7 option env="KERNELVERSION"
13 default "/lib/modules/$UNAME_RELEASE/.config"
14 default "/etc/kernel-config"
15 default "/boot/config-$UNAME_RELEASE"
16 default "$ARCH_DEFCONFIG"
17 default "arch/$ARCH/defconfig"
22 bool "Prompt for development and/or incomplete code/drivers"
24 Some of the various things that Linux supports (such as network
25 drivers, file systems, network protocols, etc.) can be in a state
26 of development where the functionality, stability, or the level of
27 testing is not yet high enough for general use. This is usually
28 known as the "alpha-test" phase among developers. If a feature is
29 currently in alpha-test, then the developers usually discourage
30 uninformed widespread use of this feature by the general public to
31 avoid "Why doesn't this work?" type mail messages. However, active
32 testing and use of these systems is welcomed. Just be aware that it
33 may not meet the normal level of reliability or it may fail to work
34 in some special cases. Detailed bug reports from people familiar
35 with the kernel internals are usually welcomed by the developers
36 (before submitting bug reports, please read the documents
37 <file:README>, <file:MAINTAINERS>, <file:REPORTING-BUGS>,
38 <file:Documentation/BUG-HUNTING>, and
39 <file:Documentation/oops-tracing.txt> in the kernel source).
41 This option will also make obsoleted drivers available. These are
42 drivers that have been replaced by something else, and/or are
43 scheduled to be removed in a future kernel release.
45 Unless you intend to help test and develop a feature or driver that
46 falls into this category, or you have a situation that requires
47 using these features, you should probably say N here, which will
48 cause the configurator to present you with fewer choices. If
49 you say Y here, you will be offered the choice of using features or
50 drivers that are currently considered to be in the alpha-test phase.
57 depends on BROKEN || !SMP
62 depends on SMP || PREEMPT
65 config INIT_ENV_ARG_LIMIT
70 Maximum of each of the number of arguments and environment
71 variables passed to init from the kernel command line.
75 string "Local version - append to kernel release"
77 Append an extra string to the end of your kernel version.
78 This will show up when you type uname, for example.
79 The string you set here will be appended after the contents of
80 any files with a filename matching localversion* in your
81 object and source tree, in that order. Your total string can
82 be a maximum of 64 characters.
84 config LOCALVERSION_AUTO
85 bool "Automatically append version information to the version string"
88 This will try to automatically determine if the current tree is a
89 release tree by looking for git tags that belong to the current
92 A string of the format -gxxxxxxxx will be added to the localversion
93 if a git-based tree is found. The string generated by this will be
94 appended after any matching localversion* files, and after the value
95 set in CONFIG_LOCALVERSION.
97 (The actual string used here is the first eight characters produced
98 by running the command:
100 $ git rev-parse --verify HEAD
102 which is done within the script "scripts/setlocalversion".)
105 prompt "Kernel compression mode"
108 The linux kernel is a kind of self-extracting executable.
109 Several compression algorithms are available, which differ
110 in efficiency, compression and decompression speed.
111 Compression speed is only relevant when building a kernel.
112 Decompression speed is relevant at each boot.
114 If you have any problems with bzip2 or lzma compressed
115 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
116 version of this functionality (bzip2 only), for 2.4, was
117 supplied by Christian Ludwig)
119 High compression options are mostly useful for users, who
120 are low on disk space (embedded systems), but for whom ram
123 If in doubt, select 'gzip'
128 The old and tried gzip compression. Its compression ratio is
129 the poorest among the 3 choices; however its speed (both
130 compression and decompression) is the fastest.
135 Its compression ratio and speed is intermediate.
136 Decompression speed is slowest among the 3.
137 The kernel size is about 10 per cent smaller with bzip2,
138 in comparison to gzip.
139 Bzip2 uses a large amount of memory. For modern kernels
140 you will need at least 8MB RAM or more for booting.
145 The most recent compression algorithm.
146 Its ratio is best, decompression speed is between the other
147 2. Compression is slowest.
148 The kernel size is about 33 per cent smaller with lzma,
149 in comparison to gzip.
155 bool "Support for paging of anonymous memory (swap)"
156 depends on MMU && BLOCK
159 This option allows you to choose whether you want to have support
160 for so called swap devices or swap files in your kernel that are
161 used to provide more virtual memory than the actual RAM present
162 in your computer. If unsure say Y.
167 Inter Process Communication is a suite of library functions and
168 system calls which let processes (running programs) synchronize and
169 exchange information. It is generally considered to be a good thing,
170 and some programs won't run unless you say Y here. In particular, if
171 you want to run the DOS emulator dosemu under Linux (read the
172 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
173 you'll need to say Y here.
175 You can find documentation about IPC with "info ipc" and also in
176 section 6.4 of the Linux Programmer's Guide, available from
177 <http://www.tldp.org/guides.html>.
179 config SYSVIPC_SYSCTL
186 bool "POSIX Message Queues"
187 depends on NET && EXPERIMENTAL
189 POSIX variant of message queues is a part of IPC. In POSIX message
190 queues every message has a priority which decides about succession
191 of receiving it by a process. If you want to compile and run
192 programs written e.g. for Solaris with use of its POSIX message
193 queues (functions mq_*) say Y here.
195 POSIX message queues are visible as a filesystem called 'mqueue'
196 and can be mounted somewhere if you want to do filesystem
197 operations on message queues.
201 config BSD_PROCESS_ACCT
202 bool "BSD Process Accounting"
204 If you say Y here, a user level program will be able to instruct the
205 kernel (via a special system call) to write process accounting
206 information to a file: whenever a process exits, information about
207 that process will be appended to the file by the kernel. The
208 information includes things such as creation time, owning user,
209 command name, memory usage, controlling terminal etc. (the complete
210 list is in the struct acct in <file:include/linux/acct.h>). It is
211 up to the user level program to do useful things with this
212 information. This is generally a good idea, so say Y.
214 config BSD_PROCESS_ACCT_V3
215 bool "BSD Process Accounting version 3 file format"
216 depends on BSD_PROCESS_ACCT
219 If you say Y here, the process accounting information is written
220 in a new file format that also logs the process IDs of each
221 process and it's parent. Note that this file format is incompatible
222 with previous v0/v1/v2 file formats, so you will need updated tools
223 for processing it. A preliminary version of these tools is available
224 at <http://www.gnu.org/software/acct/>.
227 bool "Export task/process statistics through netlink (EXPERIMENTAL)"
231 Export selected statistics for tasks/processes through the
232 generic netlink interface. Unlike BSD process accounting, the
233 statistics are available during the lifetime of tasks/processes as
234 responses to commands. Like BSD accounting, they are sent to user
239 config TASK_DELAY_ACCT
240 bool "Enable per-task delay accounting (EXPERIMENTAL)"
243 Collect information on time spent by a task waiting for system
244 resources like cpu, synchronous block I/O completion and swapping
245 in pages. Such statistics can help in setting a task's priorities
246 relative to other tasks for cpu, io, rss limits etc.
251 bool "Enable extended accounting over taskstats (EXPERIMENTAL)"
254 Collect extended task accounting data and send the data
255 to userland for processing over the taskstats interface.
259 config TASK_IO_ACCOUNTING
260 bool "Enable per-task storage I/O accounting (EXPERIMENTAL)"
261 depends on TASK_XACCT
263 Collect information on the number of bytes of storage I/O which this
269 bool "Auditing support"
272 Enable auditing infrastructure that can be used with another
273 kernel subsystem, such as SELinux (which requires this for
274 logging of avc messages output). Does not do system-call
275 auditing without CONFIG_AUDITSYSCALL.
278 bool "Enable system-call auditing support"
279 depends on AUDIT && (X86 || PPC || PPC64 || S390 || IA64 || UML || SPARC64|| SUPERH)
280 default y if SECURITY_SELINUX
282 Enable low-overhead system-call auditing infrastructure that
283 can be used independently or with another kernel subsystem,
284 such as SELinux. To use audit's filesystem watch feature, please
285 ensure that INOTIFY is configured.
289 depends on AUDITSYSCALL && INOTIFY
292 tristate "Kernel .config support"
294 This option enables the complete Linux kernel ".config" file
295 contents to be saved in the kernel. It provides documentation
296 of which kernel options are used in a running kernel or in an
297 on-disk kernel. This information can be extracted from the kernel
298 image file with the script scripts/extract-ikconfig and used as
299 input to rebuild the current kernel or to build another kernel.
300 It can also be extracted from a running kernel by reading
301 /proc/config.gz if enabled (below).
304 bool "Enable access to .config through /proc/config.gz"
305 depends on IKCONFIG && PROC_FS
307 This option enables access to the kernel configuration file
308 through /proc/config.gz.
311 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
315 Select kernel log buffer size as a power of 2.
325 bool "Control Group support"
327 This option will let you use process cgroup subsystems
333 bool "Example debug cgroup subsystem"
337 This option enables a simple cgroup subsystem that
338 exports useful debugging information about the cgroups
344 bool "Namespace cgroup subsystem"
347 Provides a simple namespace cgroup subsystem to
348 provide hierarchical naming of sets of namespaces,
349 for instance virtual servers and checkpoint/restart
352 config CGROUP_FREEZER
353 bool "control group freezer subsystem"
356 Provides a way to freeze and unfreeze all tasks in a
360 bool "Device controller for cgroups"
361 depends on CGROUPS && EXPERIMENTAL
363 Provides a cgroup implementing whitelists for devices which
364 a process in the cgroup can mknod or open.
367 bool "Cpuset support"
368 depends on SMP && CGROUPS
370 This option will let you create and manage CPUSETs which
371 allow dynamically partitioning a system into sets of CPUs and
372 Memory Nodes and assigning tasks to run only within those sets.
373 This is primarily useful on large SMP or NUMA systems.
378 # Architectures with an unreliable sched_clock() should select this:
380 config HAVE_UNSTABLE_SCHED_CLOCK
384 bool "Group CPU scheduler"
385 depends on EXPERIMENTAL
388 This feature lets CPU scheduler recognize task groups and control CPU
389 bandwidth allocation to such task groups.
391 config FAIR_GROUP_SCHED
392 bool "Group scheduling for SCHED_OTHER"
393 depends on GROUP_SCHED
396 config RT_GROUP_SCHED
397 bool "Group scheduling for SCHED_RR/FIFO"
398 depends on EXPERIMENTAL
399 depends on GROUP_SCHED
402 This feature lets you explicitly allocate real CPU bandwidth
403 to users or control groups (depending on the "Basis for grouping tasks"
404 setting below. If enabled, it will also make it impossible to
405 schedule realtime tasks for non-root users until you allocate
406 realtime bandwidth for them.
407 See Documentation/scheduler/sched-rt-group.txt for more information.
410 depends on GROUP_SCHED
411 prompt "Basis for grouping tasks"
417 This option will choose userid as the basis for grouping
418 tasks, thus providing equal CPU bandwidth to each user.
421 bool "Control groups"
424 This option allows you to create arbitrary task groups
425 using the "cgroup" pseudo filesystem and control
426 the cpu bandwidth allocated to each such task group.
427 Refer to Documentation/cgroups.txt for more information
428 on "cgroup" pseudo filesystem.
432 config CGROUP_CPUACCT
433 bool "Simple CPU accounting cgroup subsystem"
436 Provides a simple Resource Controller for monitoring the
437 total CPU consumed by the tasks in a cgroup
439 config RESOURCE_COUNTERS
440 bool "Resource counters"
442 This option enables controller independent resource accounting
443 infrastructure that works with cgroups
449 config CGROUP_MEM_RES_CTLR
450 bool "Memory Resource Controller for Control Groups"
451 depends on CGROUPS && RESOURCE_COUNTERS
454 Provides a memory resource controller that manages both anonymous
455 memory and page cache. (See Documentation/controllers/memory.txt)
457 Note that setting this option increases fixed memory overhead
458 associated with each page of memory in the system. By this,
459 20(40)bytes/PAGE_SIZE on 32(64)bit system will be occupied by memory
460 usage tracking struct at boot. Total amount of this is printed out
463 Only enable when you're ok with these trade offs and really
464 sure you need the memory resource controller. Even when you enable
465 this, you can set "cgroup_disable=memory" at your boot option to
466 disable memory resource controller and you can avoid overheads.
467 (and lose benefits of memory resource contoller)
469 This config option also selects MM_OWNER config option, which
470 could in turn add some fork/exit overhead.
472 config SYSFS_DEPRECATED
475 config SYSFS_DEPRECATED_V2
476 bool "Create deprecated sysfs files"
479 select SYSFS_DEPRECATED
481 This option creates deprecated symlinks such as the
482 "device"-link, the <subsystem>:<name>-link, and the
483 "bus"-link. It may also add deprecated key in the
485 None of these features or values should be used today, as
486 they export driver core implementation details to userspace
487 or export properties which can't be kept stable across kernel
490 If enabled, this option will also move any device structures
491 that belong to a class, back into the /sys/class hierarchy, in
492 order to support older versions of udev and some userspace
495 If you are using a distro with the most recent userspace
496 packages, it should be safe to say N here.
498 config PROC_PID_CPUSET
499 bool "Include legacy /proc/<pid>/cpuset file"
504 bool "Kernel->user space relay support (formerly relayfs)"
506 This option enables support for relay interface support in
507 certain file systems (such as debugfs).
508 It is designed to provide an efficient mechanism for tools and
509 facilities to relay large amounts of data from kernel space to
515 bool "Namespaces support" if EMBEDDED
518 Provides the way to make tasks work with different objects using
519 the same id. For example same IPC id may refer to different objects
520 or same user id or pid may refer to different tasks when used in
521 different namespaces.
525 depends on NAMESPACES
527 In this namespace tasks see different info provided with the
532 depends on NAMESPACES && SYSVIPC
534 In this namespace tasks work with IPC ids which correspond to
535 different IPC objects in different namespaces
538 bool "User namespace (EXPERIMENTAL)"
539 depends on NAMESPACES && EXPERIMENTAL
541 This allows containers, i.e. vservers, to use user namespaces
542 to provide different user info for different servers.
546 bool "PID Namespaces (EXPERIMENTAL)"
548 depends on NAMESPACES && EXPERIMENTAL
550 Support process id namespaces. This allows having multiple
551 process with the same pid as long as they are in different
552 pid namespaces. This is a building block of containers.
554 Unless you want to work with an experimental feature
557 config BLK_DEV_INITRD
558 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
559 depends on BROKEN || !FRV
561 The initial RAM filesystem is a ramfs which is loaded by the
562 boot loader (loadlin or lilo) and that is mounted as root
563 before the normal boot procedure. It is typically used to
564 load modules needed to mount the "real" root file system,
565 etc. See <file:Documentation/initrd.txt> for details.
567 If RAM disk support (BLK_DEV_RAM) is also included, this
568 also enables initial RAM disk (initrd) support and adds
569 15 Kbytes (more on some other architectures) to the kernel size.
579 config CC_OPTIMIZE_FOR_SIZE
580 bool "Optimize for size"
583 Enabling this option will pass "-Os" instead of "-O2" to gcc
584 resulting in a smaller kernel.
592 bool "Configure standard kernel features (for small systems)"
594 This option allows certain base kernel options and settings
595 to be disabled or tweaked. This is for specialized
596 environments which can tolerate a "non-standard" kernel.
597 Only use this if you really know what you are doing.
600 bool "Enable 16-bit UID system calls" if EMBEDDED
601 depends on ARM || BLACKFIN || CRIS || FRV || H8300 || X86_32 || M68K || (S390 && !64BIT) || SUPERH || SPARC32 || (SPARC64 && COMPAT) || UML || (X86_64 && IA32_EMULATION)
604 This enables the legacy 16-bit UID syscall wrappers.
606 config SYSCTL_SYSCALL
607 bool "Sysctl syscall support" if EMBEDDED
611 sys_sysctl uses binary paths that have been found challenging
612 to properly maintain and use. The interface in /proc/sys
613 using paths with ascii names is now the primary path to this
616 Almost nothing using the binary sysctl interface so if you are
617 trying to save some space it is probably safe to disable this,
618 making your kernel marginally smaller.
620 If unsure say Y here.
623 bool "Load all symbols for debugging/ksymoops" if EMBEDDED
626 Say Y here to let the kernel print out symbolic crash information and
627 symbolic stack backtraces. This increases the size of the kernel
628 somewhat, as all symbols have to be loaded into the kernel image.
631 bool "Include all symbols in kallsyms"
632 depends on DEBUG_KERNEL && KALLSYMS
634 Normally kallsyms only contains the symbols of functions, for nicer
635 OOPS messages. Some debuggers can use kallsyms for other
636 symbols too: say Y here to include all symbols, if you need them
637 and you don't care about adding 300k to the size of your kernel.
641 config KALLSYMS_STRIP_GENERATED
642 bool "Strip machine generated symbols from kallsyms"
643 depends on KALLSYMS_ALL
646 Say N if you want kallsyms to retain even machine generated symbols.
648 config KALLSYMS_EXTRA_PASS
649 bool "Do an extra kallsyms pass"
652 If kallsyms is not working correctly, the build will fail with
653 inconsistent kallsyms data. If that occurs, log a bug report and
654 turn on KALLSYMS_EXTRA_PASS which should result in a stable build.
655 Always say N here unless you find a bug in kallsyms, which must be
656 reported. KALLSYMS_EXTRA_PASS is only a temporary workaround while
657 you wait for kallsyms to be fixed.
661 bool "Support for hot-pluggable devices" if EMBEDDED
664 This option is provided for the case where no hotplug or uevent
665 capabilities is wanted by the kernel. You should only consider
666 disabling this option for embedded systems that do not use modules, a
667 dynamic /dev tree, or dynamic device discovery. Just say Y.
671 bool "Enable support for printk" if EMBEDDED
673 This option enables normal printk support. Removing it
674 eliminates most of the message strings from the kernel image
675 and makes the kernel more or less silent. As this makes it
676 very difficult to diagnose system problems, saying N here is
677 strongly discouraged.
680 bool "BUG() support" if EMBEDDED
683 Disabling this option eliminates support for BUG and WARN, reducing
684 the size of your kernel image and potentially quietly ignoring
685 numerous fatal conditions. You should only consider disabling this
686 option for embedded systems with no facilities for reporting errors.
691 bool "Enable ELF core dumps" if EMBEDDED
693 Enable support for generating core dumps. Disabling saves about 4k.
695 config PCSPKR_PLATFORM
696 bool "Enable PC-Speaker support" if EMBEDDED
697 depends on ALPHA || X86 || MIPS || PPC_PREP || PPC_CHRP || PPC_PSERIES
700 This option allows to disable the internal PC-Speaker
701 support, saving some memory.
704 bool "Disable heap randomization"
707 Randomizing heap placement makes heap exploits harder, but it
708 also breaks ancient binaries (including anything libc5 based).
709 This option changes the bootup default to heap randomization
710 disabled, and can be overriden runtime by setting
711 /proc/sys/kernel/randomize_va_space to 2.
713 On non-ancient distros (post-2000 ones) N is usually a safe choice.
717 bool "Enable full-sized data structures for core" if EMBEDDED
719 Disabling this option reduces the size of miscellaneous core
720 kernel data structures. This saves memory on small machines,
721 but may reduce performance.
724 bool "Enable futex support" if EMBEDDED
728 Disabling this option will cause the kernel to be built without
729 support for "fast userspace mutexes". The resulting kernel may not
730 run glibc-based applications correctly.
736 bool "Enable eventpoll support" if EMBEDDED
740 Disabling this option will cause the kernel to be built without
741 support for epoll family of system calls.
744 bool "Enable signalfd() system call" if EMBEDDED
748 Enable the signalfd() system call that allows to receive signals
749 on a file descriptor.
754 bool "Enable timerfd() system call" if EMBEDDED
758 Enable the timerfd() system call that allows to receive timer
759 events on a file descriptor.
764 bool "Enable eventfd() system call" if EMBEDDED
768 Enable the eventfd() system call that allows to receive both
769 kernel notification (ie. KAIO) or userspace notifications.
774 bool "Use full shmem filesystem" if EMBEDDED
778 The shmem is an internal filesystem used to manage shared memory.
779 It is backed by swap and manages resource limits. It is also exported
780 to userspace as tmpfs if TMPFS is enabled. Disabling this
781 option replaces shmem and tmpfs with the much simpler ramfs code,
782 which may be appropriate on small systems without swap.
785 bool "Enable AIO support" if EMBEDDED
788 This option enables POSIX asynchronous I/O which may by used
789 by some high performance threaded applications. Disabling
790 this option saves about 7k.
792 config VM_EVENT_COUNTERS
794 bool "Enable VM event counters for /proc/vmstat" if EMBEDDED
796 VM event counters are needed for event counts to be shown.
797 This option allows the disabling of the VM event counters
798 on EMBEDDED systems. /proc/vmstat will only show page counts
799 if VM event counters are disabled.
803 bool "Enable PCI quirk workarounds" if EMBEDDED
806 This enables workarounds for various PCI chipset
807 bugs/quirks. Disable this only if your target machine is
808 unaffected by PCI quirks.
812 bool "Enable SLUB debugging support" if EMBEDDED
813 depends on SLUB && SYSFS
815 SLUB has extensive debug support features. Disabling these can
816 result in significant savings in code size. This also disables
817 SLUB sysfs support. /sys/slab will not exist and there will be
818 no support for cache validation etc.
821 prompt "Choose SLAB allocator"
824 This option allows to select a slab allocator.
829 The regular slab allocator that is established and known to work
830 well in all environments. It organizes cache hot objects in
831 per cpu and per node queues.
834 bool "SLUB (Unqueued Allocator)"
836 SLUB is a slab allocator that minimizes cache line usage
837 instead of managing queues of cached objects (SLAB approach).
838 Per cpu caching is realized using slabs of objects instead
839 of queues of objects. SLUB can use memory efficiently
840 and has enhanced diagnostics. SLUB is the default choice for
845 bool "SLOB (Simple Allocator)"
847 SLOB replaces the stock allocator with a drastically simpler
848 allocator. SLOB is generally more space efficient but
849 does not perform as well on large systems.
854 bool "Profiling support (EXPERIMENTAL)"
856 Say Y here to enable the extended profiling support mechanisms used
857 by profilers such as OProfile.
860 # Place an empty function call at each tracepoint site. Can be
861 # dynamically changed for a probe function.
867 bool "Activate markers"
868 depends on TRACEPOINTS
870 Place an empty function call at each marker site. Can be
871 dynamically changed for a probe function.
873 source "arch/Kconfig"
875 endmenu # General setup
877 config HAVE_GENERIC_DMA_COHERENT
884 depends on SLAB || SLUB_DEBUG
897 default 0 if BASE_FULL
898 default 1 if !BASE_FULL
901 bool "Enable loadable module support"
903 Kernel modules are small pieces of compiled code which can
904 be inserted in the running kernel, rather than being
905 permanently built into the kernel. You use the "modprobe"
906 tool to add (and sometimes remove) them. If you say Y here,
907 many parts of the kernel can be built as modules (by
908 answering M instead of Y where indicated): this is most
909 useful for infrequently used options which are not required
910 for booting. For more information, see the man pages for
911 modprobe, lsmod, modinfo, insmod and rmmod.
913 If you say Y here, you will need to run "make
914 modules_install" to put the modules under /lib/modules/
915 where modprobe can find them (you may need to be root to do
922 config MODULE_FORCE_LOAD
923 bool "Forced module loading"
926 Allow loading of modules without version information (ie. modprobe
927 --force). Forced module loading sets the 'F' (forced) taint flag and
928 is usually a really bad idea.
931 bool "Module unloading"
933 Without this option you will not be able to unload any
934 modules (note that some modules may not be unloadable
935 anyway), which makes your kernel smaller, faster
936 and simpler. If unsure, say Y.
938 config MODULE_FORCE_UNLOAD
939 bool "Forced module unloading"
940 depends on MODULE_UNLOAD && EXPERIMENTAL
942 This option allows you to force a module to unload, even if the
943 kernel believes it is unsafe: the kernel will remove the module
944 without waiting for anyone to stop using it (using the -f option to
945 rmmod). This is mainly for kernel developers and desperate users.
949 bool "Module versioning support"
951 Usually, you have to use modules compiled with your kernel.
952 Saying Y here makes it sometimes possible to use modules
953 compiled for different kernels, by adding enough information
954 to the modules to (hopefully) spot any changes which would
955 make them incompatible with the kernel you are running. If
958 config MODULE_SRCVERSION_ALL
959 bool "Source checksum for all modules"
961 Modules which contain a MODULE_VERSION get an extra "srcversion"
962 field inserted into their modinfo section, which contains a
963 sum of the source files which made it. This helps maintainers
964 see exactly which source was used to build a module (since
965 others sometimes change the module source without updating
966 the version). With this option, such a "srcversion" field
967 will be created for all modules. If unsure, say N.
972 This is being removed soon. These days, CONFIG_MODULES
973 implies CONFIG_KMOD, so use that instead.
977 config INIT_ALL_POSSIBLE
980 Back when each arch used to define their own cpu_online_map and
981 cpu_possible_map, some of them chose to initialize cpu_possible_map
982 with all 1s, and others with all 0s. When they were centralised,
983 it was better to provide this option than to break all the archs
984 and have several arch maintainers persuing me down dark alleys.
989 depends on (SMP && MODULE_UNLOAD) || HOTPLUG_CPU
991 Need stop_machine() primitive.
993 source "block/Kconfig"
995 config PREEMPT_NOTIFIERS
999 prompt "RCU Implementation"
1005 This option selects the classic RCU implementation that is
1006 designed for best read-side performance on non-realtime
1009 Select this option if you are unsure.
1012 bool "Tree-based hierarchical RCU"
1014 This option selects the RCU implementation that is
1015 designed for very large SMP system with hundreds or
1019 bool "Preemptible RCU"
1022 This option reduces the latency of the kernel by making certain
1023 RCU sections preemptible. Normally RCU code is non-preemptible, if
1024 this option is selected then read-only RCU sections become
1025 preemptible. This helps latency, but may expose bugs due to
1026 now-naive assumptions about each RCU read-side critical section
1027 remaining on a given CPU through its execution.
1032 bool "Enable tracing for RCU"
1033 depends on TREE_RCU || PREEMPT_RCU
1035 This option provides tracing in RCU which presents stats
1036 in debugfs for debugging RCU implementation.
1038 Say Y here if you want to enable RCU tracing
1039 Say N if you are unsure.
1042 int "Tree-based hierarchical RCU fanout value"
1044 range 2 32 if !64BIT
1047 default 32 if !64BIT
1049 This option controls the fanout of hierarchical implementations
1050 of RCU, allowing RCU to work efficiently on machines with
1051 large numbers of CPUs. This value must be at least the cube
1052 root of NR_CPUS, which allows NR_CPUS up to 32,768 for 32-bit
1053 systems and up to 262,144 for 64-bit systems.
1055 Select a specific number if testing RCU itself.
1056 Take the default if unsure.
1058 config RCU_FANOUT_EXACT
1059 bool "Disable tree-based hierarchical RCU auto-balancing"
1063 This option forces use of the exact RCU_FANOUT value specified,
1064 regardless of imbalances in the hierarchy. This is useful for
1065 testing RCU itself, and might one day be useful on systems with
1066 strong NUMA behavior.
1068 Without RCU_FANOUT_EXACT, the code will balance the hierarchy.
1072 config TREE_RCU_TRACE
1073 def_bool RCU_TRACE && TREE_RCU
1076 This option provides tracing for the TREE_RCU implementation,
1077 permitting Makefile to trivially select kernel/rcutree_trace.c.
1079 config PREEMPT_RCU_TRACE
1080 def_bool RCU_TRACE && PREEMPT_RCU
1083 This option provides tracing for the PREEMPT_RCU implementation,
1084 permitting Makefile to trivially select kernel/rcupreempt_trace.c.