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42 drivers that have been replaced by something else, and/or are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".)
104 config HAVE_KERNEL_GZIP
107 config HAVE_KERNEL_BZIP2
110 config HAVE_KERNEL_LZMA
114 prompt "Kernel compression mode"
116 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA
118 The linux kernel is a kind of self-extracting executable.
119 Several compression algorithms are available, which differ
120 in efficiency, compression and decompression speed.
121 Compression speed is only relevant when building a kernel.
122 Decompression speed is relevant at each boot.
124 If you have any problems with bzip2 or lzma compressed
125 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
126 version of this functionality (bzip2 only), for 2.4, was
127 supplied by Christian Ludwig)
129 High compression options are mostly useful for users, who
130 are low on disk space (embedded systems), but for whom ram
133 If in doubt, select 'gzip'
137 depends on HAVE_KERNEL_GZIP
139 The old and tried gzip compression. Its compression ratio is
140 the poorest among the 3 choices; however its speed (both
141 compression and decompression) is the fastest.
145 depends on HAVE_KERNEL_BZIP2
147 Its compression ratio and speed is intermediate.
148 Decompression speed is slowest among the three. The kernel
149 size is about 10% smaller with bzip2, in comparison to gzip.
150 Bzip2 uses a large amount of memory. For modern kernels you
151 will need at least 8MB RAM or more for booting.
155 depends on HAVE_KERNEL_LZMA
157 The most recent compression algorithm.
158 Its ratio is best, decompression speed is between the other
159 two. Compression is slowest. The kernel size is about 33%
160 smaller with LZMA in comparison to gzip.
165 bool "Support for paging of anonymous memory (swap)"
166 depends on MMU && BLOCK
169 This option allows you to choose whether you want to have support
170 for so called swap devices or swap files in your kernel that are
171 used to provide more virtual memory than the actual RAM present
172 in your computer. If unsure say Y.
177 Inter Process Communication is a suite of library functions and
178 system calls which let processes (running programs) synchronize and
179 exchange information. It is generally considered to be a good thing,
180 and some programs won't run unless you say Y here. In particular, if
181 you want to run the DOS emulator dosemu under Linux (read the
182 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
183 you'll need to say Y here.
185 You can find documentation about IPC with "info ipc" and also in
186 section 6.4 of the Linux Programmer's Guide, available from
187 <http://www.tldp.org/guides.html>.
189 config SYSVIPC_SYSCTL
196 bool "POSIX Message Queues"
197 depends on NET && EXPERIMENTAL
199 POSIX variant of message queues is a part of IPC. In POSIX message
200 queues every message has a priority which decides about succession
201 of receiving it by a process. If you want to compile and run
202 programs written e.g. for Solaris with use of its POSIX message
203 queues (functions mq_*) say Y here.
205 POSIX message queues are visible as a filesystem called 'mqueue'
206 and can be mounted somewhere if you want to do filesystem
207 operations on message queues.
211 config POSIX_MQUEUE_SYSCTL
213 depends on POSIX_MQUEUE
217 config BSD_PROCESS_ACCT
218 bool "BSD Process Accounting"
220 If you say Y here, a user level program will be able to instruct the
221 kernel (via a special system call) to write process accounting
222 information to a file: whenever a process exits, information about
223 that process will be appended to the file by the kernel. The
224 information includes things such as creation time, owning user,
225 command name, memory usage, controlling terminal etc. (the complete
226 list is in the struct acct in <file:include/linux/acct.h>). It is
227 up to the user level program to do useful things with this
228 information. This is generally a good idea, so say Y.
230 config BSD_PROCESS_ACCT_V3
231 bool "BSD Process Accounting version 3 file format"
232 depends on BSD_PROCESS_ACCT
235 If you say Y here, the process accounting information is written
236 in a new file format that also logs the process IDs of each
237 process and it's parent. Note that this file format is incompatible
238 with previous v0/v1/v2 file formats, so you will need updated tools
239 for processing it. A preliminary version of these tools is available
240 at <http://www.gnu.org/software/acct/>.
243 bool "Export task/process statistics through netlink (EXPERIMENTAL)"
247 Export selected statistics for tasks/processes through the
248 generic netlink interface. Unlike BSD process accounting, the
249 statistics are available during the lifetime of tasks/processes as
250 responses to commands. Like BSD accounting, they are sent to user
255 config TASK_DELAY_ACCT
256 bool "Enable per-task delay accounting (EXPERIMENTAL)"
259 Collect information on time spent by a task waiting for system
260 resources like cpu, synchronous block I/O completion and swapping
261 in pages. Such statistics can help in setting a task's priorities
262 relative to other tasks for cpu, io, rss limits etc.
267 bool "Enable extended accounting over taskstats (EXPERIMENTAL)"
270 Collect extended task accounting data and send the data
271 to userland for processing over the taskstats interface.
275 config TASK_IO_ACCOUNTING
276 bool "Enable per-task storage I/O accounting (EXPERIMENTAL)"
277 depends on TASK_XACCT
279 Collect information on the number of bytes of storage I/O which this
285 bool "Auditing support"
288 Enable auditing infrastructure that can be used with another
289 kernel subsystem, such as SELinux (which requires this for
290 logging of avc messages output). Does not do system-call
291 auditing without CONFIG_AUDITSYSCALL.
294 bool "Enable system-call auditing support"
295 depends on AUDIT && (X86 || PPC || PPC64 || S390 || IA64 || UML || SPARC64|| SUPERH)
296 default y if SECURITY_SELINUX
298 Enable low-overhead system-call auditing infrastructure that
299 can be used independently or with another kernel subsystem,
300 such as SELinux. To use audit's filesystem watch feature, please
301 ensure that INOTIFY is configured.
305 depends on AUDITSYSCALL
311 prompt "RCU Implementation"
317 This option selects the classic RCU implementation that is
318 designed for best read-side performance on non-realtime
321 Select this option if you are unsure.
324 bool "Tree-based hierarchical RCU"
326 This option selects the RCU implementation that is
327 designed for very large SMP system with hundreds or
331 bool "Preemptible RCU"
334 This option reduces the latency of the kernel by making certain
335 RCU sections preemptible. Normally RCU code is non-preemptible, if
336 this option is selected then read-only RCU sections become
337 preemptible. This helps latency, but may expose bugs due to
338 now-naive assumptions about each RCU read-side critical section
339 remaining on a given CPU through its execution.
344 bool "Enable tracing for RCU"
345 depends on TREE_RCU || PREEMPT_RCU
347 This option provides tracing in RCU which presents stats
348 in debugfs for debugging RCU implementation.
350 Say Y here if you want to enable RCU tracing
351 Say N if you are unsure.
354 int "Tree-based hierarchical RCU fanout value"
361 This option controls the fanout of hierarchical implementations
362 of RCU, allowing RCU to work efficiently on machines with
363 large numbers of CPUs. This value must be at least the cube
364 root of NR_CPUS, which allows NR_CPUS up to 32,768 for 32-bit
365 systems and up to 262,144 for 64-bit systems.
367 Select a specific number if testing RCU itself.
368 Take the default if unsure.
370 config RCU_FANOUT_EXACT
371 bool "Disable tree-based hierarchical RCU auto-balancing"
375 This option forces use of the exact RCU_FANOUT value specified,
376 regardless of imbalances in the hierarchy. This is useful for
377 testing RCU itself, and might one day be useful on systems with
378 strong NUMA behavior.
380 Without RCU_FANOUT_EXACT, the code will balance the hierarchy.
384 config TREE_RCU_TRACE
385 def_bool RCU_TRACE && TREE_RCU
388 This option provides tracing for the TREE_RCU implementation,
389 permitting Makefile to trivially select kernel/rcutree_trace.c.
391 config PREEMPT_RCU_TRACE
392 def_bool RCU_TRACE && PREEMPT_RCU
395 This option provides tracing for the PREEMPT_RCU implementation,
396 permitting Makefile to trivially select kernel/rcupreempt_trace.c.
398 endmenu # "RCU Subsystem"
401 tristate "Kernel .config support"
403 This option enables the complete Linux kernel ".config" file
404 contents to be saved in the kernel. It provides documentation
405 of which kernel options are used in a running kernel or in an
406 on-disk kernel. This information can be extracted from the kernel
407 image file with the script scripts/extract-ikconfig and used as
408 input to rebuild the current kernel or to build another kernel.
409 It can also be extracted from a running kernel by reading
410 /proc/config.gz if enabled (below).
413 bool "Enable access to .config through /proc/config.gz"
414 depends on IKCONFIG && PROC_FS
416 This option enables access to the kernel configuration file
417 through /proc/config.gz.
420 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
424 Select kernel log buffer size as a power of 2.
434 # Architectures with an unreliable sched_clock() should select this:
436 config HAVE_UNSTABLE_SCHED_CLOCK
440 bool "Group CPU scheduler"
441 depends on EXPERIMENTAL
444 This feature lets CPU scheduler recognize task groups and control CPU
445 bandwidth allocation to such task groups.
446 In order to create a group from arbitrary set of processes, use
447 CONFIG_CGROUPS. (See Control Group support.)
449 config FAIR_GROUP_SCHED
450 bool "Group scheduling for SCHED_OTHER"
451 depends on GROUP_SCHED
454 config RT_GROUP_SCHED
455 bool "Group scheduling for SCHED_RR/FIFO"
456 depends on EXPERIMENTAL
457 depends on GROUP_SCHED
460 This feature lets you explicitly allocate real CPU bandwidth
461 to users or control groups (depending on the "Basis for grouping tasks"
462 setting below. If enabled, it will also make it impossible to
463 schedule realtime tasks for non-root users until you allocate
464 realtime bandwidth for them.
465 See Documentation/scheduler/sched-rt-group.txt for more information.
468 depends on GROUP_SCHED
469 prompt "Basis for grouping tasks"
475 This option will choose userid as the basis for grouping
476 tasks, thus providing equal CPU bandwidth to each user.
479 bool "Control groups"
482 This option allows you to create arbitrary task groups
483 using the "cgroup" pseudo filesystem and control
484 the cpu bandwidth allocated to each such task group.
485 Refer to Documentation/cgroups/cgroups.txt for more
486 information on "cgroup" pseudo filesystem.
491 boolean "Control Group support"
493 This option adds support for grouping sets of processes together, for
494 use with process control subsystems such as Cpusets, CFS, memory
495 controls or device isolation.
497 - Documentation/scheduler/sched-design-CFS.txt (CFS)
498 - Documentation/cgroups/ (features for grouping, isolation
499 and resource control)
506 bool "Example debug cgroup subsystem"
510 This option enables a simple cgroup subsystem that
511 exports useful debugging information about the cgroups
517 bool "Namespace cgroup subsystem"
520 Provides a simple namespace cgroup subsystem to
521 provide hierarchical naming of sets of namespaces,
522 for instance virtual servers and checkpoint/restart
525 config CGROUP_FREEZER
526 bool "Freezer cgroup subsystem"
529 Provides a way to freeze and unfreeze all tasks in a
533 bool "Device controller for cgroups"
534 depends on CGROUPS && EXPERIMENTAL
536 Provides a cgroup implementing whitelists for devices which
537 a process in the cgroup can mknod or open.
540 bool "Cpuset support"
543 This option will let you create and manage CPUSETs which
544 allow dynamically partitioning a system into sets of CPUs and
545 Memory Nodes and assigning tasks to run only within those sets.
546 This is primarily useful on large SMP or NUMA systems.
550 config PROC_PID_CPUSET
551 bool "Include legacy /proc/<pid>/cpuset file"
555 config CGROUP_CPUACCT
556 bool "Simple CPU accounting cgroup subsystem"
559 Provides a simple Resource Controller for monitoring the
560 total CPU consumed by the tasks in a cgroup.
562 config RESOURCE_COUNTERS
563 bool "Resource counters"
565 This option enables controller independent resource accounting
566 infrastructure that works with cgroups.
569 config CGROUP_MEM_RES_CTLR
570 bool "Memory Resource Controller for Control Groups"
571 depends on CGROUPS && RESOURCE_COUNTERS
574 Provides a memory resource controller that manages both anonymous
575 memory and page cache. (See Documentation/cgroups/memory.txt)
577 Note that setting this option increases fixed memory overhead
578 associated with each page of memory in the system. By this,
579 20(40)bytes/PAGE_SIZE on 32(64)bit system will be occupied by memory
580 usage tracking struct at boot. Total amount of this is printed out
583 Only enable when you're ok with these trade offs and really
584 sure you need the memory resource controller. Even when you enable
585 this, you can set "cgroup_disable=memory" at your boot option to
586 disable memory resource controller and you can avoid overheads.
587 (and lose benefits of memory resource controller)
589 This config option also selects MM_OWNER config option, which
590 could in turn add some fork/exit overhead.
592 config CGROUP_MEM_RES_CTLR_SWAP
593 bool "Memory Resource Controller Swap Extension(EXPERIMENTAL)"
594 depends on CGROUP_MEM_RES_CTLR && SWAP && EXPERIMENTAL
596 Add swap management feature to memory resource controller. When you
597 enable this, you can limit mem+swap usage per cgroup. In other words,
598 when you disable this, memory resource controller has no cares to
599 usage of swap...a process can exhaust all of the swap. This extension
600 is useful when you want to avoid exhaustion swap but this itself
601 adds more overheads and consumes memory for remembering information.
602 Especially if you use 32bit system or small memory system, please
603 be careful about enabling this. When memory resource controller
604 is disabled by boot option, this will be automatically disabled and
605 there will be no overhead from this. Even when you set this config=y,
606 if boot option "noswapaccount" is set, swap will not be accounted.
607 Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page
608 size is 4096bytes, 512k per 1Gbytes of swap.
615 config SYSFS_DEPRECATED
618 config SYSFS_DEPRECATED_V2
619 bool "Create deprecated sysfs layout for older userspace tools"
622 select SYSFS_DEPRECATED
624 This option switches the layout of sysfs to the deprecated
627 The current sysfs layout features a unified device tree at
628 /sys/devices/, which is able to express a hierarchy between
629 class devices. If the deprecated option is set to Y, the
630 unified device tree is split into a bus device tree at
631 /sys/devices/ and several individual class device trees at
632 /sys/class/. The class and bus devices will be connected by
633 "<subsystem>:<name>" and the "device" links. The "block"
634 class devices, will not show up in /sys/class/block/. Some
635 subsystems will suppress the creation of some devices which
636 depend on the unified device tree.
638 This option is not a pure compatibility option that can
639 be safely enabled on newer distributions. It will change the
640 layout of sysfs to the non-extensible deprecated version,
641 and disable some features, which can not be exported without
642 confusing older userspace tools. Since 2007/2008 all major
643 distributions do not enable this option, and ship no tools which
644 depend on the deprecated layout or this option.
646 If you are using a new kernel on an older distribution, or use
647 older userspace tools, you might need to say Y here. Do not say Y,
648 if the original kernel, that came with your distribution, has
649 this option set to N.
652 bool "Kernel->user space relay support (formerly relayfs)"
654 This option enables support for relay interface support in
655 certain file systems (such as debugfs).
656 It is designed to provide an efficient mechanism for tools and
657 facilities to relay large amounts of data from kernel space to
663 bool "Namespaces support" if EMBEDDED
666 Provides the way to make tasks work with different objects using
667 the same id. For example same IPC id may refer to different objects
668 or same user id or pid may refer to different tasks when used in
669 different namespaces.
673 depends on NAMESPACES
675 In this namespace tasks see different info provided with the
680 depends on NAMESPACES && (SYSVIPC || POSIX_MQUEUE)
682 In this namespace tasks work with IPC ids which correspond to
683 different IPC objects in different namespaces.
686 bool "User namespace (EXPERIMENTAL)"
687 depends on NAMESPACES && EXPERIMENTAL
689 This allows containers, i.e. vservers, to use user namespaces
690 to provide different user info for different servers.
694 bool "PID Namespaces (EXPERIMENTAL)"
696 depends on NAMESPACES && EXPERIMENTAL
698 Support process id namespaces. This allows having multiple
699 processes with the same pid as long as they are in different
700 pid namespaces. This is a building block of containers.
702 Unless you want to work with an experimental feature
706 bool "Network namespace"
708 depends on NAMESPACES && EXPERIMENTAL && NET
710 Allow user space to create what appear to be multiple instances
711 of the network stack.
713 config BLK_DEV_INITRD
714 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
715 depends on BROKEN || !FRV
717 The initial RAM filesystem is a ramfs which is loaded by the
718 boot loader (loadlin or lilo) and that is mounted as root
719 before the normal boot procedure. It is typically used to
720 load modules needed to mount the "real" root file system,
721 etc. See <file:Documentation/initrd.txt> for details.
723 If RAM disk support (BLK_DEV_RAM) is also included, this
724 also enables initial RAM disk (initrd) support and adds
725 15 Kbytes (more on some other architectures) to the kernel size.
735 config CC_OPTIMIZE_FOR_SIZE
736 bool "Optimize for size"
739 Enabling this option will pass "-Os" instead of "-O2" to gcc
740 resulting in a smaller kernel.
751 bool "Configure standard kernel features (for small systems)"
753 This option allows certain base kernel options and settings
754 to be disabled or tweaked. This is for specialized
755 environments which can tolerate a "non-standard" kernel.
756 Only use this if you really know what you are doing.
759 bool "Enable 16-bit UID system calls" if EMBEDDED
760 depends on ARM || BLACKFIN || CRIS || FRV || H8300 || X86_32 || M68K || (S390 && !64BIT) || SUPERH || SPARC32 || (SPARC64 && COMPAT) || UML || (X86_64 && IA32_EMULATION)
763 This enables the legacy 16-bit UID syscall wrappers.
765 config SYSCTL_SYSCALL
766 bool "Sysctl syscall support" if EMBEDDED
770 sys_sysctl uses binary paths that have been found challenging
771 to properly maintain and use. The interface in /proc/sys
772 using paths with ascii names is now the primary path to this
775 Almost nothing using the binary sysctl interface so if you are
776 trying to save some space it is probably safe to disable this,
777 making your kernel marginally smaller.
779 If unsure say Y here.
782 bool "Load all symbols for debugging/ksymoops" if EMBEDDED
785 Say Y here to let the kernel print out symbolic crash information and
786 symbolic stack backtraces. This increases the size of the kernel
787 somewhat, as all symbols have to be loaded into the kernel image.
790 bool "Include all symbols in kallsyms"
791 depends on DEBUG_KERNEL && KALLSYMS
793 Normally kallsyms only contains the symbols of functions, for nicer
794 OOPS messages. Some debuggers can use kallsyms for other
795 symbols too: say Y here to include all symbols, if you need them
796 and you don't care about adding 300k to the size of your kernel.
800 config KALLSYMS_EXTRA_PASS
801 bool "Do an extra kallsyms pass"
804 If kallsyms is not working correctly, the build will fail with
805 inconsistent kallsyms data. If that occurs, log a bug report and
806 turn on KALLSYMS_EXTRA_PASS which should result in a stable build.
807 Always say N here unless you find a bug in kallsyms, which must be
808 reported. KALLSYMS_EXTRA_PASS is only a temporary workaround while
809 you wait for kallsyms to be fixed.
812 config STRIP_ASM_SYMS
813 bool "Strip assembler-generated symbols during link"
816 Strip internal assembler-generated symbols during a link (symbols
817 that look like '.Lxxx') so they don't pollute the output of
818 get_wchan() and suchlike.
821 bool "Support for hot-pluggable devices" if EMBEDDED
824 This option is provided for the case where no hotplug or uevent
825 capabilities is wanted by the kernel. You should only consider
826 disabling this option for embedded systems that do not use modules, a
827 dynamic /dev tree, or dynamic device discovery. Just say Y.
831 bool "Enable support for printk" if EMBEDDED
833 This option enables normal printk support. Removing it
834 eliminates most of the message strings from the kernel image
835 and makes the kernel more or less silent. As this makes it
836 very difficult to diagnose system problems, saying N here is
837 strongly discouraged.
840 bool "BUG() support" if EMBEDDED
843 Disabling this option eliminates support for BUG and WARN, reducing
844 the size of your kernel image and potentially quietly ignoring
845 numerous fatal conditions. You should only consider disabling this
846 option for embedded systems with no facilities for reporting errors.
851 bool "Enable ELF core dumps" if EMBEDDED
853 Enable support for generating core dumps. Disabling saves about 4k.
855 config PCSPKR_PLATFORM
856 bool "Enable PC-Speaker support" if EMBEDDED
857 depends on ALPHA || X86 || MIPS || PPC_PREP || PPC_CHRP || PPC_PSERIES
860 This option allows to disable the internal PC-Speaker
861 support, saving some memory.
865 bool "Enable full-sized data structures for core" if EMBEDDED
867 Disabling this option reduces the size of miscellaneous core
868 kernel data structures. This saves memory on small machines,
869 but may reduce performance.
872 bool "Enable futex support" if EMBEDDED
876 Disabling this option will cause the kernel to be built without
877 support for "fast userspace mutexes". The resulting kernel may not
878 run glibc-based applications correctly.
881 bool "Enable eventpoll support" if EMBEDDED
885 Disabling this option will cause the kernel to be built without
886 support for epoll family of system calls.
889 bool "Enable signalfd() system call" if EMBEDDED
893 Enable the signalfd() system call that allows to receive signals
894 on a file descriptor.
899 bool "Enable timerfd() system call" if EMBEDDED
903 Enable the timerfd() system call that allows to receive timer
904 events on a file descriptor.
909 bool "Enable eventfd() system call" if EMBEDDED
913 Enable the eventfd() system call that allows to receive both
914 kernel notification (ie. KAIO) or userspace notifications.
919 bool "Use full shmem filesystem" if EMBEDDED
923 The shmem is an internal filesystem used to manage shared memory.
924 It is backed by swap and manages resource limits. It is also exported
925 to userspace as tmpfs if TMPFS is enabled. Disabling this
926 option replaces shmem and tmpfs with the much simpler ramfs code,
927 which may be appropriate on small systems without swap.
930 bool "Enable AIO support" if EMBEDDED
933 This option enables POSIX asynchronous I/O which may by used
934 by some high performance threaded applications. Disabling
935 this option saves about 7k.
937 config HAVE_PERF_COUNTERS
940 See tools/perf/design.txt for details.
942 menu "Performance Counters"
945 bool "Kernel Performance Counters"
946 depends on HAVE_PERF_COUNTERS
949 Enable kernel support for performance counter hardware.
951 Performance counters are special hardware registers available
952 on most modern CPUs. These registers count the number of certain
953 types of hw events: such as instructions executed, cachemisses
954 suffered, or branches mis-predicted - without slowing down the
955 kernel or applications. These registers can also trigger interrupts
956 when a threshold number of events have passed - and can thus be
957 used to profile the code that runs on that CPU.
959 The Linux Performance Counter subsystem provides an abstraction of
960 these hardware capabilities, available via a system call. It
961 provides per task and per CPU counters, and it provides event
962 capabilities on top of those.
967 bool "Tracepoint profile sources"
968 depends on PERF_COUNTERS && EVENT_TRACER
973 config VM_EVENT_COUNTERS
975 bool "Enable VM event counters for /proc/vmstat" if EMBEDDED
977 VM event counters are needed for event counts to be shown.
978 This option allows the disabling of the VM event counters
979 on EMBEDDED systems. /proc/vmstat will only show page counts
980 if VM event counters are disabled.
984 bool "Enable PCI quirk workarounds" if EMBEDDED
987 This enables workarounds for various PCI chipset
988 bugs/quirks. Disable this only if your target machine is
989 unaffected by PCI quirks.
993 bool "Enable SLUB debugging support" if EMBEDDED
994 depends on SLUB && SYSFS
996 SLUB has extensive debug support features. Disabling these can
997 result in significant savings in code size. This also disables
998 SLUB sysfs support. /sys/slab will not exist and there will be
999 no support for cache validation etc.
1002 bool "Disable heap randomization"
1005 Randomizing heap placement makes heap exploits harder, but it
1006 also breaks ancient binaries (including anything libc5 based).
1007 This option changes the bootup default to heap randomization
1008 disabled, and can be overridden at runtime by setting
1009 /proc/sys/kernel/randomize_va_space to 2.
1011 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1014 prompt "Choose SLAB allocator"
1017 This option allows to select a slab allocator.
1022 The regular slab allocator that is established and known to work
1023 well in all environments. It organizes cache hot objects in
1024 per cpu and per node queues.
1027 bool "SLUB (Unqueued Allocator)"
1029 SLUB is a slab allocator that minimizes cache line usage
1030 instead of managing queues of cached objects (SLAB approach).
1031 Per cpu caching is realized using slabs of objects instead
1032 of queues of objects. SLUB can use memory efficiently
1033 and has enhanced diagnostics. SLUB is the default choice for
1038 bool "SLOB (Simple Allocator)"
1040 SLOB replaces the stock allocator with a drastically simpler
1041 allocator. SLOB is generally more space efficient but
1042 does not perform as well on large systems.
1047 bool "Profiling support (EXPERIMENTAL)"
1049 Say Y here to enable the extended profiling support mechanisms used
1050 by profilers such as OProfile.
1053 # Place an empty function call at each tracepoint site. Can be
1054 # dynamically changed for a probe function.
1060 bool "Activate markers"
1063 Place an empty function call at each marker site. Can be
1064 dynamically changed for a probe function.
1066 source "arch/Kconfig"
1072 The slow work thread pool provides a number of dynamically allocated
1073 threads that can be used by the kernel to perform operations that
1074 take a relatively long time.
1076 An example of this would be CacheFiles doing a path lookup followed
1077 by a series of mkdirs and a create call, all of which have to touch
1080 See Documentation/slow-work.txt.
1082 endmenu # General setup
1084 config HAVE_GENERIC_DMA_COHERENT
1091 depends on SLAB || SLUB_DEBUG
1099 default 0 if BASE_FULL
1100 default 1 if !BASE_FULL
1103 bool "Enable loadable module support"
1105 Kernel modules are small pieces of compiled code which can
1106 be inserted in the running kernel, rather than being
1107 permanently built into the kernel. You use the "modprobe"
1108 tool to add (and sometimes remove) them. If you say Y here,
1109 many parts of the kernel can be built as modules (by
1110 answering M instead of Y where indicated): this is most
1111 useful for infrequently used options which are not required
1112 for booting. For more information, see the man pages for
1113 modprobe, lsmod, modinfo, insmod and rmmod.
1115 If you say Y here, you will need to run "make
1116 modules_install" to put the modules under /lib/modules/
1117 where modprobe can find them (you may need to be root to do
1124 config MODULE_FORCE_LOAD
1125 bool "Forced module loading"
1128 Allow loading of modules without version information (ie. modprobe
1129 --force). Forced module loading sets the 'F' (forced) taint flag and
1130 is usually a really bad idea.
1132 config MODULE_UNLOAD
1133 bool "Module unloading"
1135 Without this option you will not be able to unload any
1136 modules (note that some modules may not be unloadable
1137 anyway), which makes your kernel smaller, faster
1138 and simpler. If unsure, say Y.
1140 config MODULE_FORCE_UNLOAD
1141 bool "Forced module unloading"
1142 depends on MODULE_UNLOAD && EXPERIMENTAL
1144 This option allows you to force a module to unload, even if the
1145 kernel believes it is unsafe: the kernel will remove the module
1146 without waiting for anyone to stop using it (using the -f option to
1147 rmmod). This is mainly for kernel developers and desperate users.
1151 bool "Module versioning support"
1153 Usually, you have to use modules compiled with your kernel.
1154 Saying Y here makes it sometimes possible to use modules
1155 compiled for different kernels, by adding enough information
1156 to the modules to (hopefully) spot any changes which would
1157 make them incompatible with the kernel you are running. If
1160 config MODULE_SRCVERSION_ALL
1161 bool "Source checksum for all modules"
1163 Modules which contain a MODULE_VERSION get an extra "srcversion"
1164 field inserted into their modinfo section, which contains a
1165 sum of the source files which made it. This helps maintainers
1166 see exactly which source was used to build a module (since
1167 others sometimes change the module source without updating
1168 the version). With this option, such a "srcversion" field
1169 will be created for all modules. If unsure, say N.
1173 config INIT_ALL_POSSIBLE
1176 Back when each arch used to define their own cpu_online_map and
1177 cpu_possible_map, some of them chose to initialize cpu_possible_map
1178 with all 1s, and others with all 0s. When they were centralised,
1179 it was better to provide this option than to break all the archs
1180 and have several arch maintainers pursuing me down dark alleys.
1185 depends on (SMP && MODULE_UNLOAD) || HOTPLUG_CPU
1187 Need stop_machine() primitive.
1189 source "block/Kconfig"
1191 config PREEMPT_NOTIFIERS