blob: 3171822c22a5d4b41d15820398798da52b98a703 [file] [log] [blame]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001
2 Linux kernel coding style
3
4This is a short document describing the preferred coding style for the
5linux kernel. Coding style is very personal, and I won't _force_ my
6views on anybody, but this is what goes for anything that I have to be
7able to maintain, and I'd prefer it for most other things too. Please
8at least consider the points made here.
9
10First off, I'd suggest printing out a copy of the GNU coding standards,
11and NOT read it. Burn them, it's a great symbolic gesture.
12
13Anyway, here goes:
14
15
16 Chapter 1: Indentation
17
18Tabs are 8 characters, and thus indentations are also 8 characters.
19There are heretic movements that try to make indentations 4 (or even 2!)
20characters deep, and that is akin to trying to define the value of PI to
21be 3.
22
23Rationale: The whole idea behind indentation is to clearly define where
24a block of control starts and ends. Especially when you've been looking
25at your screen for 20 straight hours, you'll find it a lot easier to see
26how the indentation works if you have large indentations.
27
28Now, some people will claim that having 8-character indentations makes
29the code move too far to the right, and makes it hard to read on a
3080-character terminal screen. The answer to that is that if you need
31more than 3 levels of indentation, you're screwed anyway, and should fix
32your program.
33
34In short, 8-char indents make things easier to read, and have the added
35benefit of warning you when you're nesting your functions too deep.
36Heed that warning.
37
Randy Dunlapb3fc9942006-12-10 02:18:56 -080038The preferred way to ease multiple indentation levels in a switch statement is
39to align the "switch" and its subordinate "case" labels in the same column
40instead of "double-indenting" the "case" labels. E.g.:
41
42 switch (suffix) {
43 case 'G':
44 case 'g':
45 mem <<= 30;
46 break;
47 case 'M':
48 case 'm':
49 mem <<= 20;
50 break;
51 case 'K':
52 case 'k':
53 mem <<= 10;
54 /* fall through */
55 default:
56 break;
57 }
58
59
Linus Torvalds1da177e2005-04-16 15:20:36 -070060Don't put multiple statements on a single line unless you have
61something to hide:
62
63 if (condition) do_this;
64 do_something_everytime;
65
Randy Dunlapb3fc9942006-12-10 02:18:56 -080066Don't put multiple assignments on a single line either. Kernel coding style
67is super simple. Avoid tricky expressions.
68
Linus Torvalds1da177e2005-04-16 15:20:36 -070069Outside of comments, documentation and except in Kconfig, spaces are never
70used for indentation, and the above example is deliberately broken.
71
72Get a decent editor and don't leave whitespace at the end of lines.
73
74
75 Chapter 2: Breaking long lines and strings
76
77Coding style is all about readability and maintainability using commonly
78available tools.
79
Alan Coxdff49822007-10-16 23:27:33 -070080The limit on the length of lines is 80 columns and this is a strongly
81preferred limit.
Linus Torvalds1da177e2005-04-16 15:20:36 -070082
Josh Triplett6f76b6f2011-08-03 12:19:07 -070083Statements longer than 80 columns will be broken into sensible chunks, unless
84exceeding 80 columns significantly increases readability and does not hide
85information. Descendants are always substantially shorter than the parent and
86are placed substantially to the right. The same applies to function headers
87with a long argument list. However, never break user-visible strings such as
88printk messages, because that breaks the ability to grep for them.
Linus Torvalds1da177e2005-04-16 15:20:36 -070089
Linus Torvalds1da177e2005-04-16 15:20:36 -070090
Randy Dunlapb3fc9942006-12-10 02:18:56 -080091 Chapter 3: Placing Braces and Spaces
Linus Torvalds1da177e2005-04-16 15:20:36 -070092
93The other issue that always comes up in C styling is the placement of
94braces. Unlike the indent size, there are few technical reasons to
95choose one placement strategy over the other, but the preferred way, as
96shown to us by the prophets Kernighan and Ritchie, is to put the opening
97brace last on the line, and put the closing brace first, thusly:
98
99 if (x is true) {
100 we do y
101 }
102
Randy Dunlapb3fc9942006-12-10 02:18:56 -0800103This applies to all non-function statement blocks (if, switch, for,
104while, do). E.g.:
105
106 switch (action) {
107 case KOBJ_ADD:
108 return "add";
109 case KOBJ_REMOVE:
110 return "remove";
111 case KOBJ_CHANGE:
112 return "change";
113 default:
114 return NULL;
115 }
116
Linus Torvalds1da177e2005-04-16 15:20:36 -0700117However, there is one special case, namely functions: they have the
118opening brace at the beginning of the next line, thus:
119
120 int function(int x)
121 {
122 body of function
123 }
124
125Heretic people all over the world have claimed that this inconsistency
126is ... well ... inconsistent, but all right-thinking people know that
127(a) K&R are _right_ and (b) K&R are right. Besides, functions are
128special anyway (you can't nest them in C).
129
130Note that the closing brace is empty on a line of its own, _except_ in
131the cases where it is followed by a continuation of the same statement,
132ie a "while" in a do-statement or an "else" in an if-statement, like
133this:
134
135 do {
136 body of do-loop
137 } while (condition);
138
139and
140
141 if (x == y) {
142 ..
143 } else if (x > y) {
144 ...
145 } else {
146 ....
147 }
148
149Rationale: K&R.
150
151Also, note that this brace-placement also minimizes the number of empty
152(or almost empty) lines, without any loss of readability. Thus, as the
153supply of new-lines on your screen is not a renewable resource (think
15425-line terminal screens here), you have more empty lines to put
155comments on.
156
Oliver Neukume659ba42007-05-08 00:30:34 -0700157Do not unnecessarily use braces where a single statement will do.
158
159if (condition)
160 action();
161
Harry Wei38829dc2011-03-22 16:35:01 -0700162and
163
164if (condition)
165 do_this();
166else
167 do_that();
168
Antonio Ospiteb218ab02011-11-04 11:22:19 -0700169This does not apply if only one branch of a conditional statement is a single
170statement; in the latter case use braces in both branches:
Oliver Neukume659ba42007-05-08 00:30:34 -0700171
172if (condition) {
173 do_this();
174 do_that();
175} else {
176 otherwise();
177}
178
Randy Dunlapb3fc9942006-12-10 02:18:56 -0800179 3.1: Spaces
180
181Linux kernel style for use of spaces depends (mostly) on
182function-versus-keyword usage. Use a space after (most) keywords. The
183notable exceptions are sizeof, typeof, alignof, and __attribute__, which look
184somewhat like functions (and are usually used with parentheses in Linux,
185although they are not required in the language, as in: "sizeof info" after
186"struct fileinfo info;" is declared).
187
188So use a space after these keywords:
189 if, switch, case, for, do, while
190but not with sizeof, typeof, alignof, or __attribute__. E.g.,
191 s = sizeof(struct file);
192
193Do not add spaces around (inside) parenthesized expressions. This example is
194*bad*:
195
196 s = sizeof( struct file );
197
198When declaring pointer data or a function that returns a pointer type, the
199preferred use of '*' is adjacent to the data name or function name and not
200adjacent to the type name. Examples:
201
202 char *linux_banner;
203 unsigned long long memparse(char *ptr, char **retptr);
204 char *match_strdup(substring_t *s);
205
206Use one space around (on each side of) most binary and ternary operators,
207such as any of these:
208
209 = + - < > * / % | & ^ <= >= == != ? :
210
211but no space after unary operators:
212 & * + - ~ ! sizeof typeof alignof __attribute__ defined
213
214no space before the postfix increment & decrement unary operators:
215 ++ --
216
217no space after the prefix increment & decrement unary operators:
218 ++ --
219
220and no space around the '.' and "->" structure member operators.
221
Josh Tripletta923fd62007-07-15 23:41:37 -0700222Do not leave trailing whitespace at the ends of lines. Some editors with
223"smart" indentation will insert whitespace at the beginning of new lines as
224appropriate, so you can start typing the next line of code right away.
225However, some such editors do not remove the whitespace if you end up not
226putting a line of code there, such as if you leave a blank line. As a result,
227you end up with lines containing trailing whitespace.
228
229Git will warn you about patches that introduce trailing whitespace, and can
230optionally strip the trailing whitespace for you; however, if applying a series
231of patches, this may make later patches in the series fail by changing their
232context lines.
233
Linus Torvalds1da177e2005-04-16 15:20:36 -0700234
235 Chapter 4: Naming
236
237C is a Spartan language, and so should your naming be. Unlike Modula-2
238and Pascal programmers, C programmers do not use cute names like
239ThisVariableIsATemporaryCounter. A C programmer would call that
240variable "tmp", which is much easier to write, and not the least more
241difficult to understand.
242
243HOWEVER, while mixed-case names are frowned upon, descriptive names for
244global variables are a must. To call a global function "foo" is a
245shooting offense.
246
247GLOBAL variables (to be used only if you _really_ need them) need to
248have descriptive names, as do global functions. If you have a function
249that counts the number of active users, you should call that
250"count_active_users()" or similar, you should _not_ call it "cntusr()".
251
252Encoding the type of a function into the name (so-called Hungarian
253notation) is brain damaged - the compiler knows the types anyway and can
254check those, and it only confuses the programmer. No wonder MicroSoft
255makes buggy programs.
256
257LOCAL variable names should be short, and to the point. If you have
258some random integer loop counter, it should probably be called "i".
259Calling it "loop_counter" is non-productive, if there is no chance of it
260being mis-understood. Similarly, "tmp" can be just about any type of
261variable that is used to hold a temporary value.
262
263If you are afraid to mix up your local variable names, you have another
264problem, which is called the function-growth-hormone-imbalance syndrome.
Randy Dunlapb3fc9942006-12-10 02:18:56 -0800265See chapter 6 (Functions).
Linus Torvalds1da177e2005-04-16 15:20:36 -0700266
267
Randy Dunlap226a6b82006-06-23 02:05:58 -0700268 Chapter 5: Typedefs
269
270Please don't use things like "vps_t".
271
272It's a _mistake_ to use typedef for structures and pointers. When you see a
273
274 vps_t a;
275
276in the source, what does it mean?
277
278In contrast, if it says
279
280 struct virtual_container *a;
281
282you can actually tell what "a" is.
283
284Lots of people think that typedefs "help readability". Not so. They are
285useful only for:
286
287 (a) totally opaque objects (where the typedef is actively used to _hide_
288 what the object is).
289
290 Example: "pte_t" etc. opaque objects that you can only access using
291 the proper accessor functions.
292
293 NOTE! Opaqueness and "accessor functions" are not good in themselves.
294 The reason we have them for things like pte_t etc. is that there
295 really is absolutely _zero_ portably accessible information there.
296
297 (b) Clear integer types, where the abstraction _helps_ avoid confusion
298 whether it is "int" or "long".
299
300 u8/u16/u32 are perfectly fine typedefs, although they fit into
301 category (d) better than here.
302
303 NOTE! Again - there needs to be a _reason_ for this. If something is
304 "unsigned long", then there's no reason to do
305
306 typedef unsigned long myflags_t;
307
308 but if there is a clear reason for why it under certain circumstances
309 might be an "unsigned int" and under other configurations might be
310 "unsigned long", then by all means go ahead and use a typedef.
311
312 (c) when you use sparse to literally create a _new_ type for
313 type-checking.
314
315 (d) New types which are identical to standard C99 types, in certain
316 exceptional circumstances.
317
318 Although it would only take a short amount of time for the eyes and
319 brain to become accustomed to the standard types like 'uint32_t',
320 some people object to their use anyway.
321
322 Therefore, the Linux-specific 'u8/u16/u32/u64' types and their
323 signed equivalents which are identical to standard types are
324 permitted -- although they are not mandatory in new code of your
325 own.
326
327 When editing existing code which already uses one or the other set
328 of types, you should conform to the existing choices in that code.
329
330 (e) Types safe for use in userspace.
331
332 In certain structures which are visible to userspace, we cannot
333 require C99 types and cannot use the 'u32' form above. Thus, we
334 use __u32 and similar types in all structures which are shared
335 with userspace.
336
337Maybe there are other cases too, but the rule should basically be to NEVER
338EVER use a typedef unless you can clearly match one of those rules.
339
340In general, a pointer, or a struct that has elements that can reasonably
341be directly accessed should _never_ be a typedef.
342
343
344 Chapter 6: Functions
Linus Torvalds1da177e2005-04-16 15:20:36 -0700345
346Functions should be short and sweet, and do just one thing. They should
347fit on one or two screenfuls of text (the ISO/ANSI screen size is 80x24,
348as we all know), and do one thing and do that well.
349
350The maximum length of a function is inversely proportional to the
351complexity and indentation level of that function. So, if you have a
352conceptually simple function that is just one long (but simple)
353case-statement, where you have to do lots of small things for a lot of
354different cases, it's OK to have a longer function.
355
356However, if you have a complex function, and you suspect that a
357less-than-gifted first-year high-school student might not even
358understand what the function is all about, you should adhere to the
359maximum limits all the more closely. Use helper functions with
360descriptive names (you can ask the compiler to in-line them if you think
361it's performance-critical, and it will probably do a better job of it
362than you would have done).
363
364Another measure of the function is the number of local variables. They
365shouldn't exceed 5-10, or you're doing something wrong. Re-think the
366function, and split it into smaller pieces. A human brain can
367generally easily keep track of about 7 different things, anything more
368and it gets confused. You know you're brilliant, but maybe you'd like
369to understand what you did 2 weeks from now.
370
Randy Dunlapb3fc9942006-12-10 02:18:56 -0800371In source files, separate functions with one blank line. If the function is
372exported, the EXPORT* macro for it should follow immediately after the closing
373function brace line. E.g.:
374
375int system_is_up(void)
376{
377 return system_state == SYSTEM_RUNNING;
378}
379EXPORT_SYMBOL(system_is_up);
380
381In function prototypes, include parameter names with their data types.
382Although this is not required by the C language, it is preferred in Linux
383because it is a simple way to add valuable information for the reader.
384
Linus Torvalds1da177e2005-04-16 15:20:36 -0700385
Randy Dunlap226a6b82006-06-23 02:05:58 -0700386 Chapter 7: Centralized exiting of functions
Linus Torvalds1da177e2005-04-16 15:20:36 -0700387
388Albeit deprecated by some people, the equivalent of the goto statement is
389used frequently by compilers in form of the unconditional jump instruction.
390
391The goto statement comes in handy when a function exits from multiple
Dan Carpenterb57a0502013-07-03 15:08:08 -0700392locations and some common work such as cleanup has to be done. If there is no
393cleanup needed then just return directly.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700394
395The rationale is:
396
397- unconditional statements are easier to understand and follow
398- nesting is reduced
399- errors by not updating individual exit points when making
400 modifications are prevented
401- saves the compiler work to optimize redundant code away ;)
402
Jesper Juhldc3d28d2006-01-09 20:53:51 -0800403int fun(int a)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700404{
405 int result = 0;
406 char *buffer = kmalloc(SIZE);
407
408 if (buffer == NULL)
409 return -ENOMEM;
410
411 if (condition1) {
412 while (loop1) {
413 ...
414 }
415 result = 1;
416 goto out;
417 }
418 ...
419out:
420 kfree(buffer);
421 return result;
422}
423
Randy Dunlap226a6b82006-06-23 02:05:58 -0700424 Chapter 8: Commenting
Linus Torvalds1da177e2005-04-16 15:20:36 -0700425
426Comments are good, but there is also a danger of over-commenting. NEVER
427try to explain HOW your code works in a comment: it's much better to
428write the code so that the _working_ is obvious, and it's a waste of
429time to explain badly written code.
430
431Generally, you want your comments to tell WHAT your code does, not HOW.
432Also, try to avoid putting comments inside a function body: if the
433function is so complex that you need to separately comment parts of it,
Randy Dunlapb3fc9942006-12-10 02:18:56 -0800434you should probably go back to chapter 6 for a while. You can make
Linus Torvalds1da177e2005-04-16 15:20:36 -0700435small comments to note or warn about something particularly clever (or
436ugly), but try to avoid excess. Instead, put the comments at the head
437of the function, telling people what it does, and possibly WHY it does
438it.
439
Randy Dunlapb3fc9942006-12-10 02:18:56 -0800440When commenting the kernel API functions, please use the kernel-doc format.
Pekka J Enberge776eba2005-09-10 00:26:44 -0700441See the files Documentation/kernel-doc-nano-HOWTO.txt and scripts/kernel-doc
442for details.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700443
Randy Dunlapb3fc9942006-12-10 02:18:56 -0800444Linux style for comments is the C89 "/* ... */" style.
445Don't use C99-style "// ..." comments.
446
447The preferred style for long (multi-line) comments is:
448
449 /*
450 * This is the preferred style for multi-line
451 * comments in the Linux kernel source code.
452 * Please use it consistently.
453 *
454 * Description: A column of asterisks on the left side,
455 * with beginning and ending almost-blank lines.
456 */
457
Joe Perchesc4ff1b52012-10-04 17:13:36 -0700458For files in net/ and drivers/net/ the preferred style for long (multi-line)
459comments is a little different.
460
461 /* The preferred comment style for files in net/ and drivers/net
462 * looks like this.
463 *
464 * It is nearly the same as the generally preferred comment style,
465 * but there is no initial almost-blank line.
466 */
467
Randy Dunlapb3fc9942006-12-10 02:18:56 -0800468It's also important to comment data, whether they are basic types or derived
469types. To this end, use just one data declaration per line (no commas for
470multiple data declarations). This leaves you room for a small comment on each
471item, explaining its use.
472
473
Randy Dunlap226a6b82006-06-23 02:05:58 -0700474 Chapter 9: You've made a mess of it
Linus Torvalds1da177e2005-04-16 15:20:36 -0700475
476That's OK, we all do. You've probably been told by your long-time Unix
477user helper that "GNU emacs" automatically formats the C sources for
478you, and you've noticed that yes, it does do that, but the defaults it
479uses are less than desirable (in fact, they are worse than random
480typing - an infinite number of monkeys typing into GNU emacs would never
481make a good program).
482
483So, you can either get rid of GNU emacs, or change it to use saner
484values. To do the latter, you can stick the following in your .emacs file:
485
Johannes Weinera7f371e2008-07-25 01:45:51 -0700486(defun c-lineup-arglist-tabs-only (ignored)
487 "Line up argument lists by tabs, not spaces"
488 (let* ((anchor (c-langelem-pos c-syntactic-element))
489 (column (c-langelem-2nd-pos c-syntactic-element))
490 (offset (- (1+ column) anchor))
491 (steps (floor offset c-basic-offset)))
492 (* (max steps 1)
493 c-basic-offset)))
Linus Torvalds1da177e2005-04-16 15:20:36 -0700494
Teemu Likonen0acbc6c2009-01-29 16:28:16 -0800495(add-hook 'c-mode-common-hook
496 (lambda ()
497 ;; Add kernel style
498 (c-add-style
499 "linux-tabs-only"
500 '("linux" (c-offsets-alist
501 (arglist-cont-nonempty
502 c-lineup-gcc-asm-reg
503 c-lineup-arglist-tabs-only))))))
504
Johannes Weinera7f371e2008-07-25 01:45:51 -0700505(add-hook 'c-mode-hook
506 (lambda ()
507 (let ((filename (buffer-file-name)))
508 ;; Enable kernel mode for the appropriate files
509 (when (and filename
Dan Carpenter70221392009-01-29 16:28:28 -0800510 (string-match (expand-file-name "~/src/linux-trees")
511 filename))
Johannes Weinera7f371e2008-07-25 01:45:51 -0700512 (setq indent-tabs-mode t)
Teemu Likonen0acbc6c2009-01-29 16:28:16 -0800513 (c-set-style "linux-tabs-only")))))
Linus Torvalds1da177e2005-04-16 15:20:36 -0700514
Johannes Weinera7f371e2008-07-25 01:45:51 -0700515This will make emacs go better with the kernel coding style for C
516files below ~/src/linux-trees.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700517
518But even if you fail in getting emacs to do sane formatting, not
519everything is lost: use "indent".
520
521Now, again, GNU indent has the same brain-dead settings that GNU emacs
522has, which is why you need to give it a few command line options.
523However, that's not too bad, because even the makers of GNU indent
524recognize the authority of K&R (the GNU people aren't evil, they are
525just severely misguided in this matter), so you just give indent the
526options "-kr -i8" (stands for "K&R, 8 character indents"), or use
527"scripts/Lindent", which indents in the latest style.
528
529"indent" has a lot of options, and especially when it comes to comment
530re-formatting you may want to take a look at the man page. But
531remember: "indent" is not a fix for bad programming.
532
533
Robert P. J. Day6754bb42007-05-23 13:57:42 -0700534 Chapter 10: Kconfig configuration files
Linus Torvalds1da177e2005-04-16 15:20:36 -0700535
Robert P. J. Day6754bb42007-05-23 13:57:42 -0700536For all of the Kconfig* configuration files throughout the source tree,
537the indentation is somewhat different. Lines under a "config" definition
538are indented with one tab, while help text is indented an additional two
539spaces. Example:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700540
Robert P. J. Day6754bb42007-05-23 13:57:42 -0700541config AUDIT
542 bool "Auditing support"
543 depends on NET
Linus Torvalds1da177e2005-04-16 15:20:36 -0700544 help
Robert P. J. Day6754bb42007-05-23 13:57:42 -0700545 Enable auditing infrastructure that can be used with another
546 kernel subsystem, such as SELinux (which requires this for
547 logging of avc messages output). Does not do system-call
548 auditing without CONFIG_AUDITSYSCALL.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700549
Kees Cook0335cb42012-10-02 11:16:15 -0700550Seriously dangerous features (such as write support for certain
Robert P. J. Day6754bb42007-05-23 13:57:42 -0700551filesystems) should advertise this prominently in their prompt string:
552
553config ADFS_FS_RW
554 bool "ADFS write support (DANGEROUS)"
555 depends on ADFS_FS
556 ...
557
558For full documentation on the configuration files, see the file
559Documentation/kbuild/kconfig-language.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700560
561
Randy Dunlap226a6b82006-06-23 02:05:58 -0700562 Chapter 11: Data structures
Linus Torvalds1da177e2005-04-16 15:20:36 -0700563
564Data structures that have visibility outside the single-threaded
565environment they are created and destroyed in should always have
566reference counts. In the kernel, garbage collection doesn't exist (and
567outside the kernel garbage collection is slow and inefficient), which
568means that you absolutely _have_ to reference count all your uses.
569
570Reference counting means that you can avoid locking, and allows multiple
571users to have access to the data structure in parallel - and not having
572to worry about the structure suddenly going away from under them just
573because they slept or did something else for a while.
574
575Note that locking is _not_ a replacement for reference counting.
576Locking is used to keep data structures coherent, while reference
577counting is a memory management technique. Usually both are needed, and
578they are not to be confused with each other.
579
580Many data structures can indeed have two levels of reference counting,
581when there are users of different "classes". The subclass count counts
582the number of subclass users, and decrements the global count just once
583when the subclass count goes to zero.
584
585Examples of this kind of "multi-level-reference-counting" can be found in
586memory management ("struct mm_struct": mm_users and mm_count), and in
587filesystem code ("struct super_block": s_count and s_active).
588
589Remember: if another thread can find your data structure, and you don't
590have a reference count on it, you almost certainly have a bug.
591
592
Randy Dunlap226a6b82006-06-23 02:05:58 -0700593 Chapter 12: Macros, Enums and RTL
Linus Torvalds1da177e2005-04-16 15:20:36 -0700594
595Names of macros defining constants and labels in enums are capitalized.
596
597#define CONSTANT 0x12345
598
599Enums are preferred when defining several related constants.
600
601CAPITALIZED macro names are appreciated but macros resembling functions
602may be named in lower case.
603
604Generally, inline functions are preferable to macros resembling functions.
605
606Macros with multiple statements should be enclosed in a do - while block:
607
608#define macrofun(a, b, c) \
609 do { \
610 if (a == 5) \
611 do_this(b, c); \
612 } while (0)
613
614Things to avoid when using macros:
615
6161) macros that affect control flow:
617
618#define FOO(x) \
619 do { \
620 if (blah(x) < 0) \
621 return -EBUGGERED; \
622 } while(0)
623
624is a _very_ bad idea. It looks like a function call but exits the "calling"
625function; don't break the internal parsers of those who will read the code.
626
6272) macros that depend on having a local variable with a magic name:
628
629#define FOO(val) bar(index, val)
630
631might look like a good thing, but it's confusing as hell when one reads the
632code and it's prone to breakage from seemingly innocent changes.
633
6343) macros with arguments that are used as l-values: FOO(x) = y; will
635bite you if somebody e.g. turns FOO into an inline function.
636
6374) forgetting about precedence: macros defining constants using expressions
638must enclose the expression in parentheses. Beware of similar issues with
639macros using parameters.
640
641#define CONSTANT 0x4000
642#define CONSTEXP (CONSTANT | 3)
643
644The cpp manual deals with macros exhaustively. The gcc internals manual also
645covers RTL which is used frequently with assembly language in the kernel.
646
647
Randy Dunlap226a6b82006-06-23 02:05:58 -0700648 Chapter 13: Printing kernel messages
Linus Torvalds1da177e2005-04-16 15:20:36 -0700649
650Kernel developers like to be seen as literate. Do mind the spelling
651of kernel messages to make a good impression. Do not use crippled
David Brownell6b094482007-07-13 16:32:09 -0700652words like "dont"; use "do not" or "don't" instead. Make the messages
653concise, clear, and unambiguous.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700654
655Kernel messages do not have to be terminated with a period.
656
657Printing numbers in parentheses (%d) adds no value and should be avoided.
658
David Brownell6b094482007-07-13 16:32:09 -0700659There are a number of driver model diagnostic macros in <linux/device.h>
660which you should use to make sure messages are matched to the right device
661and driver, and are tagged with the right level: dev_err(), dev_warn(),
662dev_info(), and so forth. For messages that aren't associated with a
Dan Streetman6e099f52014-06-04 16:11:44 -0700663particular device, <linux/printk.h> defines pr_notice(), pr_info(),
664pr_warn(), pr_err(), etc.
David Brownell6b094482007-07-13 16:32:09 -0700665
666Coming up with good debugging messages can be quite a challenge; and once
Dan Streetman6e099f52014-06-04 16:11:44 -0700667you have them, they can be a huge help for remote troubleshooting. However
668debug message printing is handled differently than printing other non-debug
669messages. While the other pr_XXX() functions print unconditionally,
670pr_debug() does not; it is compiled out by default, unless either DEBUG is
671defined or CONFIG_DYNAMIC_DEBUG is set. That is true for dev_dbg() also,
672and a related convention uses VERBOSE_DEBUG to add dev_vdbg() messages to
673the ones already enabled by DEBUG.
674
675Many subsystems have Kconfig debug options to turn on -DDEBUG in the
676corresponding Makefile; in other cases specific files #define DEBUG. And
677when a debug message should be unconditionally printed, such as if it is
Raymond L. Rivera7c18fd72014-07-24 02:39:44 -0700678already inside a debug-related #ifdef section, printk(KERN_DEBUG ...) can be
Dan Streetman6e099f52014-06-04 16:11:44 -0700679used.
David Brownell6b094482007-07-13 16:32:09 -0700680
Linus Torvalds1da177e2005-04-16 15:20:36 -0700681
Randy Dunlap226a6b82006-06-23 02:05:58 -0700682 Chapter 14: Allocating memory
Pekka J Enbergaf4e5a22005-09-16 19:28:11 -0700683
684The kernel provides the following general purpose memory allocators:
Xi Wang158372942012-05-31 16:26:04 -0700685kmalloc(), kzalloc(), kmalloc_array(), kcalloc(), vmalloc(), and
686vzalloc(). Please refer to the API documentation for further information
687about them.
Pekka J Enbergaf4e5a22005-09-16 19:28:11 -0700688
689The preferred form for passing a size of a struct is the following:
690
691 p = kmalloc(sizeof(*p), ...);
692
693The alternative form where struct name is spelled out hurts readability and
694introduces an opportunity for a bug when the pointer variable type is changed
695but the corresponding sizeof that is passed to a memory allocator is not.
696
697Casting the return value which is a void pointer is redundant. The conversion
698from void pointer to any other pointer type is guaranteed by the C programming
699language.
700
Xi Wang158372942012-05-31 16:26:04 -0700701The preferred form for allocating an array is the following:
702
703 p = kmalloc_array(n, sizeof(...), ...);
704
705The preferred form for allocating a zeroed array is the following:
706
707 p = kcalloc(n, sizeof(...), ...);
708
709Both forms check for overflow on the allocation size n * sizeof(...),
710and return NULL if that occurred.
711
Pekka J Enbergaf4e5a22005-09-16 19:28:11 -0700712
Randy Dunlap226a6b82006-06-23 02:05:58 -0700713 Chapter 15: The inline disease
Arjan van de Vena771f2b2006-01-08 01:05:04 -0800714
715There appears to be a common misperception that gcc has a magic "make me
716faster" speedup option called "inline". While the use of inlines can be
Jesper Juhl53ab97a2007-05-08 00:31:06 -0700717appropriate (for example as a means of replacing macros, see Chapter 12), it
Arjan van de Vena771f2b2006-01-08 01:05:04 -0800718very often is not. Abundant use of the inline keyword leads to a much bigger
719kernel, which in turn slows the system as a whole down, due to a bigger
720icache footprint for the CPU and simply because there is less memory
721available for the pagecache. Just think about it; a pagecache miss causes a
Martin Olsson19af5cd2009-04-23 11:37:37 +0200722disk seek, which easily takes 5 milliseconds. There are a LOT of cpu cycles
723that can go into these 5 milliseconds.
Arjan van de Vena771f2b2006-01-08 01:05:04 -0800724
725A reasonable rule of thumb is to not put inline at functions that have more
726than 3 lines of code in them. An exception to this rule are the cases where
727a parameter is known to be a compiletime constant, and as a result of this
728constantness you *know* the compiler will be able to optimize most of your
729function away at compile time. For a good example of this later case, see
730the kmalloc() inline function.
731
732Often people argue that adding inline to functions that are static and used
733only once is always a win since there is no space tradeoff. While this is
734technically correct, gcc is capable of inlining these automatically without
735help, and the maintenance issue of removing the inline when a second user
736appears outweighs the potential value of the hint that tells gcc to do
737something it would have done anyway.
738
739
Alan Sternc16a02d2006-09-29 02:01:21 -0700740 Chapter 16: Function return values and names
741
742Functions can return values of many different kinds, and one of the
743most common is a value indicating whether the function succeeded or
744failed. Such a value can be represented as an error-code integer
745(-Exxx = failure, 0 = success) or a "succeeded" boolean (0 = failure,
746non-zero = success).
747
748Mixing up these two sorts of representations is a fertile source of
749difficult-to-find bugs. If the C language included a strong distinction
750between integers and booleans then the compiler would find these mistakes
751for us... but it doesn't. To help prevent such bugs, always follow this
752convention:
753
754 If the name of a function is an action or an imperative command,
755 the function should return an error-code integer. If the name
756 is a predicate, the function should return a "succeeded" boolean.
757
758For example, "add work" is a command, and the add_work() function returns 0
759for success or -EBUSY for failure. In the same way, "PCI device present" is
760a predicate, and the pci_dev_present() function returns 1 if it succeeds in
761finding a matching device or 0 if it doesn't.
762
763All EXPORTed functions must respect this convention, and so should all
764public functions. Private (static) functions need not, but it is
765recommended that they do.
766
767Functions whose return value is the actual result of a computation, rather
768than an indication of whether the computation succeeded, are not subject to
769this rule. Generally they indicate failure by returning some out-of-range
770result. Typical examples would be functions that return pointers; they use
771NULL or the ERR_PTR mechanism to report failure.
772
773
Robert P. J. Day58637ec2006-12-22 01:09:11 -0800774 Chapter 17: Don't re-invent the kernel macros
775
776The header file include/linux/kernel.h contains a number of macros that
777you should use, rather than explicitly coding some variant of them yourself.
778For example, if you need to calculate the length of an array, take advantage
779of the macro
780
781 #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
782
783Similarly, if you need to calculate the size of some structure member, use
784
785 #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f))
786
787There are also min() and max() macros that do strict type checking if you
788need them. Feel free to peruse that header file to see what else is already
789defined that you shouldn't reproduce in your code.
790
791
Josh Triplett4e7bd662007-07-15 23:41:37 -0700792 Chapter 18: Editor modelines and other cruft
793
794Some editors can interpret configuration information embedded in source files,
795indicated with special markers. For example, emacs interprets lines marked
796like this:
797
798-*- mode: c -*-
799
800Or like this:
801
802/*
803Local Variables:
804compile-command: "gcc -DMAGIC_DEBUG_FLAG foo.c"
805End:
806*/
807
808Vim interprets markers that look like this:
809
810/* vim:set sw=8 noet */
811
812Do not include any of these in source files. People have their own personal
813editor configurations, and your source files should not override them. This
814includes markers for indentation and mode configuration. People may use their
815own custom mode, or may have some other magic method for making indentation
816work correctly.
817
818
Josh Triplett9a7c48b2012-03-30 13:37:10 -0700819 Chapter 19: Inline assembly
820
821In architecture-specific code, you may need to use inline assembly to interface
822with CPU or platform functionality. Don't hesitate to do so when necessary.
823However, don't use inline assembly gratuitously when C can do the job. You can
824and should poke hardware from C when possible.
825
826Consider writing simple helper functions that wrap common bits of inline
827assembly, rather than repeatedly writing them with slight variations. Remember
828that inline assembly can use C parameters.
829
830Large, non-trivial assembly functions should go in .S files, with corresponding
831C prototypes defined in C header files. The C prototypes for assembly
832functions should use "asmlinkage".
833
834You may need to mark your asm statement as volatile, to prevent GCC from
835removing it if GCC doesn't notice any side effects. You don't always need to
836do so, though, and doing so unnecessarily can limit optimization.
837
838When writing a single inline assembly statement containing multiple
839instructions, put each instruction on a separate line in a separate quoted
840string, and end each string except the last with \n\t to properly indent the
841next instruction in the assembly output:
842
843 asm ("magic %reg1, #42\n\t"
844 "more_magic %reg2, %reg3"
845 : /* outputs */ : /* inputs */ : /* clobbers */);
846
847
Arjan van de Vena771f2b2006-01-08 01:05:04 -0800848
Randy Dunlap226a6b82006-06-23 02:05:58 -0700849 Appendix I: References
Linus Torvalds1da177e2005-04-16 15:20:36 -0700850
851The C Programming Language, Second Edition
852by Brian W. Kernighan and Dennis M. Ritchie.
853Prentice Hall, Inc., 1988.
854ISBN 0-13-110362-8 (paperback), 0-13-110370-9 (hardback).
855URL: http://cm.bell-labs.com/cm/cs/cbook/
856
857The Practice of Programming
858by Brian W. Kernighan and Rob Pike.
859Addison-Wesley, Inc., 1999.
860ISBN 0-201-61586-X.
861URL: http://cm.bell-labs.com/cm/cs/tpop/
862
863GNU manuals - where in compliance with K&R and this text - for cpp, gcc,
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800864gcc internals and indent, all available from http://www.gnu.org/manual/
Linus Torvalds1da177e2005-04-16 15:20:36 -0700865
866WG14 is the international standardization working group for the programming
Xose Vazquez Perez5b0ed2c2006-01-08 01:02:49 -0800867language C, URL: http://www.open-std.org/JTC1/SC22/WG14/
868
869Kernel CodingStyle, by greg@kroah.com at OLS 2002:
870http://www.kroah.com/linux/talks/ols_2002_kernel_codingstyle_talk/html/
Linus Torvalds1da177e2005-04-16 15:20:36 -0700871