3Sooner or later, the time comes when your work is ready to be presented to
4the community for review and, eventually, inclusion into the mainline
5kernel. Unsurprisingly, the kernel development community has evolved a set
6of conventions and procedures which are used in the posting of patches;
7following them will make life much easier for everybody involved. This
8document will attempt to cover these expectations in reasonable detail;
9more information can also be found in the files SubmittingPatches,
10SubmittingDrivers, and SubmitChecklist in the kernel documentation
16There is a constant temptation to avoid posting patches before they are
17completely "ready." For simple patches, that is not a problem. If the
18work being done is complex, though, there is a lot to be gained by getting
19feedback from the community before the work is complete. So you should
20consider posting in-progress work, or even making a git tree available so
21that interested developers can catch up with your work at any time.
23When posting code which is not yet considered ready for inclusion, it is a
24good idea to say so in the posting itself. Also mention any major work
25which remains to be done and any known problems. Fewer people will look at
26patches which are known to be half-baked, but those who do will come in
27with the idea that they can help you drive the work in the right direction.
32There are a number of things which should be done before you consider
33sending patches to the development community. These include:
35 - Test the code to the extent that you can. Make use of the kernel's
36   debugging tools, ensure that the kernel will build with all reasonable
37   combinations of configuration options, use cross-compilers to build for
38   different architectures, etc.
40 - Make sure your code is compliant with the kernel coding style
41   guidelines.
43 - Does your change have performance implications? If so, you should run
44   benchmarks showing what the impact (or benefit) of your change is; a
45   summary of the results should be included with the patch.
47 - Be sure that you have the right to post the code. If this work was done
48   for an employer, the employer likely has a right to the work and must be
49   agreeable with its release under the GPL.
51As a general rule, putting in some extra thought before posting code almost
52always pays back the effort in short order.
57The preparation of patches for posting can be a surprising amount of work,
58but, once again, attempting to save time here is not generally advisable
59even in the short term.
61Patches must be prepared against a specific version of the kernel. As a
62general rule, a patch should be based on the current mainline as found in
63Linus's git tree. It may become necessary to make versions against -mm,
64linux-next, or a subsystem tree, though, to facilitate wider testing and
65review. Depending on the area of your patch and what is going on
66elsewhere, basing a patch against these other trees can require a
67significant amount of work resolving conflicts and dealing with API
70Only the most simple changes should be formatted as a single patch;
71everything else should be made as a logical series of changes. Splitting
72up patches is a bit of an art; some developers spend a long time figuring
73out how to do it in the way that the community expects. There are a few
74rules of thumb, however, which can help considerably:
76 - The patch series you post will almost certainly not be the series of
77   changes found in your working revision control system. Instead, the
78   changes you have made need to be considered in their final form, then
79   split apart in ways which make sense. The developers are interested in
80   discrete, self-contained changes, not the path you took to get to those
81   changes.
83 - Each logically independent change should be formatted as a separate
84   patch. These changes can be small ("add a field to this structure") or
85   large (adding a significant new driver, for example), but they should be
86   conceptually small and amenable to a one-line description. Each patch
87   should make a specific change which can be reviewed on its own and
88   verified to do what it says it does.
90 - As a way of restating the guideline above: do not mix different types of
91   changes in the same patch. If a single patch fixes a critical security
92   bug, rearranges a few structures, and reformats the code, there is a
93   good chance that it will be passed over and the important fix will be
94   lost.
96 - Each patch should yield a kernel which builds and runs properly; if your
97   patch series is interrupted in the middle, the result should still be a
98   working kernel. Partial application of a patch series is a common
99   scenario when the "git bisect" tool is used to find regressions; if the
100   result is a broken kernel, you will make life harder for developers and
101   users who are engaging in the noble work of tracking down problems.
103 - Do not overdo it, though. One developer recently posted a set of edits
104   to a single file as 500 separate patches - an act which did not make him
105   the most popular person on the kernel mailing list. A single patch can
106   be reasonably large as long as it still contains a single *logical*
107   change.
109 - It can be tempting to add a whole new infrastructure with a series of
110   patches, but to leave that infrastructure unused until the final patch
111   in the series enables the whole thing. This temptation should be
112   avoided if possible; if that series adds regressions, bisection will
113   finger the last patch as the one which caused the problem, even though
114   the real bug is elsewhere. Whenever possible, a patch which adds new
115   code should make that code active immediately.
117Working to create the perfect patch series can be a frustrating process
118which takes quite a bit of time and thought after the "real work" has been
119done. When done properly, though, it is time well spent.
124So now you have a perfect series of patches for posting, but the work is
125not done quite yet. Each patch needs to be formatted into a message which
126quickly and clearly communicates its purpose to the rest of the world. To
127that end, each patch will be composed of the following:
129 - An optional "From" line naming the author of the patch. This line is
130   only necessary if you are passing on somebody else's patch via email,
131   but it never hurts to add it when in doubt.
133 - A one-line description of what the patch does. This message should be
134   enough for a reader who sees it with no other context to figure out the
135   scope of the patch; it is the line that will show up in the "short form"
136   changelogs. This message is usually formatted with the relevant
137   subsystem name first, followed by the purpose of the patch. For
138   example:
140    gpio: fix build on CONFIG_GPIO_SYSFS=n
142 - A blank line followed by a detailed description of the contents of the
143   patch. This description can be as long as is required; it should say
144   what the patch does and why it should be applied to the kernel.
146 - One or more tag lines, with, at a minimum, one Signed-off-by: line from
147   the author of the patch. Tags will be described in more detail below.
149The items above, together, form the changelog for the patch. Writing good
150changelogs is a crucial but often-neglected art; it's worth spending
151another moment discussing this issue. When writing a changelog, you should
152bear in mind that a number of different people will be reading your words.
153These include subsystem maintainers and reviewers who need to decide
154whether the patch should be included, distributors and other maintainers
155trying to decide whether a patch should be backported to other kernels, bug
156hunters wondering whether the patch is responsible for a problem they are
157chasing, users who want to know how the kernel has changed, and more. A
158good changelog conveys the needed information to all of these people in the
159most direct and concise way possible.
161To that end, the summary line should describe the effects of and motivation
162for the change as well as possible given the one-line constraint. The
163detailed description can then amplify on those topics and provide any
164needed additional information. If the patch fixes a bug, cite the commit
165which introduced the bug if possible. If a problem is associated with
166specific log or compiler output, include that output to help others
167searching for a solution to the same problem. If the change is meant to
168support other changes coming in later patch, say so. If internal APIs are
169changed, detail those changes and how other developers should respond. In
170general, the more you can put yourself into the shoes of everybody who will
171be reading your changelog, the better that changelog (and the kernel as a
172whole) will be.
174Needless to say, the changelog should be the text used when committing the
175change to a revision control system. It will be followed by:
177 - The patch itself, in the unified ("-u") patch format. Using the "-p"
178   option to diff will associate function names with changes, making the
179   resulting patch easier for others to read.
181You should avoid including changes to irrelevant files (those generated by
182the build process, for example, or editor backup files) in the patch. The
183file "dontdiff" in the Documentation directory can help in this regard;
184pass it to diff with the "-X" option.
186The tags mentioned above are used to describe how various developers have
187been associated with the development of this patch. They are described in
188detail in the SubmittingPatches document; what follows here is a brief
189summary. Each of these lines has the format:
191    tag: Full Name <email address> optional-other-stuff
193The tags in common use are:
195 - Signed-off-by: this is a developer's certification that he or she has
196   the right to submit the patch for inclusion into the kernel. It is an
197   agreement to the Developer's Certificate of Origin, the full text of
198   which can be found in Documentation/SubmittingPatches. Code without a
199   proper signoff cannot be merged into the mainline.
201 - Acked-by: indicates an agreement by another developer (often a
202   maintainer of the relevant code) that the patch is appropriate for
203   inclusion into the kernel.
205 - Tested-by: states that the named person has tested the patch and found
206   it to work.
208 - Reviewed-by: the named developer has reviewed the patch for correctness;
209   see the reviewer's statement in Documentation/SubmittingPatches for more
210   detail.
212 - Reported-by: names a user who reported a problem which is fixed by this
213   patch; this tag is used to give credit to the (often underappreciated)
214   people who test our code and let us know when things do not work
215   correctly.
217 - Cc: the named person received a copy of the patch and had the
218   opportunity to comment on it.
220Be careful in the addition of tags to your patches: only Cc: is appropriate
221for addition without the explicit permission of the person named.
226Before you mail your patches, there are a couple of other things you should
227take care of:
229 - Are you sure that your mailer will not corrupt the patches? Patches
230   which have had gratuitous white-space changes or line wrapping performed
231   by the mail client will not apply at the other end, and often will not
232   be examined in any detail. If there is any doubt at all, mail the patch
233   to yourself and convince yourself that it shows up intact.
235   Documentation/email-clients.txt has some helpful hints on making
236   specific mail clients work for sending patches.
238 - Are you sure your patch is free of silly mistakes? You should always
239   run patches through scripts/ and address the complaints it
240   comes up with. Please bear in mind that, while being the
241   embodiment of a fair amount of thought about what kernel patches should
242   look like, is not smarter than you. If fixing a complaint
243   would make the code worse, don't do it.
245Patches should always be sent as plain text. Please do not send them as
246attachments; that makes it much harder for reviewers to quote sections of
247the patch in their replies. Instead, just put the patch directly into your
250When mailing patches, it is important to send copies to anybody who might
251be interested in it. Unlike some other projects, the kernel encourages
252people to err on the side of sending too many copies; don't assume that the
253relevant people will see your posting on the mailing lists. In particular,
254copies should go to:
256 - The maintainer(s) of the affected subsystem(s). As described earlier,
257   the MAINTAINERS file is the first place to look for these people.
259 - Other developers who have been working in the same area - especially
260   those who might be working there now. Using git to see who else has
261   modified the files you are working on can be helpful.
263 - If you are responding to a bug report or a feature request, copy the
264   original poster as well.
266 - Send a copy to the relevant mailing list, or, if nothing else applies,
267   the linux-kernel list.
269 - If you are fixing a bug, think about whether the fix should go into the
270   next stable update. If so, should get a copy of the
271   patch. Also add a "Cc:" to the tags within the patch
272   itself; that will cause the stable team to get a notification when your
273   fix goes into the mainline.
275When selecting recipients for a patch, it is good to have an idea of who
276you think will eventually accept the patch and get it merged. While it
277is possible to send patches directly to Linus Torvalds and have him merge
278them, things are not normally done that way. Linus is busy, and there are
279subsystem maintainers who watch over specific parts of the kernel. Usually
280you will be wanting that maintainer to merge your patches. If there is no
281obvious maintainer, Andrew Morton is often the patch target of last resort.
283Patches need good subject lines. The canonical format for a patch line is
284something like:
286    [PATCH nn/mm] subsys: one-line description of the patch
288where "nn" is the ordinal number of the patch, "mm" is the total number of
289patches in the series, and "subsys" is the name of the affected subsystem.
290Clearly, nn/mm can be omitted for a single, standalone patch.
292If you have a significant series of patches, it is customary to send an
293introductory description as part zero. This convention is not universally
294followed though; if you use it, remember that information in the
295introduction does not make it into the kernel changelogs. So please ensure
296that the patches, themselves, have complete changelog information.
298In general, the second and following parts of a multi-part patch should be
299sent as a reply to the first part so that they all thread together at the
300receiving end. Tools like git and quilt have commands to mail out a set of
301patches with the proper threading. If you have a long series, though, and
302are using git, please provide the --no-chain-reply-to option to avoid
303creating exceptionally deep nesting.

Archive Download this file