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"
27 bool "Prompt for development and/or incomplete code/drivers"
29 Some of the various things that Linux supports (such as network
30 drivers, file systems, network protocols, etc.) can be in a state
31 of development where the functionality, stability, or the level of
32 testing is not yet high enough for general use. This is usually
33 known as the "alpha-test" phase among developers. If a feature is
34 currently in alpha-test, then the developers usually discourage
35 uninformed widespread use of this feature by the general public to
36 avoid "Why doesn't this work?" type mail messages. However, active
37 testing and use of these systems is welcomed. Just be aware that it
38 may not meet the normal level of reliability or it may fail to work
39 in some special cases. Detailed bug reports from people familiar
40 with the kernel internals are usually welcomed by the developers
41 (before submitting bug reports, please read the documents
42 <file:README>, <file:MAINTAINERS>, <file:REPORTING-BUGS>,
43 <file:Documentation/BUG-HUNTING>, and
44 <file:Documentation/oops-tracing.txt> in the kernel source).
46 This option will also make obsoleted drivers available. These are
47 drivers that have been replaced by something else, and/or are
48 scheduled to be removed in a future kernel release.
50 Unless you intend to help test and develop a feature or driver that
51 falls into this category, or you have a situation that requires
52 using these features, you should probably say N here, which will
53 cause the configurator to present you with fewer choices. If
54 you say Y here, you will be offered the choice of using features or
55 drivers that are currently considered to be in the alpha-test phase.
62 depends on BROKEN || !SMP
67 depends on SMP || PREEMPT
70 config INIT_ENV_ARG_LIMIT
75 Maximum of each of the number of arguments and environment
76 variables passed to init from the kernel command line.
80 string "Cross-compiler tool prefix"
82 Same as running 'make CROSS_COMPILE=prefix-' but stored for
83 default make runs in this kernel build directory. You don't
84 need to set this unless you want the configured kernel build
85 directory to select the cross-compiler automatically.
88 string "Local version - append to kernel release"
90 Append an extra string to the end of your kernel version.
91 This will show up when you type uname, for example.
92 The string you set here will be appended after the contents of
93 any files with a filename matching localversion* in your
94 object and source tree, in that order. Your total string can
95 be a maximum of 64 characters.
97 config LOCALVERSION_AUTO
98 bool "Automatically append version information to the version string"
101 This will try to automatically determine if the current tree is a
102 release tree by looking for git tags that belong to the current
103 top of tree revision.
105 A string of the format -gxxxxxxxx will be added to the localversion
106 if a git-based tree is found. The string generated by this will be
107 appended after any matching localversion* files, and after the value
108 set in CONFIG_LOCALVERSION.
110 (The actual string used here is the first eight characters produced
111 by running the command:
113 $ git rev-parse --verify HEAD
115 which is done within the script "scripts/setlocalversion".)
117 config HAVE_KERNEL_GZIP
120 config HAVE_KERNEL_BZIP2
123 config HAVE_KERNEL_LZMA
126 config HAVE_KERNEL_LZO
130 prompt "Kernel compression mode"
131 default KERNEL_LZO if ARCH_RK29
133 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_LZO
135 The linux kernel is a kind of self-extracting executable.
136 Several compression algorithms are available, which differ
137 in efficiency, compression and decompression speed.
138 Compression speed is only relevant when building a kernel.
139 Decompression speed is relevant at each boot.
141 If you have any problems with bzip2 or lzma compressed
142 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
143 version of this functionality (bzip2 only), for 2.4, was
144 supplied by Christian Ludwig)
146 High compression options are mostly useful for users, who
147 are low on disk space (embedded systems), but for whom ram
150 If in doubt, select 'gzip'
154 depends on HAVE_KERNEL_GZIP
156 The old and tried gzip compression. It provides a good balance
157 between compression ratio and decompression speed.
161 depends on HAVE_KERNEL_BZIP2
163 Its compression ratio and speed is intermediate.
164 Decompression speed is slowest among the three. The kernel
165 size is about 10% smaller with bzip2, in comparison to gzip.
166 Bzip2 uses a large amount of memory. For modern kernels you
167 will need at least 8MB RAM or more for booting.
171 depends on HAVE_KERNEL_LZMA
173 The most recent compression algorithm.
174 Its ratio is best, decompression speed is between the other
175 two. Compression is slowest. The kernel size is about 33%
176 smaller with LZMA in comparison to gzip.
180 depends on HAVE_KERNEL_LZO
182 Its compression ratio is the poorest among the 4. The kernel
183 size is about about 10% bigger than gzip; however its speed
184 (both compression and decompression) is the fastest.
189 bool "Support for paging of anonymous memory (swap)"
190 depends on MMU && BLOCK
193 This option allows you to choose whether you want to have support
194 for so called swap devices or swap files in your kernel that are
195 used to provide more virtual memory than the actual RAM present
196 in your computer. If unsure say Y.
201 Inter Process Communication is a suite of library functions and
202 system calls which let processes (running programs) synchronize and
203 exchange information. It is generally considered to be a good thing,
204 and some programs won't run unless you say Y here. In particular, if
205 you want to run the DOS emulator dosemu under Linux (read the
206 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
207 you'll need to say Y here.
209 You can find documentation about IPC with "info ipc" and also in
210 section 6.4 of the Linux Programmer's Guide, available from
211 <http://www.tldp.org/guides.html>.
213 config SYSVIPC_SYSCTL
220 bool "POSIX Message Queues"
221 depends on NET && EXPERIMENTAL
223 POSIX variant of message queues is a part of IPC. In POSIX message
224 queues every message has a priority which decides about succession
225 of receiving it by a process. If you want to compile and run
226 programs written e.g. for Solaris with use of its POSIX message
227 queues (functions mq_*) say Y here.
229 POSIX message queues are visible as a filesystem called 'mqueue'
230 and can be mounted somewhere if you want to do filesystem
231 operations on message queues.
235 config POSIX_MQUEUE_SYSCTL
237 depends on POSIX_MQUEUE
241 config BSD_PROCESS_ACCT
242 bool "BSD Process Accounting"
244 If you say Y here, a user level program will be able to instruct the
245 kernel (via a special system call) to write process accounting
246 information to a file: whenever a process exits, information about
247 that process will be appended to the file by the kernel. The
248 information includes things such as creation time, owning user,
249 command name, memory usage, controlling terminal etc. (the complete
250 list is in the struct acct in <file:include/linux/acct.h>). It is
251 up to the user level program to do useful things with this
252 information. This is generally a good idea, so say Y.
254 config BSD_PROCESS_ACCT_V3
255 bool "BSD Process Accounting version 3 file format"
256 depends on BSD_PROCESS_ACCT
259 If you say Y here, the process accounting information is written
260 in a new file format that also logs the process IDs of each
261 process and it's parent. Note that this file format is incompatible
262 with previous v0/v1/v2 file formats, so you will need updated tools
263 for processing it. A preliminary version of these tools is available
264 at <http://www.gnu.org/software/acct/>.
267 bool "Export task/process statistics through netlink (EXPERIMENTAL)"
271 Export selected statistics for tasks/processes through the
272 generic netlink interface. Unlike BSD process accounting, the
273 statistics are available during the lifetime of tasks/processes as
274 responses to commands. Like BSD accounting, they are sent to user
279 config TASK_DELAY_ACCT
280 bool "Enable per-task delay accounting (EXPERIMENTAL)"
283 Collect information on time spent by a task waiting for system
284 resources like cpu, synchronous block I/O completion and swapping
285 in pages. Such statistics can help in setting a task's priorities
286 relative to other tasks for cpu, io, rss limits etc.
291 bool "Enable extended accounting over taskstats (EXPERIMENTAL)"
294 Collect extended task accounting data and send the data
295 to userland for processing over the taskstats interface.
299 config TASK_IO_ACCOUNTING
300 bool "Enable per-task storage I/O accounting (EXPERIMENTAL)"
301 depends on TASK_XACCT
303 Collect information on the number of bytes of storage I/O which this
309 bool "Auditing support"
312 Enable auditing infrastructure that can be used with another
313 kernel subsystem, such as SELinux (which requires this for
314 logging of avc messages output). Does not do system-call
315 auditing without CONFIG_AUDITSYSCALL.
318 bool "Enable system-call auditing support"
319 depends on AUDIT && (X86 || PPC || S390 || IA64 || UML || SPARC64 || SUPERH)
320 default y if SECURITY_SELINUX
322 Enable low-overhead system-call auditing infrastructure that
323 can be used independently or with another kernel subsystem,
328 depends on AUDITSYSCALL
333 depends on AUDITSYSCALL
339 prompt "RCU Implementation"
343 bool "Tree-based hierarchical RCU"
345 This option selects the RCU implementation that is
346 designed for very large SMP system with hundreds or
347 thousands of CPUs. It also scales down nicely to
350 config TREE_PREEMPT_RCU
351 bool "Preemptable tree-based hierarchical RCU"
354 This option selects the RCU implementation that is
355 designed for very large SMP systems with hundreds or
356 thousands of CPUs, but for which real-time response
357 is also required. It also scales down nicely to
361 bool "UP-only small-memory-footprint RCU"
364 This option selects the RCU implementation that is
365 designed for UP systems from which real-time response
366 is not required. This option greatly reduces the
367 memory footprint of RCU.
372 bool "Enable tracing for RCU"
373 depends on TREE_RCU || TREE_PREEMPT_RCU
375 This option provides tracing in RCU which presents stats
376 in debugfs for debugging RCU implementation.
378 Say Y here if you want to enable RCU tracing
379 Say N if you are unsure.
382 int "Tree-based hierarchical RCU fanout value"
385 depends on TREE_RCU || TREE_PREEMPT_RCU
389 This option controls the fanout of hierarchical implementations
390 of RCU, allowing RCU to work efficiently on machines with
391 large numbers of CPUs. This value must be at least the cube
392 root of NR_CPUS, which allows NR_CPUS up to 32,768 for 32-bit
393 systems and up to 262,144 for 64-bit systems.
395 Select a specific number if testing RCU itself.
396 Take the default if unsure.
398 config RCU_FANOUT_EXACT
399 bool "Disable tree-based hierarchical RCU auto-balancing"
400 depends on TREE_RCU || TREE_PREEMPT_RCU
403 This option forces use of the exact RCU_FANOUT value specified,
404 regardless of imbalances in the hierarchy. This is useful for
405 testing RCU itself, and might one day be useful on systems with
406 strong NUMA behavior.
408 Without RCU_FANOUT_EXACT, the code will balance the hierarchy.
412 config RCU_FAST_NO_HZ
413 bool "Accelerate last non-dyntick-idle CPU's grace periods"
414 depends on TREE_RCU && NO_HZ && SMP
417 This option causes RCU to attempt to accelerate grace periods
418 in order to allow the final CPU to enter dynticks-idle state
419 more quickly. On the other hand, this option increases the
420 overhead of the dynticks-idle checking, particularly on systems
421 with large numbers of CPUs.
423 Say Y if energy efficiency is critically important, particularly
424 if you have relatively few CPUs.
426 Say N if you are unsure.
428 config TREE_RCU_TRACE
429 def_bool RCU_TRACE && ( TREE_RCU || TREE_PREEMPT_RCU )
432 This option provides tracing for the TREE_RCU and
433 TREE_PREEMPT_RCU implementations, permitting Makefile to
434 trivially select kernel/rcutree_trace.c.
436 endmenu # "RCU Subsystem"
439 tristate "Kernel .config support"
441 This option enables the complete Linux kernel ".config" file
442 contents to be saved in the kernel. It provides documentation
443 of which kernel options are used in a running kernel or in an
444 on-disk kernel. This information can be extracted from the kernel
445 image file with the script scripts/extract-ikconfig and used as
446 input to rebuild the current kernel or to build another kernel.
447 It can also be extracted from a running kernel by reading
448 /proc/config.gz if enabled (below).
451 bool "Enable access to .config through /proc/config.gz"
452 depends on IKCONFIG && PROC_FS
454 This option enables access to the kernel configuration file
455 through /proc/config.gz.
458 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
462 Select kernel log buffer size as a power of 2.
472 # Architectures with an unreliable sched_clock() should select this:
474 config HAVE_UNSTABLE_SCHED_CLOCK
478 boolean "Control Group support"
481 This option adds support for grouping sets of processes together, for
482 use with process control subsystems such as Cpusets, CFS, memory
483 controls or device isolation.
485 - Documentation/scheduler/sched-design-CFS.txt (CFS)
486 - Documentation/cgroups/ (features for grouping, isolation
487 and resource control)
494 bool "Example debug cgroup subsystem"
498 This option enables a simple cgroup subsystem that
499 exports useful debugging information about the cgroups
505 bool "Namespace cgroup subsystem"
508 Provides a simple namespace cgroup subsystem to
509 provide hierarchical naming of sets of namespaces,
510 for instance virtual servers and checkpoint/restart
513 config CGROUP_FREEZER
514 bool "Freezer cgroup subsystem"
517 Provides a way to freeze and unfreeze all tasks in a
521 bool "Device controller for cgroups"
522 depends on CGROUPS && EXPERIMENTAL
524 Provides a cgroup implementing whitelists for devices which
525 a process in the cgroup can mknod or open.
528 bool "Cpuset support"
531 This option will let you create and manage CPUSETs which
532 allow dynamically partitioning a system into sets of CPUs and
533 Memory Nodes and assigning tasks to run only within those sets.
534 This is primarily useful on large SMP or NUMA systems.
538 config PROC_PID_CPUSET
539 bool "Include legacy /proc/<pid>/cpuset file"
543 config CGROUP_CPUACCT
544 bool "Simple CPU accounting cgroup subsystem"
547 Provides a simple Resource Controller for monitoring the
548 total CPU consumed by the tasks in a cgroup.
550 config RESOURCE_COUNTERS
551 bool "Resource counters"
553 This option enables controller independent resource accounting
554 infrastructure that works with cgroups.
557 config CGROUP_MEM_RES_CTLR
558 bool "Memory Resource Controller for Control Groups"
559 depends on CGROUPS && RESOURCE_COUNTERS
562 Provides a memory resource controller that manages both anonymous
563 memory and page cache. (See Documentation/cgroups/memory.txt)
565 Note that setting this option increases fixed memory overhead
566 associated with each page of memory in the system. By this,
567 20(40)bytes/PAGE_SIZE on 32(64)bit system will be occupied by memory
568 usage tracking struct at boot. Total amount of this is printed out
571 Only enable when you're ok with these trade offs and really
572 sure you need the memory resource controller. Even when you enable
573 this, you can set "cgroup_disable=memory" at your boot option to
574 disable memory resource controller and you can avoid overheads.
575 (and lose benefits of memory resource controller)
577 This config option also selects MM_OWNER config option, which
578 could in turn add some fork/exit overhead.
580 config CGROUP_MEM_RES_CTLR_SWAP
581 bool "Memory Resource Controller Swap Extension"
582 depends on CGROUP_MEM_RES_CTLR && SWAP
584 Add swap management feature to memory resource controller. When you
585 enable this, you can limit mem+swap usage per cgroup. In other words,
586 when you disable this, memory resource controller has no cares to
587 usage of swap...a process can exhaust all of the swap. This extension
588 is useful when you want to avoid exhaustion swap but this itself
589 adds more overheads and consumes memory for remembering information.
590 Especially if you use 32bit system or small memory system, please
591 be careful about enabling this. When memory resource controller
592 is disabled by boot option, this will be automatically disabled and
593 there will be no overhead from this. Even when you set this config=y,
594 if boot option "noswapaccount" is set, swap will not be accounted.
595 Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page
596 size is 4096bytes, 512k per 1Gbytes of swap.
598 menuconfig CGROUP_SCHED
599 bool "Group CPU scheduler"
600 depends on EXPERIMENTAL && CGROUPS
603 This feature lets CPU scheduler recognize task groups and control CPU
604 bandwidth allocation to such task groups. It uses cgroups to group
608 config FAIR_GROUP_SCHED
609 bool "Group scheduling for SCHED_OTHER"
610 depends on CGROUP_SCHED
613 config RT_GROUP_SCHED
614 bool "Group scheduling for SCHED_RR/FIFO"
615 depends on EXPERIMENTAL
616 depends on CGROUP_SCHED
619 This feature lets you explicitly allocate real CPU bandwidth
620 to task groups. If enabled, it will also make it impossible to
621 schedule realtime tasks for non-root users until you allocate
622 realtime bandwidth for them.
623 See Documentation/scheduler/sched-rt-group.txt for more information.
628 tristate "Block IO controller"
629 depends on CGROUPS && BLOCK
632 Generic block IO controller cgroup interface. This is the common
633 cgroup interface which should be used by various IO controlling
636 Currently, CFQ IO scheduler uses it to recognize task groups and
637 control disk bandwidth allocation (proportional time slice allocation)
640 This option only enables generic Block IO controller infrastructure.
641 One needs to also enable actual IO controlling logic in CFQ for it
642 to take effect. (CONFIG_CFQ_GROUP_IOSCHED=y).
644 See Documentation/cgroups/blkio-controller.txt for more information.
646 config DEBUG_BLK_CGROUP
647 bool "Enable Block IO controller debugging"
648 depends on BLK_CGROUP
651 Enable some debugging help. Currently it exports additional stat
652 files in a cgroup which can be useful for debugging.
659 config SYSFS_DEPRECATED
662 config SYSFS_DEPRECATED_V2
663 bool "enable deprecated sysfs features to support old userspace tools"
666 select SYSFS_DEPRECATED
668 This option switches the layout of sysfs to the deprecated
669 version. Do not use it on recent distributions.
671 The current sysfs layout features a unified device tree at
672 /sys/devices/, which is able to express a hierarchy between
673 class devices. If the deprecated option is set to Y, the
674 unified device tree is split into a bus device tree at
675 /sys/devices/ and several individual class device trees at
676 /sys/class/. The class and bus devices will be connected by
677 "<subsystem>:<name>" and the "device" links. The "block"
678 class devices, will not show up in /sys/class/block/. Some
679 subsystems will suppress the creation of some devices which
680 depend on the unified device tree.
682 This option is not a pure compatibility option that can
683 be safely enabled on newer distributions. It will change the
684 layout of sysfs to the non-extensible deprecated version,
685 and disable some features, which can not be exported without
686 confusing older userspace tools. Since 2007/2008 all major
687 distributions do not enable this option, and ship no tools which
688 depend on the deprecated layout or this option.
690 If you are using a new kernel on an older distribution, or use
691 older userspace tools, you might need to say Y here. Do not say Y,
692 if the original kernel, that came with your distribution, has
693 this option set to N.
696 bool "Kernel->user space relay support (formerly relayfs)"
698 This option enables support for relay interface support in
699 certain file systems (such as debugfs).
700 It is designed to provide an efficient mechanism for tools and
701 facilities to relay large amounts of data from kernel space to
707 bool "Namespaces support" if EMBEDDED
710 Provides the way to make tasks work with different objects using
711 the same id. For example same IPC id may refer to different objects
712 or same user id or pid may refer to different tasks when used in
713 different namespaces.
717 depends on NAMESPACES
719 In this namespace tasks see different info provided with the
724 depends on NAMESPACES && (SYSVIPC || POSIX_MQUEUE)
726 In this namespace tasks work with IPC ids which correspond to
727 different IPC objects in different namespaces.
730 bool "User namespace (EXPERIMENTAL)"
731 depends on NAMESPACES && EXPERIMENTAL
733 This allows containers, i.e. vservers, to use user namespaces
734 to provide different user info for different servers.
738 bool "PID Namespaces (EXPERIMENTAL)"
740 depends on NAMESPACES && EXPERIMENTAL
742 Support process id namespaces. This allows having multiple
743 processes with the same pid as long as they are in different
744 pid namespaces. This is a building block of containers.
746 Unless you want to work with an experimental feature
750 bool "Network namespace"
752 depends on NAMESPACES && EXPERIMENTAL && NET
754 Allow user space to create what appear to be multiple instances
755 of the network stack.
757 config BLK_DEV_INITRD
758 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
759 depends on BROKEN || !FRV
761 The initial RAM filesystem is a ramfs which is loaded by the
762 boot loader (loadlin or lilo) and that is mounted as root
763 before the normal boot procedure. It is typically used to
764 load modules needed to mount the "real" root file system,
765 etc. See <file:Documentation/initrd.txt> for details.
767 If RAM disk support (BLK_DEV_RAM) is also included, this
768 also enables initial RAM disk (initrd) support and adds
769 15 Kbytes (more on some other architectures) to the kernel size.
779 config CC_OPTIMIZE_FOR_SIZE
780 bool "Optimize for size"
783 Enabling this option will pass "-Os" instead of "-O2" to gcc
784 resulting in a smaller kernel.
795 bool "Configure standard kernel features (for small systems)"
797 This option allows certain base kernel options and settings
798 to be disabled or tweaked. This is for specialized
799 environments which can tolerate a "non-standard" kernel.
800 Only use this if you really know what you are doing.
803 bool "Enable 16-bit UID system calls" if EMBEDDED
804 depends on ARM || BLACKFIN || CRIS || FRV || H8300 || X86_32 || M68K || (S390 && !64BIT) || SUPERH || SPARC32 || (SPARC64 && COMPAT) || UML || (X86_64 && IA32_EMULATION)
807 This enables the legacy 16-bit UID syscall wrappers.
809 config SYSCTL_SYSCALL
810 bool "Sysctl syscall support" if EMBEDDED
811 depends on PROC_SYSCTL
815 sys_sysctl uses binary paths that have been found challenging
816 to properly maintain and use. The interface in /proc/sys
817 using paths with ascii names is now the primary path to this
820 Almost nothing using the binary sysctl interface so if you are
821 trying to save some space it is probably safe to disable this,
822 making your kernel marginally smaller.
824 If unsure say Y here.
827 bool "Load all symbols for debugging/ksymoops" if EMBEDDED
830 Say Y here to let the kernel print out symbolic crash information and
831 symbolic stack backtraces. This increases the size of the kernel
832 somewhat, as all symbols have to be loaded into the kernel image.
835 bool "Include all symbols in kallsyms"
836 depends on DEBUG_KERNEL && KALLSYMS
838 Normally kallsyms only contains the symbols of functions, for nicer
839 OOPS messages. Some debuggers can use kallsyms for other
840 symbols too: say Y here to include all symbols, if you need them
841 and you don't care about adding 300k to the size of your kernel.
845 config KALLSYMS_EXTRA_PASS
846 bool "Do an extra kallsyms pass"
849 If kallsyms is not working correctly, the build will fail with
850 inconsistent kallsyms data. If that occurs, log a bug report and
851 turn on KALLSYMS_EXTRA_PASS which should result in a stable build.
852 Always say N here unless you find a bug in kallsyms, which must be
853 reported. KALLSYMS_EXTRA_PASS is only a temporary workaround while
854 you wait for kallsyms to be fixed.
858 bool "Support for hot-pluggable devices" if EMBEDDED
861 This option is provided for the case where no hotplug or uevent
862 capabilities is wanted by the kernel. You should only consider
863 disabling this option for embedded systems that do not use modules, a
864 dynamic /dev tree, or dynamic device discovery. Just say Y.
868 bool "Enable support for printk" if EMBEDDED
870 This option enables normal printk support. Removing it
871 eliminates most of the message strings from the kernel image
872 and makes the kernel more or less silent. As this makes it
873 very difficult to diagnose system problems, saying N here is
874 strongly discouraged.
877 bool "BUG() support" if EMBEDDED
880 Disabling this option eliminates support for BUG and WARN, reducing
881 the size of your kernel image and potentially quietly ignoring
882 numerous fatal conditions. You should only consider disabling this
883 option for embedded systems with no facilities for reporting errors.
888 bool "Enable ELF core dumps" if EMBEDDED
890 Enable support for generating core dumps. Disabling saves about 4k.
892 config PCSPKR_PLATFORM
893 bool "Enable PC-Speaker support" if EMBEDDED
894 depends on ALPHA || X86 || MIPS || PPC_PREP || PPC_CHRP || PPC_PSERIES
897 This option allows to disable the internal PC-Speaker
898 support, saving some memory.
902 bool "Enable full-sized data structures for core" if EMBEDDED
904 Disabling this option reduces the size of miscellaneous core
905 kernel data structures. This saves memory on small machines,
906 but may reduce performance.
909 bool "Enable futex support" if EMBEDDED
913 Disabling this option will cause the kernel to be built without
914 support for "fast userspace mutexes". The resulting kernel may not
915 run glibc-based applications correctly.
918 bool "Enable eventpoll support" if EMBEDDED
922 Disabling this option will cause the kernel to be built without
923 support for epoll family of system calls.
926 bool "Enable signalfd() system call" if EMBEDDED
930 Enable the signalfd() system call that allows to receive signals
931 on a file descriptor.
936 bool "Enable timerfd() system call" if EMBEDDED
940 Enable the timerfd() system call that allows to receive timer
941 events on a file descriptor.
946 bool "Enable eventfd() system call" if EMBEDDED
950 Enable the eventfd() system call that allows to receive both
951 kernel notification (ie. KAIO) or userspace notifications.
956 bool "Use full shmem filesystem" if EMBEDDED
960 The shmem is an internal filesystem used to manage shared memory.
961 It is backed by swap and manages resource limits. It is also exported
962 to userspace as tmpfs if TMPFS is enabled. Disabling this
963 option replaces shmem and tmpfs with the much simpler ramfs code,
964 which may be appropriate on small systems without swap.
967 bool "Enable AIO support" if EMBEDDED
970 This option enables POSIX asynchronous I/O which may by used
971 by some high performance threaded applications. Disabling
972 this option saves about 7k.
974 config HAVE_PERF_EVENTS
977 See tools/perf/design.txt for details.
979 config PERF_USE_VMALLOC
982 See tools/perf/design.txt for details
984 menu "Kernel Performance Events And Counters"
987 bool "Kernel performance events and counters"
988 default y if (PROFILING || PERF_COUNTERS)
989 depends on HAVE_PERF_EVENTS
992 Enable kernel support for various performance events provided
993 by software and hardware.
995 Software events are supported either built-in or via the
996 use of generic tracepoints.
998 Most modern CPUs support performance events via performance
999 counter registers. These registers count the number of certain
1000 types of hw events: such as instructions executed, cachemisses
1001 suffered, or branches mis-predicted - without slowing down the
1002 kernel or applications. These registers can also trigger interrupts
1003 when a threshold number of events have passed - and can thus be
1004 used to profile the code that runs on that CPU.
1006 The Linux Performance Event subsystem provides an abstraction of
1007 these software and hardware event capabilities, available via a
1008 system call and used by the "perf" utility in tools/perf/. It
1009 provides per task and per CPU counters, and it provides event
1010 capabilities on top of those.
1014 config PERF_COUNTERS
1015 bool "Kernel performance counters (old config option)"
1016 depends on HAVE_PERF_EVENTS
1018 This config has been obsoleted by the PERF_EVENTS
1019 config option - please see that one for details.
1021 It has no effect on the kernel whether you enable
1022 it or not, it is a compatibility placeholder.
1026 config DEBUG_PERF_USE_VMALLOC
1028 bool "Debug: use vmalloc to back perf mmap() buffers"
1029 depends on PERF_EVENTS && DEBUG_KERNEL
1030 select PERF_USE_VMALLOC
1032 Use vmalloc memory to back perf mmap() buffers.
1034 Mostly useful for debugging the vmalloc code on platforms
1035 that don't require it.
1041 config VM_EVENT_COUNTERS
1043 bool "Enable VM event counters for /proc/vmstat" if EMBEDDED
1045 VM event counters are needed for event counts to be shown.
1046 This option allows the disabling of the VM event counters
1047 on EMBEDDED systems. /proc/vmstat will only show page counts
1048 if VM event counters are disabled.
1052 bool "Enable PCI quirk workarounds" if EMBEDDED
1055 This enables workarounds for various PCI chipset
1056 bugs/quirks. Disable this only if your target machine is
1057 unaffected by PCI quirks.
1061 bool "Enable SLUB debugging support" if EMBEDDED
1062 depends on SLUB && SYSFS
1064 SLUB has extensive debug support features. Disabling these can
1065 result in significant savings in code size. This also disables
1066 SLUB sysfs support. /sys/slab will not exist and there will be
1067 no support for cache validation etc.
1070 bool "Disable heap randomization"
1073 Randomizing heap placement makes heap exploits harder, but it
1074 also breaks ancient binaries (including anything libc5 based).
1075 This option changes the bootup default to heap randomization
1076 disabled, and can be overridden at runtime by setting
1077 /proc/sys/kernel/randomize_va_space to 2.
1079 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1082 prompt "Choose SLAB allocator"
1085 This option allows to select a slab allocator.
1090 The regular slab allocator that is established and known to work
1091 well in all environments. It organizes cache hot objects in
1092 per cpu and per node queues.
1095 bool "SLUB (Unqueued Allocator)"
1097 SLUB is a slab allocator that minimizes cache line usage
1098 instead of managing queues of cached objects (SLAB approach).
1099 Per cpu caching is realized using slabs of objects instead
1100 of queues of objects. SLUB can use memory efficiently
1101 and has enhanced diagnostics. SLUB is the default choice for
1106 bool "SLOB (Simple Allocator)"
1108 SLOB replaces the stock allocator with a drastically simpler
1109 allocator. SLOB is generally more space efficient but
1110 does not perform as well on large systems.
1114 config MMAP_ALLOW_UNINITIALIZED
1115 bool "Allow mmapped anonymous memory to be uninitialized"
1116 depends on EMBEDDED && !MMU
1119 Normally, and according to the Linux spec, anonymous memory obtained
1120 from mmap() has it's contents cleared before it is passed to
1121 userspace. Enabling this config option allows you to request that
1122 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
1123 providing a huge performance boost. If this option is not enabled,
1124 then the flag will be ignored.
1126 This is taken advantage of by uClibc's malloc(), and also by
1127 ELF-FDPIC binfmt's brk and stack allocator.
1129 Because of the obvious security issues, this option should only be
1130 enabled on embedded devices where you control what is run in
1131 userspace. Since that isn't generally a problem on no-MMU systems,
1132 it is normally safe to say Y here.
1134 See Documentation/nommu-mmap.txt for more information.
1137 bool "Profiling support"
1139 Say Y here to enable the extended profiling support mechanisms used
1140 by profilers such as OProfile.
1143 # Place an empty function call at each tracepoint site. Can be
1144 # dynamically changed for a probe function.
1149 source "arch/Kconfig"
1151 endmenu # General setup
1153 config HAVE_GENERIC_DMA_COHERENT
1160 depends on SLAB || SLUB_DEBUG
1168 default 0 if BASE_FULL
1169 default 1 if !BASE_FULL
1172 bool "Enable loadable module support"
1174 Kernel modules are small pieces of compiled code which can
1175 be inserted in the running kernel, rather than being
1176 permanently built into the kernel. You use the "modprobe"
1177 tool to add (and sometimes remove) them. If you say Y here,
1178 many parts of the kernel can be built as modules (by
1179 answering M instead of Y where indicated): this is most
1180 useful for infrequently used options which are not required
1181 for booting. For more information, see the man pages for
1182 modprobe, lsmod, modinfo, insmod and rmmod.
1184 If you say Y here, you will need to run "make
1185 modules_install" to put the modules under /lib/modules/
1186 where modprobe can find them (you may need to be root to do
1193 config MODULE_FORCE_LOAD
1194 bool "Forced module loading"
1197 Allow loading of modules without version information (ie. modprobe
1198 --force). Forced module loading sets the 'F' (forced) taint flag and
1199 is usually a really bad idea.
1201 config MODULE_UNLOAD
1202 bool "Module unloading"
1204 Without this option you will not be able to unload any
1205 modules (note that some modules may not be unloadable
1206 anyway), which makes your kernel smaller, faster
1207 and simpler. If unsure, say Y.
1209 config MODULE_FORCE_UNLOAD
1210 bool "Forced module unloading"
1211 depends on MODULE_UNLOAD && EXPERIMENTAL
1213 This option allows you to force a module to unload, even if the
1214 kernel believes it is unsafe: the kernel will remove the module
1215 without waiting for anyone to stop using it (using the -f option to
1216 rmmod). This is mainly for kernel developers and desperate users.
1220 bool "Module versioning support"
1222 Usually, you have to use modules compiled with your kernel.
1223 Saying Y here makes it sometimes possible to use modules
1224 compiled for different kernels, by adding enough information
1225 to the modules to (hopefully) spot any changes which would
1226 make them incompatible with the kernel you are running. If
1229 config MODULE_SRCVERSION_ALL
1230 bool "Source checksum for all modules"
1232 Modules which contain a MODULE_VERSION get an extra "srcversion"
1233 field inserted into their modinfo section, which contains a
1234 sum of the source files which made it. This helps maintainers
1235 see exactly which source was used to build a module (since
1236 others sometimes change the module source without updating
1237 the version). With this option, such a "srcversion" field
1238 will be created for all modules. If unsure, say N.
1242 config INIT_ALL_POSSIBLE
1245 Back when each arch used to define their own cpu_online_map and
1246 cpu_possible_map, some of them chose to initialize cpu_possible_map
1247 with all 1s, and others with all 0s. When they were centralised,
1248 it was better to provide this option than to break all the archs
1249 and have several arch maintainers pursuing me down dark alleys.
1254 depends on (SMP && MODULE_UNLOAD) || HOTPLUG_CPU
1256 Need stop_machine() primitive.
1258 source "block/Kconfig"
1260 config PREEMPT_NOTIFIERS
1267 source "kernel/Kconfig.locks"