Version:  2.0.40 2.2.26 2.4.37 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12 3.13 3.14 3.15 3.16 3.17 3.18 3.19


  2         How to Get Your Change Into the Linux Kernel
  3                 or
  4         Care And Operation Of Your Linus Torvalds
  8 For a person or company who wishes to submit a change to the Linux
  9 kernel, the process can sometimes be daunting if you're not familiar
 10 with "the system."  This text is a collection of suggestions which
 11 can greatly increase the chances of your change being accepted.
 13 Read Documentation/SubmitChecklist for a list of items to check
 14 before submitting code.  If you are submitting a driver, also read
 15 Documentation/SubmittingDrivers.
 17 Many of these steps describe the default behavior of the git version
 18 control system; if you use git to prepare your patches, you'll find much
 19 of the mechanical work done for you, though you'll still need to prepare
 20 and document a sensible set of patches.
 22 --------------------------------------------
 24 --------------------------------------------
 28 1) "diff -up"
 29 ------------
 31 Use "diff -up" or "diff -uprN" to create patches.  git generates patches
 32 in this form by default; if you're using git, you can skip this section
 33 entirely.
 35 All changes to the Linux kernel occur in the form of patches, as
 36 generated by diff(1).  When creating your patch, make sure to create it
 37 in "unified diff" format, as supplied by the '-u' argument to diff(1).
 38 Also, please use the '-p' argument which shows which C function each
 39 change is in - that makes the resultant diff a lot easier to read.
 40 Patches should be based in the root kernel source directory,
 41 not in any lower subdirectory.
 43 To create a patch for a single file, it is often sufficient to do:
 45         SRCTREE= linux-2.6
 46         MYFILE=  drivers/net/mydriver.c
 48         cd $SRCTREE
 49         cp $MYFILE $MYFILE.orig
 50         vi $MYFILE      # make your change
 51         cd ..
 52         diff -up $SRCTREE/$MYFILE{.orig,} > /tmp/patch
 54 To create a patch for multiple files, you should unpack a "vanilla",
 55 or unmodified kernel source tree, and generate a diff against your
 56 own source tree.  For example:
 58         MYSRC= /devel/linux-2.6
 60         tar xvfz linux-2.6.12.tar.gz
 61         mv linux-2.6.12 linux-2.6.12-vanilla
 62         diff -uprN -X linux-2.6.12-vanilla/Documentation/dontdiff \
 63                 linux-2.6.12-vanilla $MYSRC > /tmp/patch
 65 "dontdiff" is a list of files which are generated by the kernel during
 66 the build process, and should be ignored in any diff(1)-generated
 67 patch.  The "dontdiff" file is included in the kernel tree in
 68 2.6.12 and later.
 70 Make sure your patch does not include any extra files which do not
 71 belong in a patch submission.  Make sure to review your patch -after-
 72 generated it with diff(1), to ensure accuracy.
 74 If your changes produce a lot of deltas, you need to split them into
 75 individual patches which modify things in logical stages; see section
 76 #3.  This will facilitate easier reviewing by other kernel developers,
 77 very important if you want your patch accepted.
 79 If you're using git, "git rebase -i" can help you with this process.  If
 80 you're not using git, quilt <>
 81 is another popular alternative.
 85 2) Describe your changes.
 87 Describe your problem.  Whether your patch is a one-line bug fix or
 88 5000 lines of a new feature, there must be an underlying problem that
 89 motivated you to do this work.  Convince the reviewer that there is a
 90 problem worth fixing and that it makes sense for them to read past the
 91 first paragraph.
 93 Describe user-visible impact.  Straight up crashes and lockups are
 94 pretty convincing, but not all bugs are that blatant.  Even if the
 95 problem was spotted during code review, describe the impact you think
 96 it can have on users.  Keep in mind that the majority of Linux
 97 installations run kernels from secondary stable trees or
 98 vendor/product-specific trees that cherry-pick only specific patches
 99 from upstream, so include anything that could help route your change
100 downstream: provoking circumstances, excerpts from dmesg, crash
101 descriptions, performance regressions, latency spikes, lockups, etc.
103 Quantify optimizations and trade-offs.  If you claim improvements in
104 performance, memory consumption, stack footprint, or binary size,
105 include numbers that back them up.  But also describe non-obvious
106 costs.  Optimizations usually aren't free but trade-offs between CPU,
107 memory, and readability; or, when it comes to heuristics, between
108 different workloads.  Describe the expected downsides of your
109 optimization so that the reviewer can weigh costs against benefits.
111 Once the problem is established, describe what you are actually doing
112 about it in technical detail.  It's important to describe the change
113 in plain English for the reviewer to verify that the code is behaving
114 as you intend it to.
116 The maintainer will thank you if you write your patch description in a
117 form which can be easily pulled into Linux's source code management
118 system, git, as a "commit log".  See #15, below.
120 Solve only one problem per patch.  If your description starts to get
121 long, that's a sign that you probably need to split up your patch.
122 See #3, next.
124 When you submit or resubmit a patch or patch series, include the
125 complete patch description and justification for it.  Don't just
126 say that this is version N of the patch (series).  Don't expect the
127 patch merger to refer back to earlier patch versions or referenced
128 URLs to find the patch description and put that into the patch.
129 I.e., the patch (series) and its description should be self-contained.
130 This benefits both the patch merger(s) and reviewers.  Some reviewers
131 probably didn't even receive earlier versions of the patch.
133 Describe your changes in imperative mood, e.g. "make xyzzy do frotz"
134 instead of "[This patch] makes xyzzy do frotz" or "[I] changed xyzzy
135 to do frotz", as if you are giving orders to the codebase to change
136 its behaviour.
138 If the patch fixes a logged bug entry, refer to that bug entry by
139 number and URL.  If the patch follows from a mailing list discussion,
140 give a URL to the mailing list archive; use the
141 redirector with a Message-Id, to ensure that the links cannot become
142 stale.
144 However, try to make your explanation understandable without external
145 resources.  In addition to giving a URL to a mailing list archive or
146 bug, summarize the relevant points of the discussion that led to the
147 patch as submitted.
149 If you want to refer to a specific commit, don't just refer to the
150 SHA-1 ID of the commit. Please also include the oneline summary of
151 the commit, to make it easier for reviewers to know what it is about.
152 Example:
154         Commit e21d2170f36602ae2708 ("video: remove unnecessary
155         platform_set_drvdata()") removed the unnecessary
156         platform_set_drvdata(), but left the variable "dev" unused,
157         delete it.
159 If your patch fixes a bug in a specific commit, e.g. you found an issue using
160 git-bisect, please use the 'Fixes:' tag with the first 12 characters of the
161 SHA-1 ID, and the one line summary.
162 Example:
164         Fixes: e21d2170f366 ("video: remove unnecessary platform_set_drvdata()")
166 The following git-config settings can be used to add a pretty format for
167 outputting the above style in the git log or git show commands
169         [core]
170                 abbrev = 12
171         [pretty]
172                 fixes = Fixes: %h (\"%s\")
174 3) Separate your changes.
176 Separate _logical changes_ into a single patch file.
178 For example, if your changes include both bug fixes and performance
179 enhancements for a single driver, separate those changes into two
180 or more patches.  If your changes include an API update, and a new
181 driver which uses that new API, separate those into two patches.
183 On the other hand, if you make a single change to numerous files,
184 group those changes into a single patch.  Thus a single logical change
185 is contained within a single patch.
187 If one patch depends on another patch in order for a change to be
188 complete, that is OK.  Simply note "this patch depends on patch X"
189 in your patch description.
191 If you cannot condense your patch set into a smaller set of patches,
192 then only post say 15 or so at a time and wait for review and integration.
196 4) Style check your changes.
198 Check your patch for basic style violations, details of which can be
199 found in Documentation/CodingStyle.  Failure to do so simply wastes
200 the reviewers time and will get your patch rejected, probably
201 without even being read.
203 At a minimum you should check your patches with the patch style
204 checker prior to submission (scripts/  You should
205 be able to justify all violations that remain in your patch.
209 5) Select e-mail destination.
211 Look through the MAINTAINERS file and the source code, and determine
212 if your change applies to a specific subsystem of the kernel, with
213 an assigned maintainer.  If so, e-mail that person.  The script
214 scripts/ can be very useful at this step.
216 If no maintainer is listed, or the maintainer does not respond, send
217 your patch to the primary Linux kernel developer's mailing list,
218  Most kernel developers monitor this
219 e-mail list, and can comment on your changes.
222 Do not send more than 15 patches at once to the vger mailing lists!!!
225 Linus Torvalds is the final arbiter of all changes accepted into the
226 Linux kernel.  His e-mail address is <>. 
227 He gets a lot of e-mail, so typically you should do your best to -avoid-
228 sending him e-mail. 
230 Patches which are bug fixes, are "obvious" changes, or similarly
231 require little discussion should be sent or CC'd to Linus.  Patches
232 which require discussion or do not have a clear advantage should
233 usually be sent first to linux-kernel.  Only after the patch is
234 discussed should the patch then be submitted to Linus.
238 6) Select your CC (e-mail carbon copy) list.
240 Unless you have a reason NOT to do so, CC
242 Other kernel developers besides Linus need to be aware of your change,
243 so that they may comment on it and offer code review and suggestions.
244 linux-kernel is the primary Linux kernel developer mailing list.
245 Other mailing lists are available for specific subsystems, such as
246 USB, framebuffer devices, the VFS, the SCSI subsystem, etc.  See the
247 MAINTAINERS file for a mailing list that relates specifically to
248 your change.
250 Majordomo lists of VGER.KERNEL.ORG at:
251         <>
253 If changes affect userland-kernel interfaces, please send
254 the MAN-PAGES maintainer (as listed in the MAINTAINERS file)
255 a man-pages patch, or at least a notification of the change,
256 so that some information makes its way into the manual pages.
258 Even if the maintainer did not respond in step #5, make sure to ALWAYS
259 copy the maintainer when you change their code.
261 For small patches you may want to CC the Trivial Patch Monkey
262 which collects "trivial" patches. Have a look
263 into the MAINTAINERS file for its current manager.
264 Trivial patches must qualify for one of the following rules:
265  Spelling fixes in documentation
266  Spelling fixes which could break grep(1)
267  Warning fixes (cluttering with useless warnings is bad)
268  Compilation fixes (only if they are actually correct)
269  Runtime fixes (only if they actually fix things)
270  Removing use of deprecated functions/macros (eg. check_region)
271  Contact detail and documentation fixes
272  Non-portable code replaced by portable code (even in arch-specific,
273  since people copy, as long as it's trivial)
274  Any fix by the author/maintainer of the file (ie. patch monkey
275  in re-transmission mode)
279 7) No MIME, no links, no compression, no attachments.  Just plain text.
281 Linus and other kernel developers need to be able to read and comment
282 on the changes you are submitting.  It is important for a kernel
283 developer to be able to "quote" your changes, using standard e-mail
284 tools, so that they may comment on specific portions of your code.
286 For this reason, all patches should be submitting e-mail "inline".
287 WARNING:  Be wary of your editor's word-wrap corrupting your patch,
288 if you choose to cut-n-paste your patch.
290 Do not attach the patch as a MIME attachment, compressed or not.
291 Many popular e-mail applications will not always transmit a MIME
292 attachment as plain text, making it impossible to comment on your
293 code.  A MIME attachment also takes Linus a bit more time to process,
294 decreasing the likelihood of your MIME-attached change being accepted.
296 Exception:  If your mailer is mangling patches then someone may ask
297 you to re-send them using MIME.
299 See Documentation/email-clients.txt for hints about configuring
300 your e-mail client so that it sends your patches untouched.
302 8) E-mail size.
304 When sending patches to Linus, always follow step #7.
306 Large changes are not appropriate for mailing lists, and some
307 maintainers.  If your patch, uncompressed, exceeds 300 kB in size,
308 it is preferred that you store your patch on an Internet-accessible
309 server, and provide instead a URL (link) pointing to your patch.
313 9) Name your kernel version.
315 It is important to note, either in the subject line or in the patch
316 description, the kernel version to which this patch applies.
318 If the patch does not apply cleanly to the latest kernel version,
319 Linus will not apply it.
323 10) Don't get discouraged.  Re-submit.
325 After you have submitted your change, be patient and wait.  If Linus
326 likes your change and applies it, it will appear in the next version
327 of the kernel that he releases.
329 However, if your change doesn't appear in the next version of the
330 kernel, there could be any number of reasons.  It's YOUR job to
331 narrow down those reasons, correct what was wrong, and submit your
332 updated change.
334 It is quite common for Linus to "drop" your patch without comment.
335 That's the nature of the system.  If he drops your patch, it could be
336 due to
337 * Your patch did not apply cleanly to the latest kernel version.
338 * Your patch was not sufficiently discussed on linux-kernel.
339 * A style issue (see section 2).
340 * An e-mail formatting issue (re-read this section).
341 * A technical problem with your change.
342 * He gets tons of e-mail, and yours got lost in the shuffle.
343 * You are being annoying.
345 When in doubt, solicit comments on linux-kernel mailing list.
349 11) Include PATCH in the subject
351 Due to high e-mail traffic to Linus, and to linux-kernel, it is common
352 convention to prefix your subject line with [PATCH].  This lets Linus
353 and other kernel developers more easily distinguish patches from other
354 e-mail discussions.
358 12) Sign your work
360 To improve tracking of who did what, especially with patches that can
361 percolate to their final resting place in the kernel through several
362 layers of maintainers, we've introduced a "sign-off" procedure on
363 patches that are being emailed around.
365 The sign-off is a simple line at the end of the explanation for the
366 patch, which certifies that you wrote it or otherwise have the right to
367 pass it on as an open-source patch.  The rules are pretty simple: if you
368 can certify the below:
370         Developer's Certificate of Origin 1.1
372         By making a contribution to this project, I certify that:
374         (a) The contribution was created in whole or in part by me and I
375             have the right to submit it under the open source license
376             indicated in the file; or
378         (b) The contribution is based upon previous work that, to the best
379             of my knowledge, is covered under an appropriate open source
380             license and I have the right under that license to submit that
381             work with modifications, whether created in whole or in part
382             by me, under the same open source license (unless I am
383             permitted to submit under a different license), as indicated
384             in the file; or
386         (c) The contribution was provided directly to me by some other
387             person who certified (a), (b) or (c) and I have not modified
388             it.
390         (d) I understand and agree that this project and the contribution
391             are public and that a record of the contribution (including all
392             personal information I submit with it, including my sign-off) is
393             maintained indefinitely and may be redistributed consistent with
394             this project or the open source license(s) involved.
396 then you just add a line saying
398         Signed-off-by: Random J Developer <>
400 using your real name (sorry, no pseudonyms or anonymous contributions.)
402 Some people also put extra tags at the end.  They'll just be ignored for
403 now, but you can do this to mark internal company procedures or just
404 point out some special detail about the sign-off. 
406 If you are a subsystem or branch maintainer, sometimes you need to slightly
407 modify patches you receive in order to merge them, because the code is not
408 exactly the same in your tree and the submitters'. If you stick strictly to
409 rule (c), you should ask the submitter to rediff, but this is a totally
410 counter-productive waste of time and energy. Rule (b) allows you to adjust
411 the code, but then it is very impolite to change one submitter's code and
412 make him endorse your bugs. To solve this problem, it is recommended that
413 you add a line between the last Signed-off-by header and yours, indicating
414 the nature of your changes. While there is nothing mandatory about this, it
415 seems like prepending the description with your mail and/or name, all
416 enclosed in square brackets, is noticeable enough to make it obvious that
417 you are responsible for last-minute changes. Example :
419         Signed-off-by: Random J Developer <>
420         [ struct foo moved from foo.c to foo.h]
421         Signed-off-by: Lucky K Maintainer <>
423 This practice is particularly helpful if you maintain a stable branch and
424 want at the same time to credit the author, track changes, merge the fix,
425 and protect the submitter from complaints. Note that under no circumstances
426 can you change the author's identity (the From header), as it is the one
427 which appears in the changelog.
429 Special note to back-porters: It seems to be a common and useful practice
430 to insert an indication of the origin of a patch at the top of the commit
431 message (just after the subject line) to facilitate tracking. For instance,
432 here's what we see in 2.6-stable :
434     Date:   Tue May 13 19:10:30 2008 +0000
436         SCSI: libiscsi regression in 2.6.25: fix nop timer handling
438         commit 4cf1043593db6a337f10e006c23c69e5fc93e722 upstream
440 And here's what appears in 2.4 :
442     Date:   Tue May 13 22:12:27 2008 +0200
444         wireless, airo: waitbusy() won't delay
446         [backport of 2.6 commit b7acbdfbd1f277c1eb23f344f899cfa4cd0bf36a]
448 Whatever the format, this information provides a valuable help to people
449 tracking your trees, and to people trying to trouble-shoot bugs in your
450 tree.
453 13) When to use Acked-by: and Cc:
455 The Signed-off-by: tag indicates that the signer was involved in the
456 development of the patch, or that he/she was in the patch's delivery path.
458 If a person was not directly involved in the preparation or handling of a
459 patch but wishes to signify and record their approval of it then they can
460 arrange to have an Acked-by: line added to the patch's changelog.
462 Acked-by: is often used by the maintainer of the affected code when that
463 maintainer neither contributed to nor forwarded the patch.
465 Acked-by: is not as formal as Signed-off-by:.  It is a record that the acker
466 has at least reviewed the patch and has indicated acceptance.  Hence patch
467 mergers will sometimes manually convert an acker's "yep, looks good to me"
468 into an Acked-by:.
470 Acked-by: does not necessarily indicate acknowledgement of the entire patch.
471 For example, if a patch affects multiple subsystems and has an Acked-by: from
472 one subsystem maintainer then this usually indicates acknowledgement of just
473 the part which affects that maintainer's code.  Judgement should be used here.
474 When in doubt people should refer to the original discussion in the mailing
475 list archives.
477 If a person has had the opportunity to comment on a patch, but has not
478 provided such comments, you may optionally add a "Cc:" tag to the patch.
479 This is the only tag which might be added without an explicit action by the
480 person it names.  This tag documents that potentially interested parties
481 have been included in the discussion
484 14) Using Reported-by:, Tested-by:, Reviewed-by:, Suggested-by: and Fixes:
486 The Reported-by tag gives credit to people who find bugs and report them and it
487 hopefully inspires them to help us again in the future.  Please note that if
488 the bug was reported in private, then ask for permission first before using the
489 Reported-by tag.
491 A Tested-by: tag indicates that the patch has been successfully tested (in
492 some environment) by the person named.  This tag informs maintainers that
493 some testing has been performed, provides a means to locate testers for
494 future patches, and ensures credit for the testers.
496 Reviewed-by:, instead, indicates that the patch has been reviewed and found
497 acceptable according to the Reviewer's Statement:
499         Reviewer's statement of oversight
501         By offering my Reviewed-by: tag, I state that:
503          (a) I have carried out a technical review of this patch to
504              evaluate its appropriateness and readiness for inclusion into
505              the mainline kernel.
507          (b) Any problems, concerns, or questions relating to the patch
508              have been communicated back to the submitter.  I am satisfied
509              with the submitter's response to my comments.
511          (c) While there may be things that could be improved with this
512              submission, I believe that it is, at this time, (1) a
513              worthwhile modification to the kernel, and (2) free of known
514              issues which would argue against its inclusion.
516          (d) While I have reviewed the patch and believe it to be sound, I
517              do not (unless explicitly stated elsewhere) make any
518              warranties or guarantees that it will achieve its stated
519              purpose or function properly in any given situation.
521 A Reviewed-by tag is a statement of opinion that the patch is an
522 appropriate modification of the kernel without any remaining serious
523 technical issues.  Any interested reviewer (who has done the work) can
524 offer a Reviewed-by tag for a patch.  This tag serves to give credit to
525 reviewers and to inform maintainers of the degree of review which has been
526 done on the patch.  Reviewed-by: tags, when supplied by reviewers known to
527 understand the subject area and to perform thorough reviews, will normally
528 increase the likelihood of your patch getting into the kernel.
530 A Suggested-by: tag indicates that the patch idea is suggested by the person
531 named and ensures credit to the person for the idea. Please note that this
532 tag should not be added without the reporter's permission, especially if the
533 idea was not posted in a public forum. That said, if we diligently credit our
534 idea reporters, they will, hopefully, be inspired to help us again in the
535 future.
537 A Fixes: tag indicates that the patch fixes an issue in a previous commit. It
538 is used to make it easy to determine where a bug originated, which can help
539 review a bug fix. This tag also assists the stable kernel team in determining
540 which stable kernel versions should receive your fix. This is the preferred
541 method for indicating a bug fixed by the patch. See #2 above for more details.
544 15) The canonical patch format
546 The canonical patch subject line is:
548     Subject: [PATCH 001/123] subsystem: summary phrase
550 The canonical patch message body contains the following:
552   - A "from" line specifying the patch author.
554   - An empty line.
556   - The body of the explanation, which will be copied to the
557     permanent changelog to describe this patch.
559   - The "Signed-off-by:" lines, described above, which will
560     also go in the changelog.
562   - A marker line containing simply "---".
564   - Any additional comments not suitable for the changelog.
566   - The actual patch (diff output).
568 The Subject line format makes it very easy to sort the emails
569 alphabetically by subject line - pretty much any email reader will
570 support that - since because the sequence number is zero-padded,
571 the numerical and alphabetic sort is the same.
573 The "subsystem" in the email's Subject should identify which
574 area or subsystem of the kernel is being patched.
576 The "summary phrase" in the email's Subject should concisely
577 describe the patch which that email contains.  The "summary
578 phrase" should not be a filename.  Do not use the same "summary
579 phrase" for every patch in a whole patch series (where a "patch
580 series" is an ordered sequence of multiple, related patches).
582 Bear in mind that the "summary phrase" of your email becomes a
583 globally-unique identifier for that patch.  It propagates all the way
584 into the git changelog.  The "summary phrase" may later be used in
585 developer discussions which refer to the patch.  People will want to
586 google for the "summary phrase" to read discussion regarding that
587 patch.  It will also be the only thing that people may quickly see
588 when, two or three months later, they are going through perhaps
589 thousands of patches using tools such as "gitk" or "git log
590 --oneline".
592 For these reasons, the "summary" must be no more than 70-75
593 characters, and it must describe both what the patch changes, as well
594 as why the patch might be necessary.  It is challenging to be both
595 succinct and descriptive, but that is what a well-written summary
596 should do.
598 The "summary phrase" may be prefixed by tags enclosed in square
599 brackets: "Subject: [PATCH tag] <summary phrase>".  The tags are not
600 considered part of the summary phrase, but describe how the patch
601 should be treated.  Common tags might include a version descriptor if
602 the multiple versions of the patch have been sent out in response to
603 comments (i.e., "v1, v2, v3"), or "RFC" to indicate a request for
604 comments.  If there are four patches in a patch series the individual
605 patches may be numbered like this: 1/4, 2/4, 3/4, 4/4.  This assures
606 that developers understand the order in which the patches should be
607 applied and that they have reviewed or applied all of the patches in
608 the patch series.
610 A couple of example Subjects:
612     Subject: [patch 2/5] ext2: improve scalability of bitmap searching
613     Subject: [PATCHv2 001/207] x86: fix eflags tracking
615 The "from" line must be the very first line in the message body,
616 and has the form:
618         From: Original Author <>
620 The "from" line specifies who will be credited as the author of the
621 patch in the permanent changelog.  If the "from" line is missing,
622 then the "From:" line from the email header will be used to determine
623 the patch author in the changelog.
625 The explanation body will be committed to the permanent source
626 changelog, so should make sense to a competent reader who has long
627 since forgotten the immediate details of the discussion that might
628 have led to this patch.  Including symptoms of the failure which the
629 patch addresses (kernel log messages, oops messages, etc.) is
630 especially useful for people who might be searching the commit logs
631 looking for the applicable patch.  If a patch fixes a compile failure,
632 it may not be necessary to include _all_ of the compile failures; just
633 enough that it is likely that someone searching for the patch can find
634 it.  As in the "summary phrase", it is important to be both succinct as
635 well as descriptive.
637 The "---" marker line serves the essential purpose of marking for patch
638 handling tools where the changelog message ends.
640 One good use for the additional comments after the "---" marker is for
641 a diffstat, to show what files have changed, and the number of
642 inserted and deleted lines per file.  A diffstat is especially useful
643 on bigger patches.  Other comments relevant only to the moment or the
644 maintainer, not suitable for the permanent changelog, should also go
645 here.  A good example of such comments might be "patch changelogs"
646 which describe what has changed between the v1 and v2 version of the
647 patch.
649 If you are going to include a diffstat after the "---" marker, please
650 use diffstat options "-p 1 -w 70" so that filenames are listed from
651 the top of the kernel source tree and don't use too much horizontal
652 space (easily fit in 80 columns, maybe with some indentation).  (git
653 generates appropriate diffstats by default.)
655 See more details on the proper patch format in the following
656 references.
659 16) Sending "git pull" requests  (from Linus emails)
661 Please write the git repo address and branch name alone on the same line
662 so that I can't even by mistake pull from the wrong branch, and so
663 that a triple-click just selects the whole thing.
665 So the proper format is something along the lines of:
667         "Please pull from
669                 git:// i2c-for-linus
671          to get these changes:"
673 so that I don't have to hunt-and-peck for the address and inevitably
674 get it wrong (actually, I've only gotten it wrong a few times, and
675 checking against the diffstat tells me when I get it wrong, but I'm
676 just a lot more comfortable when I don't have to "look for" the right
677 thing to pull, and double-check that I have the right branch-name).
680 Please use "git diff -M --stat --summary" to generate the diffstat:
681 the -M enables rename detection, and the summary enables a summary of
682 new/deleted or renamed files.
684 With rename detection, the statistics are rather different [...]
685 because git will notice that a fair number of the changes are renames.
687 -----------------------------------
689 -----------------------------------
691 This section lists many of the common "rules" associated with code
692 submitted to the kernel.  There are always exceptions... but you must
693 have a really good reason for doing so.  You could probably call this
694 section Linus Computer Science 101.
698 1) Read Documentation/CodingStyle
700 Nuff said.  If your code deviates too much from this, it is likely
701 to be rejected without further review, and without comment.
703 One significant exception is when moving code from one file to
704 another -- in this case you should not modify the moved code at all in
705 the same patch which moves it.  This clearly delineates the act of
706 moving the code and your changes.  This greatly aids review of the
707 actual differences and allows tools to better track the history of
708 the code itself.
710 Check your patches with the patch style checker prior to submission
711 (scripts/  The style checker should be viewed as
712 a guide not as the final word.  If your code looks better with
713 a violation then its probably best left alone.
715 The checker reports at three levels:
716  - ERROR: things that are very likely to be wrong
717  - WARNING: things requiring careful review
718  - CHECK: things requiring thought
720 You should be able to justify all violations that remain in your
721 patch.
725 2) #ifdefs are ugly
727 Code cluttered with ifdefs is difficult to read and maintain.  Don't do
728 it.  Instead, put your ifdefs in a header, and conditionally define
729 'static inline' functions, or macros, which are used in the code.
730 Let the compiler optimize away the "no-op" case.
732 Simple example, of poor code:
734         dev = alloc_etherdev (sizeof(struct funky_private));
735         if (!dev)
736                 return -ENODEV;
737         #ifdef CONFIG_NET_FUNKINESS
738         init_funky_net(dev);
739         #endif
741 Cleaned-up example:
743 (in header)
744         #ifndef CONFIG_NET_FUNKINESS
745         static inline void init_funky_net (struct net_device *d) {}
746         #endif
748 (in the code itself)
749         dev = alloc_etherdev (sizeof(struct funky_private));
750         if (!dev)
751                 return -ENODEV;
752         init_funky_net(dev);
756 3) 'static inline' is better than a macro
758 Static inline functions are greatly preferred over macros.
759 They provide type safety, have no length limitations, no formatting
760 limitations, and under gcc they are as cheap as macros.
762 Macros should only be used for cases where a static inline is clearly
763 suboptimal [there are a few, isolated cases of this in fast paths],
764 or where it is impossible to use a static inline function [such as
765 string-izing].
767 'static inline' is preferred over 'static __inline__', 'extern inline',
768 and 'extern __inline__'.
772 4) Don't over-design.
774 Don't try to anticipate nebulous future cases which may or may not
775 be useful:  "Make it as simple as you can, and no simpler."
779 ----------------------
781 ----------------------
783 Andrew Morton, "The perfect patch" (tpp).
784   <>
786 Jeff Garzik, "Linux kernel patch submission format".
787   <>
789 Greg Kroah-Hartman, "How to piss off a kernel subsystem maintainer".
790   <>
791   <>
792   <>
793   <>
794   <>
795   <>
797 NO!!!! No more huge patch bombs to people!
798   <>
800 Kernel Documentation/CodingStyle:
801   <>
803 Linus Torvalds's mail on the canonical patch format:
804   <>
806 Andi Kleen, "On submitting kernel patches"
807   Some strategies to get difficult or controversial changes in.
810 --

This page was automatically generated by LXR 0.3.1 (source).  •  Linux is a registered trademark of Linus Torvalds  •  Contact us