -0700702Jeff Garzik, "Linux kernel patch submission format".
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800703 <http://linux.yyz.us/patch-format.html>
704
Pavel Machek8e9cb8f2006-09-29 02:01:29 -0700705Greg Kroah-Hartman, "How to piss off a kernel subsystem maintainer".
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800706 <http://www.kroah.com/log/2005/03/31/>
707 <http://www.kroah.com/log/2005/07/08/>
708 <http://www.kroah.com/log/2005/10/19/>
Xose Vazquez Pereze1b114e2006-01-14 12:57:41 +0100709 <http://www.kroah.com/log/2006/01/11/>
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800710
Randy Dunlapbc7455f2006-07-30 03:03:45 -0700711NO!!!! No more huge patch bombs to linux-kernel@vger.kernel.org people!
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800712 <http://marc.theaimsgroup.com/?l=linux-kernel&m=112112749912944&w=2>
713
Pavel Machek8e9cb8f2006-09-29 02:01:29 -0700714Kernel Documentation/CodingStyle:
Qi Yong4db29c12007-06-12 13:06:49 +0800715 <http://users.sosdg.org/~qiyong/lxr/source/Documentation/CodingStyle>
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800716
Pavel Machek8e9cb8f2006-09-29 02:01:29 -0700717Linus Torvalds's mail on the canonical patch format:
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800718 <http://lkml.org/lkml/2005/4/7/183>
Andi Kleen95367272008-10-15 22:02:02 -0700719
720Andi Kleen, "On submitting kernel patches"
721 Some strategies to get difficult or controversal changes in.
722 http://halobates.de/on-submitting-patches.pdf
723
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800724--