1 menu "Code maturity level options"
4 bool "Prompt for development and/or incomplete code/drivers"
6 Some of the various things that Linux supports (such as network
7 drivers, file systems, network protocols, etc.) can be in a state
8 of development where the functionality, stability, or the level of
9 testing is not yet high enough for general use. This is usually
10 known as the "alpha-test" phase among developers. If a feature is
11 currently in alpha-test, then the developers usually discourage
12 uninformed widespread use of this feature by the general public to
13 avoid "Why doesn't this work?" type mail messages. However, active
14 testing and use of these systems is welcomed. Just be aware that it
15 may not meet the normal level of reliability or it may fail to work
16 in some special cases. Detailed bug reports from people familiar
17 with the kernel internals are usually welcomed by the developers
18 (before submitting bug reports, please read the documents
19 <file:README>, <file:MAINTAINERS>, <file:REPORTING-BUGS>,
20 <file:Documentation/BUG-HUNTING>, and
21 <file:Documentation/oops-tracing.txt> in the kernel source).
23 This option will also make obsoleted drivers available. These are
24 drivers that have been replaced by something else, and/or are
25 scheduled to be removed in a future kernel release.
27 Unless you intend to help test and develop a feature or driver that
28 falls into this category, or you have a situation that requires
29 using these features, you should probably say N here, which will
30 cause the configurator to present you with fewer choices. If
31 you say Y here, you will be offered the choice of using features or
32 drivers that are currently considered to be in the alpha-test phase.
35 bool "Select only drivers expected to compile cleanly" if EXPERIMENTAL
38 Select this option if you don't even want to see the option
39 to configure known-broken drivers.
45 depends on !CLEAN_COMPILE
50 depends on BROKEN || !SMP
55 depends on SMP || PREEMPT
63 string "Local version - append to kernel release"
65 Append an extra string to the end of your kernel version.
66 This will show up when you type uname, for example.
67 The string you set here will be appended after the contents of
68 any files with a filename matching localversion* in your
69 object and source tree, in that order. Your total string can
70 be a maximum of 64 characters.
73 bool "Support for paging of anonymous memory (swap)"
77 This option allows you to choose whether you want to have support
78 for socalled swap devices or swap files in your kernel that are
79 used to provide more virtual memory than the actual RAM present
80 in your computer. If unsure say Y.
86 Inter Process Communication is a suite of library functions and
87 system calls which let processes (running programs) synchronize and
88 exchange information. It is generally considered to be a good thing,
89 and some programs won't run unless you say Y here. In particular, if
90 you want to run the DOS emulator dosemu under Linux (read the
91 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
92 you'll need to say Y here.
94 You can find documentation about IPC with "info ipc" and also in
95 section 6.4 of the Linux Programmer's Guide, available from
96 <http://www.tldp.org/guides.html>.
99 bool "POSIX Message Queues"
100 depends on NET && EXPERIMENTAL
102 POSIX variant of message queues is a part of IPC. In POSIX message
103 queues every message has a priority which decides about succession
104 of receiving it by a process. If you want to compile and run
105 programs written e.g. for Solaris with use of its POSIX message
106 queues (functions mq_*) say Y here. To use this feature you will
107 also need mqueue library, available from
108 <http://www.mat.uni.torun.pl/~wrona/posix_ipc/>
110 POSIX message queues are visible as a filesystem called 'mqueue'
111 and can be mounted somewhere if you want to do filesystem
112 operations on message queues.
116 config BSD_PROCESS_ACCT
117 bool "BSD Process Accounting"
119 If you say Y here, a user level program will be able to instruct the
120 kernel (via a special system call) to write process accounting
121 information to a file: whenever a process exits, information about
122 that process will be appended to the file by the kernel. The
123 information includes things such as creation time, owning user,
124 command name, memory usage, controlling terminal etc. (the complete
125 list is in the struct acct in <file:include/linux/acct.h>). It is
126 up to the user level program to do useful things with this
127 information. This is generally a good idea, so say Y.
129 config BSD_PROCESS_ACCT_V3
130 bool "BSD Process Accounting version 3 file format"
131 depends on BSD_PROCESS_ACCT
134 If you say Y here, the process accounting information is written
135 in a new file format that also logs the process IDs of each
136 process and it's parent. Note that this file format is incompatible
137 with previous v0/v1/v2 file formats, so you will need updated tools
138 for processing it. A preliminary version of these tools is available
139 at <http://www.physik3.uni-rostock.de/tim/kernel/utils/acct/>.
141 menu "Class Based Kernel Resource Management"
144 bool "Class Based Kernel Resource Management Core"
145 depends on EXPERIMENTAL
147 Class-based Kernel Resource Management is a framework for controlling
148 and monitoring resource allocation of user-defined groups of tasks or
149 incoming socket connections. For more information, please visit
152 If you say Y here, enable the Resource Class File System and atleast
153 one of the resource controllers below. Say N if you are unsure.
156 tristate "Resource Class File System (User API)"
159 RCFS is the filesystem API for CKRM. This separate configuration
160 option is provided only for debugging and will eventually disappear
161 since rcfs will be automounted whenever CKRM is configured.
163 Say N if unsure, Y if you've enabled CKRM, M to debug rcfs
166 config CKRM_TYPE_TASKCLASS
167 bool "Class Manager for Task Groups"
168 depends on CKRM && RCFS_FS
170 TASKCLASS provides the extensions for CKRM to track task classes
171 This is the base to enable task class based resource control for
172 cpu, memory and disk I/O.
177 tristate "Null Tasks Resource Manager"
178 depends on CKRM_TYPE_TASKCLASS
182 bool "Class based physical memory controller"
186 Provide the basic support for collecting physical memory usage
187 information among classes. Say Y if you want to know the memory
190 config CKRM_TYPE_SOCKETCLASS
191 bool "Class Manager for socket groups"
192 depends on CKRM && RCFS_FS
194 Provides a Null Resource Controller for CKRM that is purely for
195 demonstration purposes.
197 Say N if unsure, Y to use the feature.
200 config CKRM_RES_NUMTASKS
201 tristate "Number of Tasks Resource Manager"
202 depends on CKRM_TYPE_TASKCLASS
205 Provides a Resource Controller for CKRM that allows limiting number of
206 tasks a task class can have.
208 Say N if unsure, Y to use the feature.
210 config CKRM_RES_NUMTASKS_FORKRATE
211 tristate "Number of Tasks Resource Manager for Fork Rate"
212 depends on CKRM_RES_NUMTASKS
215 Provides a Resource Controller for CKRM that allows limiting the rate
216 of tasks a task class can fork per hour.
218 Say N if unsure, Y to use the feature.
221 config CKRM_CPU_SCHEDULE
222 bool "CKRM CPU scheduler"
223 depends on CKRM_TYPE_TASKCLASS
226 Use CKRM CPU scheduler instead of Linux Scheduler
228 Say N if unsure, Y to use the feature.
230 config CKRM_RES_BLKIO
231 tristate " Disk I/O Resource Controller"
232 depends on CKRM_TYPE_TASKCLASS && IOSCHED_CFQ
235 Provides a resource controller for best-effort block I/O
236 bandwidth control. The controller attempts this by proportional
237 servicing of requests in the I/O scheduler. However, seek
238 optimizations and reordering by device drivers/disk controllers may
239 alter the actual bandwidth delivered to a class.
241 Say N if unsure, Y to use the feature.
243 config CKRM_CPU_SCHEDULE_AT_BOOT
244 bool "Turn on at boot time"
245 depends on CKRM_CPU_SCHEDULE
248 Enable CKRM CPU Scheduler at boot time. Otherwise
249 it can be turned on dynamically at runtime. If not
250 turned on the default Linux Scheduler behavior
253 Say N if unsure, Y to use this feature
255 config CKRM_TYPE_SOCKETCLASS
256 bool "Class Manager for socket groups"
259 SOCKET provides the extensions for CKRM to track per socket
260 classes. This is the base to enable socket based resource
261 control for inbound connection control, bandwidth control etc.
265 config CKRM_RES_LISTENAQ
266 tristate "Multiple Accept Queues Resource Manager"
267 depends on CKRM_TYPE_SOCKETCLASS && ACCEPT_QUEUES
270 Provides a resource controller for CKRM to prioritize inbound
271 connection requests. See inbound control description for
272 "IP: TCP Multiple accept queues support". If you choose that
273 option choose this option to control the queue weights.
278 prompt "Classification Engine"
279 depends on CKRM && RCFS_FS
282 Select a classification engine (CE) that assists in
283 automatic classification of kernel objects managed by CKRM when
284 they are created. Without a CE, a user must manually
285 classify objects into classes. Processes inherit their parent's
288 Only one engine can be built into the kernel though all can be
289 built as modules (only one will load).
294 tristate "Vanilla RBCE"
296 Vanilla Rule-based Classification Engine (RBCE). Rules for
297 classifying kernel objects are created/deleted/modified through
298 a RCFS directory using a filesystem interface.
300 Any CE is optional. If unsure, say N.
303 tristate "Enhanced RBCE"
304 depends on DELAY_ACCT && RELAYFS_FS
306 Enhanced Rule-based Classification Engine (CRBCE). Like the Vanilla
307 RBCE, rules for classifying kernel objects are created, deleted and
308 modified through a RCFS directory using a filesystem interface
309 (requires CKRM_RCFS configured).
311 In addition, CRBCE provides per-process delay data
312 (requires DELAY_ACCT configured) and makes information on significant
313 kernel events available to userspace tools through relayfs
314 (requires RELAYFS_FS configured).
316 Any CE is optional. If unsure, say N.
323 bool "Sysctl support"
325 The sysctl interface provides a means of dynamically changing
326 certain kernel parameters and variables on the fly without requiring
327 a recompile of the kernel or reboot of the system. The primary
328 interface consists of a system call, but if you say Y to "/proc
329 file system support", a tree of modifiable sysctl entries will be
330 generated beneath the /proc/sys directory. They are explained in the
331 files in <file:Documentation/sysctl/>. Note that enabling this
332 option will enlarge the kernel by at least 8 KB.
334 As it is generally a good thing, you should say Y here unless
335 building a kernel for install/rescue disks or your system is very
339 bool "Auditing support"
340 default y if SECURITY_SELINUX
343 Enable auditing infrastructure that can be used with another
344 kernel subsystem, such as SELinux (which requires this for
345 logging of avc messages output). Does not do system-call
346 auditing without CONFIG_AUDITSYSCALL.
349 bool "Enable system-call auditing support"
350 depends on AUDIT && (X86 || PPC64 || ARCH_S390 || IA64)
351 default y if SECURITY_SELINUX
354 Enable low-overhead system-call auditing infrastructure that
355 can be used independently or with another kernel subsystem,
359 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)" if DEBUG_KERNEL
361 default 17 if ARCH_S390
362 default 16 if X86_NUMAQ || IA64
366 Select kernel log buffer size as a power of 2.
367 Defaults and Examples:
368 17 => 128 KB for S/390
369 16 => 64 KB for x86 NUMAQ or IA-64
371 14 => 16 KB for uniprocessor
376 bool "Support for hot-pluggable devices" if !ARCH_S390
379 Say Y here if you want to plug devices into your computer while
380 the system is running, and be able to use them quickly. In many
381 cases, the devices can likewise be unplugged at any time too.
383 One well known example of this is PCMCIA- or PC-cards, credit-card
384 size devices such as network cards, modems or hard drives which are
385 plugged into slots found on all modern laptop computers. Another
386 example, used on modern desktops as well as laptops, is USB.
388 Enable HOTPLUG and KMOD, and build a modular kernel. Get agent
389 software (at <http://linux-hotplug.sourceforge.net/>) and install it.
390 Then your kernel will automatically call out to a user mode "policy
391 agent" (/sbin/hotplug) to load modules and set up software needed
392 to use devices as you hotplug them.
394 config KOBJECT_UEVENT
395 bool "Kernel Userspace Events"
399 This option enables the kernel userspace event layer, which is a
400 simple mechanism for kernel-to-user communication over a netlink
402 The goal of the kernel userspace events layer is to provide a simple
403 and efficient events system, that notifies userspace about kobject
404 state changes. This will enable applications to just listen for
405 events instead of polling system devices and files.
406 Hotplug events (kobject addition and removal) are also available on
407 the netlink socket in addition to the execution of /sbin/hotplug if
408 CONFIG_HOTPLUG is enabled.
410 Say Y, unless you are building a system requiring minimal memory
414 bool "Kernel .config support"
416 This option enables the complete Linux kernel ".config" file
417 contents to be saved in the kernel. It provides documentation
418 of which kernel options are used in a running kernel or in an
419 on-disk kernel. This information can be extracted from the kernel
420 image file with the script scripts/extract-ikconfig and used as
421 input to rebuild the current kernel or to build another kernel.
422 It can also be extracted from a running kernel by reading
423 /proc/config.gz if enabled (below).
426 bool "Enable access to .config through /proc/config.gz"
427 depends on IKCONFIG && PROC_FS
429 This option enables access to the kernel configuration file
430 through /proc/config.gz.
436 This option enables panic() to be called when a system is out of
437 memory. This feature along with /proc/sys/kernel/panic allows a
438 different behavior on out-of-memory conditions when the standard
439 behavior (killing processes in an attempt to recover) does not
446 depends on !OOM_PANIC
450 bool "Configure standard kernel features (for small systems)"
452 This option allows certain base kernel options and settings
453 to be disabled or tweaked. This is for specialized
454 environments which can tolerate a "non-standard" kernel.
455 Only use this if you really know what you are doing.
458 bool "Enable delay accounting (EXPERIMENTAL)"
460 In addition to counting frequency the total delay in ns is also
461 recorded. CPU delays are specified as cpu-wait and cpu-run.
462 I/O delays are recorded for memory and regular I/O.
463 Information is accessible through /proc/<pid>/delay.
466 bool "Load all symbols for debugging/kksymoops" if EMBEDDED
469 Say Y here to let the kernel print out symbolic crash information and
470 symbolic stack backtraces. This increases the size of the kernel
471 somewhat, as all symbols have to be loaded into the kernel image.
474 bool "Include all symbols in kallsyms"
475 depends on DEBUG_KERNEL && KALLSYMS
477 Normally kallsyms only contains the symbols of functions, for nicer
478 OOPS messages. Some debuggers can use kallsyms for other
479 symbols too: say Y here to include all symbols, and you
480 don't care about adding 300k to the size of your kernel.
484 config KALLSYMS_EXTRA_PASS
485 bool "Do an extra kallsyms pass"
488 If kallsyms is not working correctly, the build will fail with
489 inconsistent kallsyms data. If that occurs, log a bug report and
490 turn on KALLSYMS_EXTRA_PASS which should result in a stable build.
491 Always say N here unless you find a bug in kallsyms, which must be
492 reported. KALLSYMS_EXTRA_PASS is only a temporary workaround while
493 you wait for kallsyms to be fixed.
496 bool "Enable futex support" if EMBEDDED
499 Disabling this option will cause the kernel to be built without
500 support for "fast userspace mutexes". The resulting kernel may not
501 run glibc-based applications correctly.
504 bool "Enable eventpoll support" if EMBEDDED
507 Disabling this option will cause the kernel to be built without
508 support for epoll family of system calls.
510 config CC_OPTIMIZE_FOR_SIZE
511 bool "Optimize for size"
512 default y if ARM || H8300
515 Enabling this option will pass "-Os" instead of "-O2" to gcc
516 resulting in a smaller kernel.
518 WARNING: some versions of gcc may generate incorrect code with this
519 option. If problems are observed, a gcc upgrade may be needed.
525 bool "Use full shmem filesystem" if EMBEDDED && MMU
527 The shmem is an internal filesystem used to manage shared memory.
528 It is backed by swap and manages resource limits. It is also exported
529 to userspace as tmpfs if TMPFS is enabled. Disabling this
530 option replaces shmem and tmpfs with the much simpler ramfs code,
531 which may be appropriate on small systems without swap.
533 config CC_ALIGN_FUNCTIONS
534 int "Function alignment" if EMBEDDED
537 Align the start of functions to the next power-of-two greater than n,
538 skipping up to n bytes. For instance, 32 aligns functions
539 to the next 32-byte boundary, but 24 would align to the next
540 32-byte boundary only if this can be done by skipping 23 bytes or less.
541 Zero means use compiler's default.
543 config CC_ALIGN_LABELS
544 int "Label alignment" if EMBEDDED
547 Align all branch targets to a power-of-two boundary, skipping
548 up to n bytes like ALIGN_FUNCTIONS. This option can easily
549 make code slower, because it must insert dummy operations for
550 when the branch target is reached in the usual flow of the code.
551 Zero means use compiler's default.
553 config CC_ALIGN_LOOPS
554 int "Loop alignment" if EMBEDDED
557 Align loops to a power-of-two boundary, skipping up to n bytes.
558 Zero means use compiler's default.
560 config CC_ALIGN_JUMPS
561 int "Jump alignment" if EMBEDDED
564 Align branch targets to a power-of-two boundary, for branch
565 targets where the targets can only be reached by jumping,
566 skipping up to n bytes like ALIGN_FUNCTIONS. In this case,
567 no dummy operations need be executed.
568 Zero means use compiler's default.
570 endmenu # General setup
576 menu "Loadable module support"
579 bool "Enable loadable module support"
581 Kernel modules are small pieces of compiled code which can
582 be inserted in the running kernel, rather than being
583 permanently built into the kernel. You use the "modprobe"
584 tool to add (and sometimes remove) them. If you say Y here,
585 many parts of the kernel can be built as modules (by
586 answering M instead of Y where indicated): this is most
587 useful for infrequently used options which are not required
588 for booting. For more information, see the man pages for
589 modprobe, lsmod, modinfo, insmod and rmmod.
591 If you say Y here, you will need to run "make
592 modules_install" to put the modules under /lib/modules/
593 where modprobe can find them (you may need to be root to do
599 bool "Module unloading"
602 Without this option you will not be able to unload any
603 modules (note that some modules may not be unloadable
604 anyway), which makes your kernel slightly smaller and
605 simpler. If unsure, say Y.
607 config MODULE_FORCE_UNLOAD
608 bool "Forced module unloading"
609 depends on MODULE_UNLOAD && EXPERIMENTAL
611 This option allows you to force a module to unload, even if the
612 kernel believes it is unsafe: the kernel will remove the module
613 without waiting for anyone to stop using it (using the -f option to
614 rmmod). This is mainly for kernel developers and desperate users.
617 config OBSOLETE_MODPARM
622 You need this option to use module parameters on modules which
623 have not been converted to the new module parameter system yet.
627 bool "Module versioning support (EXPERIMENTAL)"
628 depends on MODULES && EXPERIMENTAL
630 Usually, you have to use modules compiled with your kernel.
631 Saying Y here makes it sometimes possible to use modules
632 compiled for different kernels, by adding enough information
633 to the modules to (hopefully) spot any changes which would
634 make them incompatible with the kernel you are running. If
637 config MODULE_SRCVERSION_ALL
638 bool "Source checksum for all modules"
641 Modules which contain a MODULE_VERSION get an extra "srcversion"
642 field inserted into their modinfo section, which contains a
643 sum of the source files which made it. This helps maintainers
644 see exactly which source was used to build a module (since
645 others sometimes change the module source without updating
646 the version). With this option, such a "srcversion" field
647 will be created for all modules. If unsure, say N.
650 bool "Module signature verification (EXPERIMENTAL)"
651 depends on MODULES && EXPERIMENTAL
654 select CRYPTO_SIGNATURE
656 Check modules for valid signatures upon load.
658 config MODULE_SIG_FORCE
659 bool "Required modules to be validly signed (EXPERIMENTAL)"
660 depends on MODULE_SIG
662 Reject unsigned modules or signed modules for which we don't have a
666 bool "Automatic kernel module loading"
669 Normally when you have selected some parts of the kernel to
670 be created as kernel modules, you must load them (using the
671 "modprobe" command) before you can use them. If you say Y
672 here, some parts of the kernel will be able to load modules
673 automatically: when a part of the kernel needs a module, it
674 runs modprobe with the appropriate arguments, thereby
675 loading the module if it is available. If unsure, say Y.
680 depends on (SMP && MODULE_UNLOAD) || HOTPLUG_CPU
682 Need stop_machine() primitive.