8 The configuration database is a collection of configuration options
9 organized in a tree structure::
11 +- Code maturity level options
12 | +- Prompt for development and/or incomplete code/drivers
14 | +- Networking support
16 | +- BSD Process Accounting
18 +- Loadable module support
19 | +- Enable loadable module support
20 | +- Set version information on all module symbols
21 | +- Kernel module loader
24 Every entry has its own dependencies. These dependencies are used
25 to determine the visibility of an entry. Any child entry is only
26 visible if its parent entry is also visible.
31 Most entries define a config option; all other entries help to organize
32 them. A single configuration option is defined like this::
35 bool "Set version information on all module symbols"
38 Usually, modules have to be recompiled whenever you switch to a new
41 Every line starts with a key word and can be followed by multiple
42 arguments. "config" starts a new config entry. The following lines
43 define attributes for this config option. Attributes can be the type of
44 the config option, input prompt, dependencies, help text and default
45 values. A config option can be defined multiple times with the same
46 name, but every definition can have only a single input prompt and the
47 type must not conflict.
52 A menu entry can have a number of attributes. Not all of them are
53 applicable everywhere (see syntax).
55 - type definition: "bool"/"tristate"/"string"/"hex"/"int"
57 Every config option must have a type. There are only two basic types:
58 tristate and string; the other types are based on these two. The type
59 definition optionally accepts an input prompt, so these two examples
62 bool "Networking support"
67 prompt "Networking support"
69 - input prompt: "prompt" <prompt> ["if" <expr>]
71 Every menu entry can have at most one prompt, which is used to display
72 to the user. Optionally dependencies only for this prompt can be added
75 - default value: "default" <expr> ["if" <expr>]
77 A config option can have any number of default values. If multiple
78 default values are visible, only the first defined one is active.
79 Default values are not limited to the menu entry where they are
80 defined. This means the default can be defined somewhere else or be
81 overridden by an earlier definition.
82 The default value is only assigned to the config symbol if no other
83 value was set by the user (via the input prompt above). If an input
84 prompt is visible the default value is presented to the user and can
86 Optionally, dependencies only for this default value can be added with
89 The default value deliberately defaults to 'n' in order to avoid bloating the
90 build. With few exceptions, new config options should not change this. The
91 intent is for "make oldconfig" to add as little as possible to the config from
95 Things that merit "default y/m" include:
97 a) A new Kconfig option for something that used to always be built
98 should be "default y".
100 b) A new gatekeeping Kconfig option that hides/shows other Kconfig
101 options (but does not generate any code of its own), should be
102 "default y" so people will see those other options.
104 c) Sub-driver behavior or similar options for a driver that is
105 "default n". This allows you to provide sane defaults.
107 d) Hardware or infrastructure that everybody expects, such as CONFIG_NET
108 or CONFIG_BLOCK. These are rare exceptions.
110 - type definition + default value::
112 "def_bool"/"def_tristate" <expr> ["if" <expr>]
114 This is a shorthand notation for a type definition plus a value.
115 Optionally dependencies for this default value can be added with "if".
117 - dependencies: "depends on" <expr>
119 This defines a dependency for this menu entry. If multiple
120 dependencies are defined, they are connected with '&&'. Dependencies
121 are applied to all other options within this menu entry (which also
122 accept an "if" expression), so these two examples are equivalent::
133 - reverse dependencies: "select" <symbol> ["if" <expr>]
135 While normal dependencies reduce the upper limit of a symbol (see
136 below), reverse dependencies can be used to force a lower limit of
137 another symbol. The value of the current menu symbol is used as the
138 minimal value <symbol> can be set to. If <symbol> is selected multiple
139 times, the limit is set to the largest selection.
140 Reverse dependencies can only be used with boolean or tristate
144 select should be used with care. select will force
145 a symbol to a value without visiting the dependencies.
146 By abusing select you are able to select a symbol FOO even
147 if FOO depends on BAR that is not set.
148 In general use select only for non-visible symbols
149 (no prompts anywhere) and for symbols with no dependencies.
150 That will limit the usefulness but on the other hand avoid
151 the illegal configurations all over.
153 If "select" <symbol> is followed by "if" <expr>, <symbol> will be
154 selected by the logical AND of the value of the current menu symbol
155 and <expr>. This means, the lower limit can be downgraded due to the
156 presence of "if" <expr>. This behavior may seem weird, but we rely on
157 it. (The future of this behavior is undecided.)
159 - weak reverse dependencies: "imply" <symbol> ["if" <expr>]
161 This is similar to "select" as it enforces a lower limit on another
162 symbol except that the "implied" symbol's value may still be set to n
163 from a direct dependency or with a visible prompt.
165 Given the following example::
175 The following values are possible:
177 === === ============= ==============
178 FOO BAR BAZ's default choice for BAZ
179 === === ============= ==============
187 === === ============= ==============
189 This is useful e.g. with multiple drivers that want to indicate their
190 ability to hook into a secondary subsystem while allowing the user to
191 configure that subsystem out without also having to unset these drivers.
193 Note: If the combination of FOO=y and BAZ=m causes a link error,
194 you can guard the function call with IS_REACHABLE()::
198 if (IS_REACHABLE(CONFIG_BAZ))
203 Note: If the feature provided by BAZ is highly desirable for FOO,
204 FOO should imply not only BAZ, but also its dependency BAR::
211 Note: If "imply" <symbol> is followed by "if" <expr>, the default of <symbol>
212 will be the logical AND of the value of the current menu symbol and <expr>.
213 (The future of this behavior is undecided.)
215 - limiting menu display: "visible if" <expr>
217 This attribute is only applicable to menu blocks, if the condition is
218 false, the menu block is not displayed to the user (the symbols
219 contained there can still be selected by other symbols, though). It is
220 similar to a conditional "prompt" attribute for individual menu
221 entries. Default value of "visible" is true.
223 - numerical ranges: "range" <symbol> <symbol> ["if" <expr>]
225 This allows to limit the range of possible input values for int
226 and hex symbols. The user can only input a value which is larger than
227 or equal to the first symbol and smaller than or equal to the second
232 This defines a help text. The end of the help text is determined by
233 the indentation level, this means it ends at the first line which has
234 a smaller indentation than the first line of the help text.
236 - module attribute: "modules"
237 This declares the symbol to be used as the MODULES symbol, which
238 enables the third modular state for all config symbols.
239 At most one symbol may have the "modules" option set.
244 Dependencies define the visibility of a menu entry and can also reduce
245 the input range of tristate symbols. The tristate logic used in the
246 expressions uses one more state than normal boolean logic to express the
247 module state. Dependency expressions have the following syntax::
249 <expr> ::= <symbol> (1)
250 <symbol> '=' <symbol> (2)
251 <symbol> '!=' <symbol> (3)
252 <symbol1> '<' <symbol2> (4)
253 <symbol1> '>' <symbol2> (4)
254 <symbol1> '<=' <symbol2> (4)
255 <symbol1> '>=' <symbol2> (4)
258 <expr> '&&' <expr> (7)
259 <expr> '||' <expr> (8)
261 Expressions are listed in decreasing order of precedence.
263 (1) Convert the symbol into an expression. Boolean and tristate symbols
264 are simply converted into the respective expression values. All
265 other symbol types result in 'n'.
266 (2) If the values of both symbols are equal, it returns 'y',
268 (3) If the values of both symbols are equal, it returns 'n',
270 (4) If value of <symbol1> is respectively lower, greater, lower-or-equal,
271 or greater-or-equal than value of <symbol2>, it returns 'y',
273 (5) Returns the value of the expression. Used to override precedence.
274 (6) Returns the result of (2-/expr/).
275 (7) Returns the result of min(/expr/, /expr/).
276 (8) Returns the result of max(/expr/, /expr/).
278 An expression can have a value of 'n', 'm' or 'y' (or 0, 1, 2
279 respectively for calculations). A menu entry becomes visible when its
280 expression evaluates to 'm' or 'y'.
282 There are two types of symbols: constant and non-constant symbols.
283 Non-constant symbols are the most common ones and are defined with the
284 'config' statement. Non-constant symbols consist entirely of alphanumeric
285 characters or underscores.
286 Constant symbols are only part of expressions. Constant symbols are
287 always surrounded by single or double quotes. Within the quote, any
288 other character is allowed and the quotes can be escaped using '\'.
293 The position of a menu entry in the tree is determined in two ways. First
294 it can be specified explicitly::
296 menu "Network device support"
304 All entries within the "menu" ... "endmenu" block become a submenu of
305 "Network device support". All subentries inherit the dependencies from
306 the menu entry, e.g. this means the dependency "NET" is added to the
307 dependency list of the config option NETDEVICES.
309 The other way to generate the menu structure is done by analyzing the
310 dependencies. If a menu entry somehow depends on the previous entry, it
311 can be made a submenu of it. First, the previous (parent) symbol must
312 be part of the dependency list and then one of these two conditions
315 - the child entry must become invisible, if the parent is set to 'n'
316 - the child entry must only be visible, if the parent is visible::
319 bool "Enable loadable module support"
322 bool "Set version information on all module symbols"
325 comment "module support disabled"
328 MODVERSIONS directly depends on MODULES, this means it's only visible if
329 MODULES is different from 'n'. The comment on the other hand is only
330 visible when MODULES is set to 'n'.
336 The configuration file describes a series of menu entries, where every
337 line starts with a keyword (except help texts). The following keywords
348 The first five also start the definition of a menu entry.
355 This defines a config symbol <symbol> and accepts any of above
356 attributes as options.
360 "menuconfig" <symbol>
363 This is similar to the simple config entry above, but it also gives a
364 hint to front ends, that all suboptions should be displayed as a
365 separate list of options. To make sure all the suboptions will really
366 show up under the menuconfig entry and not outside of it, every item
367 from the <config options> list must depend on the menuconfig symbol.
368 In practice, this is achieved by using one of the next two constructs::
384 In the following examples (3) and (4), C1 and C2 still have the M
385 dependency, but will not appear under menuconfig M anymore, because
386 of C0, which doesn't depend on M::
411 This defines a choice group and accepts any of the above attributes as
414 A choice only allows a single config entry to be selected.
421 This defines a comment which is displayed to the user during the
422 configuration process and is also echoed to the output files. The only
423 possible options are dependencies.
432 This defines a menu block, see "Menu structure" above for more
433 information. The only possible options are dependencies and "visible"
442 This defines an if block. The dependency expression <expr> is appended
443 to all enclosed menu entries.
449 This reads the specified configuration file. This file is always parsed.
455 This sets the config program's title bar if the config program chooses
456 to use it. It should be placed at the top of the configuration, before any
459 '#' Kconfig source file comment:
461 An unquoted '#' character anywhere in a source file line indicates
462 the beginning of a source file comment. The remainder of that line
468 This is a collection of Kconfig tips, most of which aren't obvious at
469 first glance and most of which have become idioms in several Kconfig
472 Adding common features and make the usage configurable
473 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
474 It is a common idiom to implement a feature/functionality that are
475 relevant for some architectures but not all.
476 The recommended way to do so is to use a config variable named HAVE_*
477 that is defined in a common Kconfig file and selected by the relevant
479 An example is the generic IOMAP functionality.
481 We would in lib/Kconfig see::
483 # Generic IOMAP is used to ...
484 config HAVE_GENERIC_IOMAP
487 depends on HAVE_GENERIC_IOMAP && FOO
489 And in lib/Makefile we would see::
491 obj-$(CONFIG_GENERIC_IOMAP) += iomap.o
493 For each architecture using the generic IOMAP functionality we would see::
497 select HAVE_GENERIC_IOMAP
500 Note: we use the existing config option and avoid creating a new
501 config variable to select HAVE_GENERIC_IOMAP.
503 Note: the use of the internal config variable HAVE_GENERIC_IOMAP, it is
504 introduced to overcome the limitation of select which will force a
505 config option to 'y' no matter the dependencies.
506 The dependencies are moved to the symbol GENERIC_IOMAP and we avoid the
507 situation where select forces a symbol equals to 'y'.
509 Adding features that need compiler support
510 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
512 There are several features that need compiler support. The recommended way
513 to describe the dependency on the compiler feature is to use "depends on"
514 followed by a test macro::
516 config STACKPROTECTOR
517 bool "Stack Protector buffer overflow detection"
518 depends on $(cc-option,-fstack-protector)
521 If you need to expose a compiler capability to makefiles and/or C source files,
522 `CC_HAS_` is the recommended prefix for the config option::
525 def_bool $(success,$(srctree)/scripts/cc-check-foo.sh $(CC))
529 To restrict a component build to module-only, qualify its config symbol
530 with "depends on m". E.g.::
535 limits FOO to module (=m) or disabled (=n).
539 If a config symbol has a dependency, but the code controlled by the config
540 symbol can still be compiled if the dependency is not met, it is encouraged to
541 increase build coverage by adding an "|| COMPILE_TEST" clause to the
542 dependency. This is especially useful for drivers for more exotic hardware, as
543 it allows continuous-integration systems to compile-test the code on a more
544 common system, and detect bugs that way.
545 Note that compile-tested code should avoid crashing when run on a system where
546 the dependency is not met.
548 Architecture and platform dependencies
549 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
550 Due to the presence of stubs, most drivers can now be compiled on most
551 architectures. However, this does not mean it makes sense to have all drivers
552 available everywhere, as the actual hardware may only exist on specific
553 architectures and platforms. This is especially true for on-SoC IP cores,
554 which may be limited to a specific vendor or SoC family.
556 To prevent asking the user about drivers that cannot be used on the system(s)
557 the user is compiling a kernel for, and if it makes sense, config symbols
558 controlling the compilation of a driver should contain proper dependencies,
559 limiting the visibility of the symbol to (a superset of) the platform(s) the
560 driver can be used on. The dependency can be an architecture (e.g. ARM) or
561 platform (e.g. ARCH_OMAP4) dependency. This makes life simpler not only for
562 distro config owners, but also for every single developer or user who
565 Such a dependency can be relaxed by combining it with the compile-testing rule
569 bool "Support for foo hardware"
570 depends on ARCH_FOO_VENDOR || COMPILE_TEST
572 Optional dependencies
573 ~~~~~~~~~~~~~~~~~~~~~
575 Some drivers are able to optionally use a feature from another module
576 or build cleanly with that module disabled, but cause a link failure
577 when trying to use that loadable module from a built-in driver.
579 The most common way to express this optional dependency in Kconfig logic
580 uses the slightly counterintuitive::
583 tristate "Support for foo hardware"
584 depends on BAR || !BAR
586 This means that there is either a dependency on BAR that disallows
587 the combination of FOO=y with BAR=m, or BAR is completely disabled.
588 For a more formalized approach if there are multiple drivers that have
589 the same dependency, a helper symbol can be used, like::
592 tristate "Support for foo hardware"
593 depends on BAR_OPTIONAL
596 def_tristate BAR || !BAR
598 Kconfig recursive dependency limitations
599 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
601 If you've hit the Kconfig error: "recursive dependency detected" you've run
602 into a recursive dependency issue with Kconfig, a recursive dependency can be
603 summarized as a circular dependency. The kconfig tools need to ensure that
604 Kconfig files comply with specified configuration requirements. In order to do
605 that kconfig must determine the values that are possible for all Kconfig
606 symbols, this is currently not possible if there is a circular relation
607 between two or more Kconfig symbols. For more details refer to the "Simple
608 Kconfig recursive issue" subsection below. Kconfig does not do recursive
609 dependency resolution; this has a few implications for Kconfig file writers.
610 We'll first explain why this issues exists and then provide an example
611 technical limitation which this brings upon Kconfig developers. Eager
612 developers wishing to try to address this limitation should read the next
615 Simple Kconfig recursive issue
616 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
618 Read: Documentation/kbuild/Kconfig.recursion-issue-01
622 make KBUILD_KCONFIG=Documentation/kbuild/Kconfig.recursion-issue-01 allnoconfig
624 Cumulative Kconfig recursive issue
625 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
627 Read: Documentation/kbuild/Kconfig.recursion-issue-02
631 make KBUILD_KCONFIG=Documentation/kbuild/Kconfig.recursion-issue-02 allnoconfig
633 Practical solutions to kconfig recursive issue
634 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
636 Developers who run into the recursive Kconfig issue have two options
637 at their disposal. We document them below and also provide a list of
638 historical issues resolved through these different solutions.
640 a) Remove any superfluous "select FOO" or "depends on FOO"
641 b) Match dependency semantics:
643 b1) Swap all "select FOO" to "depends on FOO" or,
645 b2) Swap all "depends on FOO" to "select FOO"
647 The resolution to a) can be tested with the sample Kconfig file
648 Documentation/kbuild/Kconfig.recursion-issue-01 through the removal
649 of the "select CORE" from CORE_BELL_A_ADVANCED as that is implicit already
650 since CORE_BELL_A depends on CORE. At times it may not be possible to remove
651 some dependency criteria, for such cases you can work with solution b).
653 The two different resolutions for b) can be tested in the sample Kconfig file
654 Documentation/kbuild/Kconfig.recursion-issue-02.
656 Below is a list of examples of prior fixes for these types of recursive issues;
657 all errors appear to involve one or more "select" statements and one or more
660 ============ ===================================
662 ============ ===================================
663 06b718c01208 select A -> depends on A
664 c22eacfe82f9 depends on A -> depends on B
665 6a91e854442c select A -> depends on A
666 118c565a8f2e select A -> select B
667 f004e5594705 select A -> depends on A
668 c7861f37b4c6 depends on A -> (null)
669 80c69915e5fb select A -> (null) (1)
670 c2218e26c0d0 select A -> depends on A (1)
671 d6ae99d04e1c select A -> depends on A
672 95ca19cf8cbf select A -> depends on A
673 8f057d7bca54 depends on A -> (null)
674 8f057d7bca54 depends on A -> select A
675 a0701f04846e select A -> depends on A
676 0c8b92f7f259 depends on A -> (null)
677 e4e9e0540928 select A -> depends on A (2)
678 7453ea886e87 depends on A > (null) (1)
679 7b1fff7e4fdf select A -> depends on A
680 86c747d2a4f0 select A -> depends on A
681 d9f9ab51e55e select A -> depends on A
682 0c51a4d8abd6 depends on A -> select A (3)
683 e98062ed6dc4 select A -> depends on A (3)
684 91e5d284a7f1 select A -> (null)
685 ============ ===================================
687 (1) Partial (or no) quote of error.
688 (2) That seems to be the gist of that fix.
694 Work on kconfig is welcomed on both areas of clarifying semantics and on
695 evaluating the use of a full SAT solver for it. A full SAT solver can be
696 desirable to enable more complex dependency mappings and / or queries,
697 for instance one possible use case for a SAT solver could be that of handling
698 the current known recursive dependency issues. It is not known if this would
699 address such issues but such evaluation is desirable. If support for a full SAT
700 solver proves too complex or that it cannot address recursive dependency issues
701 Kconfig should have at least clear and well defined semantics which also
702 addresses and documents limitations or requirements such as the ones dealing
703 with recursive dependencies.
705 Further work on both of these areas is welcomed on Kconfig. We elaborate
706 on both of these in the next two subsections.
711 The use of Kconfig is broad, Linux is now only one of Kconfig's users:
712 one study has completed a broad analysis of Kconfig use in 12 projects [0]_.
713 Despite its widespread use, and although this document does a reasonable job
714 in documenting basic Kconfig syntax a more precise definition of Kconfig
715 semantics is welcomed. One project deduced Kconfig semantics through
716 the use of the xconfig configurator [1]_. Work should be done to confirm if
717 the deduced semantics matches our intended Kconfig design goals.
718 Another project formalized a denotational semantics of a core subset of
719 the Kconfig language [10]_.
721 Having well defined semantics can be useful for tools for practical
722 evaluation of dependencies, for instance one such case was work to
723 express in boolean abstraction of the inferred semantics of Kconfig to
724 translate Kconfig logic into boolean formulas and run a SAT solver on this to
725 find dead code / features (always inactive), 114 dead features were found in
726 Linux using this methodology [1]_ (Section 8: Threats to validity).
727 The kismet tool, based on the semantics in [10]_, finds abuses of reverse
728 dependencies and has led to dozens of committed fixes to Linux Kconfig files [11]_.
730 Confirming this could prove useful as Kconfig stands as one of the leading
731 industrial variability modeling languages [1]_ [2]_. Its study would help
732 evaluate practical uses of such languages, their use was only theoretical
733 and real world requirements were not well understood. As it stands though
734 only reverse engineering techniques have been used to deduce semantics from
735 variability modeling languages such as Kconfig [3]_.
737 .. [0] https://www.eng.uwaterloo.ca/~shshe/kconfig_semantics.pdf
738 .. [1] https://gsd.uwaterloo.ca/sites/default/files/vm-2013-berger.pdf
739 .. [2] https://gsd.uwaterloo.ca/sites/default/files/ase241-berger_0.pdf
740 .. [3] https://gsd.uwaterloo.ca/sites/default/files/icse2011.pdf
742 Full SAT solver for Kconfig
743 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
745 Although SAT solvers [4]_ haven't yet been used by Kconfig directly, as noted
746 in the previous subsection, work has been done however to express in boolean
747 abstraction the inferred semantics of Kconfig to translate Kconfig logic into
748 boolean formulas and run a SAT solver on it [5]_. Another known related project
749 is CADOS [6]_ (former VAMOS [7]_) and the tools, mainly undertaker [8]_, which
750 has been introduced first with [9]_. The basic concept of undertaker is to
751 extract variability models from Kconfig and put them together with a
752 propositional formula extracted from CPP #ifdefs and build-rules into a SAT
753 solver in order to find dead code, dead files, and dead symbols. If using a SAT
754 solver is desirable on Kconfig one approach would be to evaluate repurposing
755 such efforts somehow on Kconfig. There is enough interest from mentors of
756 existing projects to not only help advise how to integrate this work upstream
757 but also help maintain it long term. Interested developers should visit:
759 https://kernelnewbies.org/KernelProjects/kconfig-sat
761 .. [4] https://www.cs.cornell.edu/~sabhar/chapters/SATSolvers-KR-Handbook.pdf
762 .. [5] https://gsd.uwaterloo.ca/sites/default/files/vm-2013-berger.pdf
763 .. [6] https://cados.cs.fau.de
764 .. [7] https://vamos.cs.fau.de
765 .. [8] https://undertaker.cs.fau.de
766 .. [9] https://www4.cs.fau.de/Publications/2011/tartler_11_eurosys.pdf
767 .. [10] https://paulgazzillo.com/papers/esecfse21.pdf
768 .. [11] https://github.com/paulgazz/kmax