1 .. _development_followthrough:
6 At this point, you have followed the guidelines given so far and, with the
7 addition of your own engineering skills, have posted a perfect series of
8 patches. One of the biggest mistakes that even experienced kernel
9 developers can make is to conclude that their work is now done. In truth,
10 posting patches indicates a transition into the next stage of the process,
11 with, possibly, quite a bit of work yet to be done.
13 It is a rare patch which is so good at its first posting that there is no
14 room for improvement. The kernel development process recognizes this fact,
15 and, as a result, is heavily oriented toward the improvement of posted
16 code. You, as the author of that code, will be expected to work with the
17 kernel community to ensure that your code is up to the kernel's quality
18 standards. A failure to participate in this process is quite likely to
19 prevent the inclusion of your patches into the mainline.
22 Working with reviewers
23 ----------------------
25 A patch of any significance will result in a number of comments from other
26 developers as they review the code. Working with reviewers can be, for
27 many developers, the most intimidating part of the kernel development
28 process. Life can be made much easier, though, if you keep a few things in
31 - If you have explained your patch well, reviewers will understand its
32 value and why you went to the trouble of writing it. But that value
33 will not keep them from asking a fundamental question: what will it be
34 like to maintain a kernel with this code in it five or ten years later?
35 Many of the changes you may be asked to make - from coding style tweaks
36 to substantial rewrites - come from the understanding that Linux will
37 still be around and under development a decade from now.
39 - Code review is hard work, and it is a relatively thankless occupation;
40 people remember who wrote kernel code, but there is little lasting fame
41 for those who reviewed it. So reviewers can get grumpy, especially when
42 they see the same mistakes being made over and over again. If you get a
43 review which seems angry, insulting, or outright offensive, resist the
44 impulse to respond in kind. Code review is about the code, not about
45 the people, and code reviewers are not attacking you personally.
47 - Similarly, code reviewers are not trying to promote their employers'
48 agendas at the expense of your own. Kernel developers often expect to
49 be working on the kernel years from now, but they understand that their
50 employer could change. They truly are, almost without exception,
51 working toward the creation of the best kernel they can; they are not
52 trying to create discomfort for their employers' competitors.
54 What all of this comes down to is that, when reviewers send you comments,
55 you need to pay attention to the technical observations that they are
56 making. Do not let their form of expression or your own pride keep that
57 from happening. When you get review comments on a patch, take the time to
58 understand what the reviewer is trying to say. If possible, fix the things
59 that the reviewer is asking you to fix. And respond back to the reviewer:
60 thank them, and describe how you will answer their questions.
62 Note that you do not have to agree with every change suggested by
63 reviewers. If you believe that the reviewer has misunderstood your code,
64 explain what is really going on. If you have a technical objection to a
65 suggested change, describe it and justify your solution to the problem. If
66 your explanations make sense, the reviewer will accept them. Should your
67 explanation not prove persuasive, though, especially if others start to
68 agree with the reviewer, take some time to think things over again. It can
69 be easy to become blinded by your own solution to a problem to the point
70 that you don't realize that something is fundamentally wrong or, perhaps,
71 you're not even solving the right problem.
73 Andrew Morton has suggested that every review comment which does not result
74 in a code change should result in an additional code comment instead; that
75 can help future reviewers avoid the questions which came up the first time
78 One fatal mistake is to ignore review comments in the hope that they will
79 go away. They will not go away. If you repost code without having
80 responded to the comments you got the time before, you're likely to find
81 that your patches go nowhere.
83 Speaking of reposting code: please bear in mind that reviewers are not
84 going to remember all the details of the code you posted the last time
85 around. So it is always a good idea to remind reviewers of previously
86 raised issues and how you dealt with them; the patch changelog is a good
87 place for this kind of information. Reviewers should not have to search
88 through list archives to familiarize themselves with what was said last
89 time; if you help them get a running start, they will be in a better mood
90 when they revisit your code.
92 What if you've tried to do everything right and things still aren't going
93 anywhere? Most technical disagreements can be resolved through discussion,
94 but there are times when somebody simply has to make a decision. If you
95 honestly believe that this decision is going against you wrongly, you can
96 always try appealing to a higher power. As of this writing, that higher
97 power tends to be Andrew Morton. Andrew has a great deal of respect in the
98 kernel development community; he can often unjam a situation which seems to
99 be hopelessly blocked. Appealing to Andrew should not be done lightly,
100 though, and not before all other alternatives have been explored. And bear
101 in mind, of course, that he may not agree with you either.
107 If a patch is considered to be a good thing to add to the kernel, and once
108 most of the review issues have been resolved, the next step is usually
109 entry into a subsystem maintainer's tree. How that works varies from one
110 subsystem to the next; each maintainer has his or her own way of doing
111 things. In particular, there may be more than one tree - one, perhaps,
112 dedicated to patches planned for the next merge window, and another for
115 For patches applying to areas for which there is no obvious subsystem tree
116 (memory management patches, for example), the default tree often ends up
117 being -mm. Patches which affect multiple subsystems can also end up going
118 through the -mm tree.
120 Inclusion into a subsystem tree can bring a higher level of visibility to a
121 patch. Now other developers working with that tree will get the patch by
122 default. Subsystem trees typically feed linux-next as well, making their
123 contents visible to the development community as a whole. At this point,
124 there's a good chance that you will get more comments from a new set of
125 reviewers; these comments need to be answered as in the previous round.
127 What may also happen at this point, depending on the nature of your patch,
128 is that conflicts with work being done by others turn up. In the worst
129 case, heavy patch conflicts can result in some work being put on the back
130 burner so that the remaining patches can be worked into shape and merged.
131 Other times, conflict resolution will involve working with the other
132 developers and, possibly, moving some patches between trees to ensure that
133 everything applies cleanly. This work can be a pain, but count your
134 blessings: before the advent of the linux-next tree, these conflicts often
135 only turned up during the merge window and had to be addressed in a hurry.
136 Now they can be resolved at leisure, before the merge window opens.
138 Some day, if all goes well, you'll log on and see that your patch has been
139 merged into the mainline kernel. Congratulations! Once the celebration is
140 complete (and you have added yourself to the MAINTAINERS file), though, it
141 is worth remembering an important little fact: the job still is not done.
142 Merging into the mainline brings its own challenges.
144 To begin with, the visibility of your patch has increased yet again. There
145 may be a new round of comments from developers who had not been aware of
146 the patch before. It may be tempting to ignore them, since there is no
147 longer any question of your code being merged. Resist that temptation,
148 though; you still need to be responsive to developers who have questions or
151 More importantly, though: inclusion into the mainline puts your code into
152 the hands of a much larger group of testers. Even if you have contributed
153 a driver for hardware which is not yet available, you will be surprised by
154 how many people will build your code into their kernels. And, of course,
155 where there are testers, there will be bug reports.
157 The worst sort of bug reports are regressions. If your patch causes a
158 regression, you'll find an uncomfortable number of eyes upon you;
159 regressions need to be fixed as soon as possible. If you are unwilling or
160 unable to fix the regression (and nobody else does it for you), your patch
161 will almost certainly be removed during the stabilization period. Beyond
162 negating all of the work you have done to get your patch into the mainline,
163 having a patch pulled as the result of a failure to fix a regression could
164 well make it harder for you to get work merged in the future.
166 After any regressions have been dealt with, there may be other, ordinary
167 bugs to deal with. The stabilization period is your best opportunity to
168 fix these bugs and ensure that your code's debut in a mainline kernel
169 release is as solid as possible. So, please, answer bug reports, and fix
170 the problems if at all possible. That's what the stabilization period is
171 for; you can start creating cool new patches once any problems with the old
172 ones have been taken care of.
174 And don't forget that there are other milestones which may also create bug
175 reports: the next mainline stable release, when prominent distributors pick
176 up a version of the kernel containing your patch, etc. Continuing to
177 respond to these reports is a matter of basic pride in your work. If that
178 is insufficient motivation, though, it's also worth considering that the
179 development community remembers developers who lose interest in their code
180 after it's merged. The next time you post a patch, they will be evaluating
181 it with the assumption that you will not be around to maintain it
185 Other things that can happen
186 -----------------------------
188 One day, you may open your mail client and see that somebody has mailed you
189 a patch to your code. That is one of the advantages of having your code
190 out there in the open, after all. If you agree with the patch, you can
191 either forward it on to the subsystem maintainer (be sure to include a
192 proper From: line so that the attribution is correct, and add a signoff of
193 your own), or send an Acked-by: response back and let the original poster
196 If you disagree with the patch, send a polite response explaining why. If
197 possible, tell the author what changes need to be made to make the patch
198 acceptable to you. There is a certain resistance to merging patches which
199 are opposed by the author and maintainer of the code, but it only goes so
200 far. If you are seen as needlessly blocking good work, those patches will
201 eventually flow around you and get into the mainline anyway. In the Linux
202 kernel, nobody has absolute veto power over any code. Except maybe Linus.
204 On very rare occasion, you may see something completely different: another
205 developer posts a different solution to your problem. At that point,
206 chances are that one of the two patches will not be merged, and "mine was
207 here first" is not considered to be a compelling technical argument. If
208 somebody else's patch displaces yours and gets into the mainline, there is
209 really only one way to respond: be pleased that your problem got solved and
210 get on with your work. Having one's work shoved aside in this manner can
211 be hurtful and discouraging, but the community will remember your reaction
212 long after they have forgotten whose patch actually got merged.