blob: a173cd5f93d284f8ca252bdb9b0a7261a470d4ab [file] [log] [blame]
Mauro Carvalho Chehabf7c9fe42016-09-19 08:07:36 -03001.. _development_followthrough:
2
3Followthrough
4=============
Jonathan Corbet75b02142008-09-30 15:15:56 -06005
6At this point, you have followed the guidelines given so far and, with the
7addition of your own engineering skills, have posted a perfect series of
8patches. One of the biggest mistakes that even experienced kernel
9developers can make is to conclude that their work is now done. In truth,
10posting patches indicates a transition into the next stage of the process,
11with, possibly, quite a bit of work yet to be done.
12
13It is a rare patch which is so good at its first posting that there is no
14room for improvement. The kernel development process recognizes this fact,
15and, as a result, is heavily oriented toward the improvement of posted
16code. You, as the author of that code, will be expected to work with the
17kernel community to ensure that your code is up to the kernel's quality
18standards. A failure to participate in this process is quite likely to
19prevent the inclusion of your patches into the mainline.
20
21
Mauro Carvalho Chehabf7c9fe42016-09-19 08:07:36 -030022Working with reviewers
23----------------------
Jonathan Corbet75b02142008-09-30 15:15:56 -060024
25A patch of any significance will result in a number of comments from other
26developers as they review the code. Working with reviewers can be, for
27many developers, the most intimidating part of the kernel development
28process. Life can be made much easier, though, if you keep a few things in
29mind:
30
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.
38
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.
46
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.
53
54What all of this comes down to is that, when reviewers send you comments,
55you need to pay attention to the technical observations that they are
56making. Do not let their form of expression or your own pride keep that
57from happening. When you get review comments on a patch, take the time to
58understand what the reviewer is trying to say. If possible, fix the things
59that the reviewer is asking you to fix. And respond back to the reviewer:
60thank them, and describe how you will answer their questions.
61
62Note that you do not have to agree with every change suggested by
63reviewers. If you believe that the reviewer has misunderstood your code,
64explain what is really going on. If you have a technical objection to a
65suggested change, describe it and justify your solution to the problem. If
66your explanations make sense, the reviewer will accept them. Should your
67explanation not prove persuasive, though, especially if others start to
68agree with the reviewer, take some time to think things over again. It can
69be easy to become blinded by your own solution to a problem to the point
70that you don't realize that something is fundamentally wrong or, perhaps,
71you're not even solving the right problem.
72
Jonathan Corbet5c050fb2011-03-25 12:17:53 -060073Andrew Morton has suggested that every review comment which does not result
74in a code change should result in an additional code comment instead; that
75can help future reviewers avoid the questions which came up the first time
76around.
77
Jonathan Corbet75b02142008-09-30 15:15:56 -060078One fatal mistake is to ignore review comments in the hope that they will
79go away. They will not go away. If you repost code without having
80responded to the comments you got the time before, you're likely to find
81that your patches go nowhere.
82
83Speaking of reposting code: please bear in mind that reviewers are not
84going to remember all the details of the code you posted the last time
85around. So it is always a good idea to remind reviewers of previously
86raised issues and how you dealt with them; the patch changelog is a good
87place for this kind of information. Reviewers should not have to search
88through list archives to familiarize themselves with what was said last
89time; if you help them get a running start, they will be in a better mood
90when they revisit your code.
91
92What if you've tried to do everything right and things still aren't going
93anywhere? Most technical disagreements can be resolved through discussion,
94but there are times when somebody simply has to make a decision. If you
95honestly believe that this decision is going against you wrongly, you can
96always try appealing to a higher power. As of this writing, that higher
97power tends to be Andrew Morton. Andrew has a great deal of respect in the
98kernel development community; he can often unjam a situation which seems to
99be hopelessly blocked. Appealing to Andrew should not be done lightly,
100though, and not before all other alternatives have been explored. And bear
101in mind, of course, that he may not agree with you either.
102
103
Mauro Carvalho Chehabf7c9fe42016-09-19 08:07:36 -0300104What happens next
105-----------------
Jonathan Corbet75b02142008-09-30 15:15:56 -0600106
107If a patch is considered to be a good thing to add to the kernel, and once
108most of the review issues have been resolved, the next step is usually
109entry into a subsystem maintainer's tree. How that works varies from one
110subsystem to the next; each maintainer has his or her own way of doing
111things. In particular, there may be more than one tree - one, perhaps,
112dedicated to patches planned for the next merge window, and another for
Jonathan Corbet5c050fb2011-03-25 12:17:53 -0600113longer-term work.
Jonathan Corbet75b02142008-09-30 15:15:56 -0600114
115For patches applying to areas for which there is no obvious subsystem tree
116(memory management patches, for example), the default tree often ends up
117being -mm. Patches which affect multiple subsystems can also end up going
118through the -mm tree.
119
120Inclusion into a subsystem tree can bring a higher level of visibility to a
121patch. Now other developers working with that tree will get the patch by
Jonathan Corbet5c050fb2011-03-25 12:17:53 -0600122default. Subsystem trees typically feed linux-next as well, making their
123contents visible to the development community as a whole. At this point,
124there's a good chance that you will get more comments from a new set of
125reviewers; these comments need to be answered as in the previous round.
Jonathan Corbet75b02142008-09-30 15:15:56 -0600126
127What may also happen at this point, depending on the nature of your patch,
128is that conflicts with work being done by others turn up. In the worst
129case, heavy patch conflicts can result in some work being put on the back
130burner so that the remaining patches can be worked into shape and merged.
131Other times, conflict resolution will involve working with the other
132developers and, possibly, moving some patches between trees to ensure that
133everything applies cleanly. This work can be a pain, but count your
134blessings: before the advent of the linux-next tree, these conflicts often
135only turned up during the merge window and had to be addressed in a hurry.
136Now they can be resolved at leisure, before the merge window opens.
137
138Some day, if all goes well, you'll log on and see that your patch has been
139merged into the mainline kernel. Congratulations! Once the celebration is
140complete (and you have added yourself to the MAINTAINERS file), though, it
141is worth remembering an important little fact: the job still is not done.
142Merging into the mainline brings its own challenges.
143
144To begin with, the visibility of your patch has increased yet again. There
145may be a new round of comments from developers who had not been aware of
146the patch before. It may be tempting to ignore them, since there is no
147longer any question of your code being merged. Resist that temptation,
148though; you still need to be responsive to developers who have questions or
149suggestions.
150
151More importantly, though: inclusion into the mainline puts your code into
152the hands of a much larger group of testers. Even if you have contributed
153a driver for hardware which is not yet available, you will be surprised by
154how many people will build your code into their kernels. And, of course,
155where there are testers, there will be bug reports.
156
157The worst sort of bug reports are regressions. If your patch causes a
158regression, you'll find an uncomfortable number of eyes upon you;
159regressions need to be fixed as soon as possible. If you are unwilling or
160unable to fix the regression (and nobody else does it for you), your patch
161will almost certainly be removed during the stabilization period. Beyond
162negating all of the work you have done to get your patch into the mainline,
163having a patch pulled as the result of a failure to fix a regression could
164well make it harder for you to get work merged in the future.
165
166After any regressions have been dealt with, there may be other, ordinary
167bugs to deal with. The stabilization period is your best opportunity to
168fix these bugs and ensure that your code's debut in a mainline kernel
169release is as solid as possible. So, please, answer bug reports, and fix
170the problems if at all possible. That's what the stabilization period is
171for; you can start creating cool new patches once any problems with the old
172ones have been taken care of.
173
174And don't forget that there are other milestones which may also create bug
175reports: the next mainline stable release, when prominent distributors pick
176up a version of the kernel containing your patch, etc. Continuing to
177respond to these reports is a matter of basic pride in your work. If that
178is insufficient motivation, though, it's also worth considering that the
179development community remembers developers who lose interest in their code
180after it's merged. The next time you post a patch, they will be evaluating
181it with the assumption that you will not be around to maintain it
182afterward.
183
184
Mauro Carvalho Chehabf7c9fe42016-09-19 08:07:36 -0300185Other things that can happen
186-----------------------------
Jonathan Corbet75b02142008-09-30 15:15:56 -0600187
188One day, you may open your mail client and see that somebody has mailed you
189a patch to your code. That is one of the advantages of having your code
190out there in the open, after all. If you agree with the patch, you can
191either forward it on to the subsystem maintainer (be sure to include a
192proper From: line so that the attribution is correct, and add a signoff of
193your own), or send an Acked-by: response back and let the original poster
194send it upward.
195
196If you disagree with the patch, send a polite response explaining why. If
197possible, tell the author what changes need to be made to make the patch
198acceptable to you. There is a certain resistance to merging patches which
199are opposed by the author and maintainer of the code, but it only goes so
200far. If you are seen as needlessly blocking good work, those patches will
201eventually flow around you and get into the mainline anyway. In the Linux
202kernel, nobody has absolute veto power over any code. Except maybe Linus.
203
204On very rare occasion, you may see something completely different: another
205developer posts a different solution to your problem. At that point,
206chances are that one of the two patches will not be merged, and "mine was
207here first" is not considered to be a compelling technical argument. If
208somebody else's patch displaces yours and gets into the mainline, there is
209really only one way to respond: be pleased that your problem got solved and
210get on with your work. Having one's work shoved aside in this manner can
211be hurtful and discouraging, but the community will remember your reaction
212long after they have forgotten whose patch actually got merged.