aboutsummaryrefslogtreecommitdiffstats
path: root/Documentation/development-process/2.Process
blob: d750321acd5a42560b0746d31c189ee3b3688f3e (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
2: HOW THE DEVELOPMENT PROCESS WORKS

Linux kernel development in the early 1990's was a pretty loose affair,
with relatively small numbers of users and developers involved.  With a
user base in the millions and with some 2,000 developers involved over the
course of one year, the kernel has since had to evolve a number of
processes to keep development happening smoothly.  A solid understanding of
how the process works is required in order to be an effective part of it.


2.1: THE BIG PICTURE

The kernel developers use a loosely time-based release process, with a new
major kernel release happening every two or three months.  The recent
release history looks like this:

	2.6.26	July 13, 2008
	2.6.25	April 16, 2008
	2.6.24	January 24, 2008
	2.6.23	October 9, 2007
	2.6.22	July 8, 2007
	2.6.21	April 25, 2007
	2.6.20	February 4, 2007

Every 2.6.x release is a major kernel release with new features, internal
API changes, and more.  A typical 2.6 release can contain over 10,000
changesets with changes to several hundred thousand lines of code.  2.6 is
thus the leading edge of Linux kernel development; the kernel uses a
rolling development model which is continually integrating major changes.

A relatively straightforward discipline is followed with regard to the
merging of patches for each release.  At the beginning of each development
cycle, the "merge window" is said to be open.  At that time, code which is
deemed to be sufficiently stable (and which is accepted by the development
community) is merged into the mainline kernel.  The bulk of changes for a
new development cycle (and all of the major changes) will be merged during
this time, at a rate approaching 1,000 changes ("patches," or "changesets")
per day.

(As an aside, it is worth noting that the changes integrated during the
merge window do not come out of thin air; they have been collected, tested,
and staged ahead of time.  How that process works will be described in
detail later on).

The merge window lasts for two weeks.  At the end of this time, Linus
Torvalds will declare that the window is closed and release the first of
the "rc" kernels.  For the kernel which is destined to be 2.6.26, for
example, the release which happens at the end of the merge window will be
called 2.6.26-rc1.  The -rc1 release is the signal that the time to merge
new features has passed, and that the time to stabilize the next kernel has
begun.

Over the next six to ten weeks, only patches which fix problems should be
submitted to the mainline.  On occasion a more significant change will be
allowed, but such occasions are rare; developers who try to merge new
features outside of the merge window tend to get an unfriendly reception.
As a general rule, if you miss the merge window for a given feature, the
best thing to do is to wait for the next development cycle.  (An occasional
exception is made for drivers for previously-unsupported hardware; if they
touch no in-tree code, they cannot cause regressions and should be safe to
add at any time).

As fixes make their way into the mainline, the patch rate will slow over
time.  Linus releases new -rc kernels about once a week; a normal series
will get up to somewhere between -rc6 and -rc9 before the kernel is
considered to be sufficiently stable and the final 2.6.x release is made.
At that point the whole process starts over again.

As an example, here is how the 2.6.25 development cycle went (all dates in
2008): 

	January 24	2.6.24 stable release
	February 10	2.6.25-rc1, merge window closes
	February 15	2.6.25-rc2
	February 24	2.6.25-rc3
	March 4	 	2.6.25-rc4
	March 9		2.6.25-rc5
	March 16	2.6.25-rc6
	March 25	2.6.25-rc7
	April 1		2.6.25-rc8
	April 11	2.6.25-rc9
	April 16	2.6.25 stable release

How do the developers decide when to close the development cycle and create
the stable release?  The most significant metric used is the list of
regressions from previous releases.  No bugs are welcome, but those which
break systems which worked in the past are considered to be especially
serious.  For this reason, patches which cause regressions are looked upon
unfavorably and are quite likely to be reverted during the stabilization
period. 

The developers' goal is to fix all known regressions before the stable
release is made.  In the real world, this kind of perfection is hard to
achieve; there are just too many variables in a project of this size.
There comes a point where delaying the final release just makes the problem
worse; the pile of changes waiting for the next merge window will grow
larger, creating even more regressions the next time around.  So most 2.6.x
kernels go out with a handful of known regressions though, hopefully, none
of them are serious.

Once a stable release is made, its ongoing maintenance is passed off to the
"stable team," currently comprised of Greg Kroah-Hartman and Chris Wright.
The stable team will release occasional updates to the stable release using
the 2.6.x.y numbering scheme.  To be considered for an update release, a
patch must (1) fix a significant bug, and (2) already be merged into the
mainline for the next development kernel.  Continuing our 2.6.25 example,
the history (as of this writing) is:

	May 1		2.6.25.1
	May 6		2.6.25.2 
	May 9		2.6.25.3 
	May 15		2.6.25.4
	June 7		2.6.25.5
	June 9		2.6.25.6
	June 16		2.6.25.7
	June 21		2.6.25.8
	June 24		2.6.25.9

Stable updates for a given kernel are made for approximately six months;
after that, the maintenance of stable releases is solely the responsibility
of the distributors which have shipped that particular kernel.


2.2: THE LIFECYCLE OF A PATCH

Patches do not go directly from the developer's keyboard into the mainline
kernel.  There is, instead, a somewhat involved (if somewhat informal)
process designed to ensure that each patch is reviewed for quality and that
each patch implements a change which is desirable to have in the mainline.
This process can happen quickly for minor fixes, or, in the case of large
and controversial changes, go on for years.  Much developer frustration
comes from a lack of understanding of this process or from attempts to
circumvent it.  

In the hopes of reducing that frustration, this document will describe how
a patch gets into the kernel.  What follows below is an introduction which
describes the process in a somewhat idealized way.  A much more detailed
treatment will come in later sections.

The stages that a patch goes through are, generally:

 - Design.  This is where the real requirements for the patch - and the way
   those requirements will be met - are laid out.  Design work is often
   done without involving the community, but it is better to do this work
   in the open if at all possible; it can save a lot of time redesigning
   things later.

 - Early review.  Patches are posted to the relevant mailing list, and
   developers on that list reply with any comments they may have.  This
   process should turn up any major problems with a patch if all goes
   well.

 - Wider review.  When the patch is getting close to ready for mainline
   inclusion, it will be accepted by a relevant subsystem maintainer -
   though this acceptance is not a guarantee that the patch will make it
   all the way to the mainline.  The patch will show up in the maintainer's
   subsystem tree and into the staging trees (described below).  When the
   process works, this step leads to more extensive review of the patch and
   the discovery of any problems resulting from the integration of this
   patch with work being done by others.

 - Merging into the mainline.  Eventually, a successful patch will be
   merged into the mainline repository managed by Linus Torvalds.  More
   comments and/or problems may surface at this time; it is important that
   the developer be responsive to these and fix any issues which arise.

 - Stable release.  The number of users potentially affected by the patch
   is now large, so, once again, new problems may arise.

 - Long-term maintenance.  While it is certainly possible for a developer
   to forget about code after merging it, that sort of behavior tends to
   leave a poor impression in the development community.  Merging code
   eliminates some of the maintenance burden, in that others will fix
   problems caused by API changes.  But the original developer should
   continue to take responsibility for the code if it is to remain useful
   in the longer term.

One of the largest mistakes made by kernel developers (or their employers)
is to try to cut the process down to a single "merging into the mainline"
step.  This approach invariably leads to frustration for everybody
involved.


2.3: HOW PATCHES GET INTO THE KERNEL

There is exactly one person who can merge patches into the mainline kernel
repository: Linus Torvalds.  But, of the over 12,000 patches which went
into the 2.6.25 kernel, only 250 (around 2%) were directly chosen by Linus
himself.  The kernel project has long since grown to a size where no single
developer could possibly inspect and select every patch unassisted.  The
way the kernel developers have addressed this growth is through the use of
a lieutenant system built around a chain of trust.

The kernel code base is logically broken down into a set of subsystems:
networking, specific architecture support, memory management, video
devices, etc.  Most subsystems have a designated maintainer, a developer
who has overall responsibility for the code within that subsystem.  These
subsystem maintainers are the gatekeepers (in a loose way) for the portion
of the kernel they manage; they are the ones who will (usually) accept a
patch for inclusion into the mainline kernel.

Subsystem maintainers each manage their own version of the kernel source
tree, usually (but certainly not always) using the git source management
tool.  Tools like git (and related tools like quilt or mercurial) allow
maintainers to track a list of patches, including authorship information
and other metadata.  At any given time, the maintainer can identify which
patches in his or her repository are not found in the mainline.

When the merge window opens, top-level maintainers will ask Linus to "pull"
the patches they have selected for merging from their repositories.  If
Linus agrees, the stream of patches will flow up into his repository,
becoming part of the mainline kernel.  The amount of attention that Linus
pays to specific patches received in a pull operation varies.  It is clear
that, sometimes, he looks quite closely.  But, as a general rule, Linus
trusts the subsystem maintainers to not send bad patches upstream.

Subsystem maintainers, in turn, can pull patches from other maintainers.
For example, the networking tree is built from patches which accumulated
first in trees dedicated to network device drivers, wireless networking,
etc.  This chain of repositories can be arbitrarily long, though it rarely
exceeds two or three links.  Since each maintainer in the chain trusts
those managing lower-level trees, this process is known as the "chain of
trust." 

Clearly, in a system like this, getting patches into the kernel depends on
finding the right maintainer.  Sending patches directly to Linus is not
normally the right way to go.


2.4: STAGING TREES

The chain of subsystem trees guides the flow of patches into the kernel,
but it also raises an interesting question: what if somebody wants to look
at all of the patches which are being prepared for the next merge window?
Developers will be interested in what other changes are pending to see
whether there are any conflicts to worry about; a patch which changes a
core kernel function prototype, for example, will conflict with any other
patches which use the older form of that function.  Reviewers and testers
want access to the changes in their integrated form before all of those
changes land in the mainline kernel.  One could pull changes from all of
the interesting subsystem trees, but that would be a big and error-prone
job.

The answer comes in the form of staging trees, where subsystem trees are
collected for testing and review.  The older of these trees, maintained by
Andrew Morton, is called "-mm" (for memory management, which is how it got
started).  The -mm tree integrates patches from a long list of subsystem
trees; it also has some patches aimed at helping with debugging.  

Beyond that, -mm contains a significant collection of patches which have
been selected by Andrew directly.  These patches may have been posted on a
mailing list, or they may apply to a part of the kernel for which there is
no designated subsystem tree.  As a result, -mm operates as a sort of
subsystem tree of last resort; if there is no other obvious path for a
patch into the mainline, it is likely to end up in -mm.  Miscellaneous
patches which accumulate in -mm will eventually either be forwarded on to
an appropriate subsystem tree or be sent directly to Linus.  In a typical
development cycle, approximately 10% of the patches going into the mainline
get there via -mm.

The current -mm patch can always be found from the front page of

	http://kernel.org/

Those who want to see the current state of -mm can get the "-mm of the
moment" tree, found at:

	http://userweb.kernel.org/~akpm/mmotm/

Use of the MMOTM tree is likely to be a frustrating experience, though;
there is a definite chance that it will not even compile.

The other staging tree, started more recently, is linux-next, maintained by
Stephen Rothwell.  The linux-next tree is, by design, a snapshot of what
the mainline is expected to look like after the next merge window closes.
Linux-next trees are announced on the linux-kernel and linux-next mailing
lists when they are assembled; they can be downloaded from:

	http://www.kernel.org/pub/linux/kernel/people/sfr/linux-next/

Some information about linux-next has been gathered at:

	http://linux.f-seidel.de/linux-next/pmwiki/

How the linux-next tree will fit into the development process is still
changing.  As of this writing, the first full development cycle involving
linux-next (2.6.26) is coming to an end; thus far, it has proved to be a
valuable resource for finding and fixing integration problems before the
beginning of the merge window.  See http://lwn.net/Articles/287155/ for
more information on how linux-next has worked to set up the 2.6.27 merge
window.

Some developers have begun to suggest that linux-next should be used as the
target for future development as well.  The linux-next tree does tend to be
far ahead of the mainline and is more representative of the tree into which
any new work will be merged.  The downside to this idea is that the
volatility of linux-next tends to make it a difficult development target.
See http://lwn.net/Articles/289013/ for more information on this topic, and
stay tuned; much is still in flux where linux-next is involved.


2.5: TOOLS

As can be seen from the above text, the kernel development process depends
heavily on the ability to herd collections of patches in various
directions.  The whole thing would not work anywhere near as well as it
does without suitably powerful tools.  Tutorials on how to use these tools
are well beyond the scope of this document, but there is space for a few
pointers.

By far the dominant source code management system used by the kernel
community is git.  Git is one of a number of distributed version control
systems being developed in the free software community.  It is well tuned
for kernel development, in that it performs quite well when dealing with
large repositories and large numbers of patches.  It also has a reputation
for being difficult to learn and use, though it has gotten better over
time.  Some sort of familiarity with git is almost a requirement for kernel
developers; even if they do not use it for their own work, they'll need git
to keep up with what other developers (and the mainline) are doing.

Git is now packaged by almost all Linux distributions.  There is a home
page at 

	http://git.or.cz/

That page has pointers to documentation and tutorials.  One should be
aware, in particular, of the Kernel Hacker's Guide to git, which has
information specific to kernel development:

	http://linux.yyz.us/git-howto.html

Among the kernel developers who do not use git, the most popular choice is
almost certainly Mercurial:

	http://www.selenic.com/mercurial/

Mercurial shares many features with git, but it provides an interface which
many find easier to use.

The other tool worth knowing about is Quilt:

	http://savannah.nongnu.org/projects/quilt/

Quilt is a patch management system, rather than a source code management
system.  It does not track history over time; it is, instead, oriented
toward tracking a specific set of changes against an evolving code base.
Some major subsystem maintainers use quilt to manage patches intended to go
upstream.  For the management of certain kinds of trees (-mm, for example),
quilt is the best tool for the job.


2.6: MAILING LISTS

A great deal of Linux kernel development work is done by way of mailing
lists.  It is hard to be a fully-functioning member of the community
without joining at least one list somewhere.  But Linux mailing lists also
represent a potential hazard to developers, who risk getting buried under a
load of electronic mail, running afoul of the conventions used on the Linux
lists, or both.

Most kernel mailing lists are run on vger.kernel.org; the master list can
be found at:

	http://vger.kernel.org/vger-lists.html

There are lists hosted elsewhere, though; a number of them are at
lists.redhat.com.

The core mailing list for kernel development is, of course, linux-kernel.
This list is an intimidating place to be; volume can reach 500 messages per
day, the amount of noise is high, the conversation can be severely
technical, and participants are not always concerned with showing a high
degree of politeness.  But there is no other place where the kernel
development community comes together as a whole; developers who avoid this
list will miss important information.

There are a few hints which can help with linux-kernel survival:

- Have the list delivered to a separate folder, rather than your main
  mailbox.  One must be able to ignore the stream for sustained periods of
  time.

- Do not try to follow every conversation - nobody else does.  It is
  important to filter on both the topic of interest (though note that
  long-running conversations can drift away from the original subject
  without changing the email subject line) and the people who are
  participating.  

- Do not feed the trolls.  If somebody is trying to stir up an angry
  response, ignore them.

- When responding to linux-kernel email (or that on other lists) preserve
  the Cc: header for all involved.  In the absence of a strong reason (such
  as an explicit request), you should never remove recipients.  Always make
  sure that the person you are responding to is in the Cc: list.  This
  convention also makes it unnecessary to explicitly ask to be copied on
  replies to your postings.

- Search the list archives (and the net as a whole) before asking
  questions.  Some developers can get impatient with people who clearly
  have not done their homework.

- Avoid top-posting (the practice of putting your answer above the quoted
  text you are responding to).  It makes your response harder to read and
  makes a poor impression.

- Ask on the correct mailing list.  Linux-kernel may be the general meeting
  point, but it is not the best place to find developers from all
  subsystems.

The last point - finding the correct mailing list - is a common place for
beginning developers to go wrong.  Somebody who asks a networking-related
question on linux-kernel will almost certainly receive a polite suggestion
to ask on the netdev list instead, as that is the list frequented by most
networking developers.  Other lists exist for the SCSI, video4linux, IDE,
filesystem, etc. subsystems.  The best place to look for mailing lists is
in the MAINTAINERS file packaged with the kernel source.


2.7: GETTING STARTED WITH KERNEL DEVELOPMENT

Questions about how to get started with the kernel development process are
common - from both individuals and companies.  Equally common are missteps
which make the beginning of the relationship harder than it has to be.

Companies often look to hire well-known developers to get a development
group started.  This can, in fact, be an effective technique.  But it also
tends to be expensive and does not do much to grow the pool of experienced
kernel developers.  It is possible to bring in-house developers up to speed
on Linux kernel development, given the investment of a bit of time.  Taking
this time can endow an employer with a group of developers who understand
the kernel and the company both, and who can help to train others as well.
Over the medium term, this is often the more profitable approach.

Individual developers are often, understandably, at a loss for a place to
start.  Beginning with a large project can be intimidating; one often wants
to test the waters with something smaller first.  This is the point where
some developers jump into the creation of patches fixing spelling errors or
minor coding style issues.  Unfortunately, such patches create a level of
noise which is distracting for the development community as a whole, so,
increasingly, they are looked down upon.  New developers wishing to
introduce themselves to the community will not get the sort of reception
they wish for by these means.

Andrew Morton gives this advice for aspiring kernel developers

	The #1 project for all kernel beginners should surely be "make sure
	that the kernel runs perfectly at all times on all machines which
	you can lay your hands on".  Usually the way to do this is to work
	with others on getting things fixed up (this can require
	persistence!) but that's fine - it's a part of kernel development.

(http://lwn.net/Articles/283982/).

In the absence of obvious problems to fix, developers are advised to look
at the current lists of regressions and open bugs in general.  There is
never any shortage of issues in need of fixing; by addressing these issues,
developers will gain experience with the process while, at the same time,
building respect with the rest of the development community.