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47 drivers that have been replaced by something else, and/or are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 "Local version - append to kernel release"
82 Append an extra string to the end of your kernel version.
83 This will show up when you type uname, for example.
84 The string you set here will be appended after the contents of
85 any files with a filename matching localversion* in your
86 object and source tree, in that order. Your total string can
87 be a maximum of 64 characters.
89 config LOCALVERSION_AUTO
90 bool "Automatically append version information to the version string"
93 This will try to automatically determine if the current tree is a
94 release tree by looking for git tags that belong to the current
97 A string of the format -gxxxxxxxx will be added to the localversion
98 if a git-based tree is found. The string generated by this will be
99 appended after any matching localversion* files, and after the value
100 set in CONFIG_LOCALVERSION.
102 (The actual string used here is the first eight characters produced
103 by running the command:
105 $ git rev-parse --verify HEAD
107 which is done within the script "scripts/setlocalversion".)
109 config HAVE_KERNEL_GZIP
112 config HAVE_KERNEL_BZIP2
115 config HAVE_KERNEL_LZMA
119 prompt "Kernel compression mode"
121 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA
123 The linux kernel is a kind of self-extracting executable.
124 Several compression algorithms are available, which differ
125 in efficiency, compression and decompression speed.
126 Compression speed is only relevant when building a kernel.
127 Decompression speed is relevant at each boot.
129 If you have any problems with bzip2 or lzma compressed
130 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
131 version of this functionality (bzip2 only), for 2.4, was
132 supplied by Christian Ludwig)
134 High compression options are mostly useful for users, who
135 are low on disk space (embedded systems), but for whom ram
138 If in doubt, select 'gzip'
142 depends on HAVE_KERNEL_GZIP
144 The old and tried gzip compression. Its compression ratio is
145 the poorest among the 3 choices; however its speed (both
146 compression and decompression) is the fastest.
150 depends on HAVE_KERNEL_BZIP2
152 Its compression ratio and speed is intermediate.
153 Decompression speed is slowest among the three. The kernel
154 size is about 10% smaller with bzip2, in comparison to gzip.
155 Bzip2 uses a large amount of memory. For modern kernels you
156 will need at least 8MB RAM or more for booting.
160 depends on HAVE_KERNEL_LZMA
162 The most recent compression algorithm.
163 Its ratio is best, decompression speed is between the other
164 two. Compression is slowest. The kernel size is about 33%
165 smaller with LZMA in comparison to gzip.
170 bool "Support for paging of anonymous memory (swap)"
171 depends on MMU && BLOCK
174 This option allows you to choose whether you want to have support
175 for so called swap devices or swap files in your kernel that are
176 used to provide more virtual memory than the actual RAM present
177 in your computer. If unsure say Y.
182 Inter Process Communication is a suite of library functions and
183 system calls which let processes (running programs) synchronize and
184 exchange information. It is generally considered to be a good thing,
185 and some programs won't run unless you say Y here. In particular, if
186 you want to run the DOS emulator dosemu under Linux (read the
187 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
188 you'll need to say Y here.
190 You can find documentation about IPC with "info ipc" and also in
191 section 6.4 of the Linux Programmer's Guide, available from
192 <http://www.tldp.org/guides.html>.
194 config SYSVIPC_SYSCTL
201 bool "POSIX Message Queues"
202 depends on NET && EXPERIMENTAL
204 POSIX variant of message queues is a part of IPC. In POSIX message
205 queues every message has a priority which decides about succession
206 of receiving it by a process. If you want to compile and run
207 programs written e.g. for Solaris with use of its POSIX message
208 queues (functions mq_*) say Y here.
210 POSIX message queues are visible as a filesystem called 'mqueue'
211 and can be mounted somewhere if you want to do filesystem
212 operations on message queues.
216 config POSIX_MQUEUE_SYSCTL
218 depends on POSIX_MQUEUE
222 config BSD_PROCESS_ACCT
223 bool "BSD Process Accounting"
225 If you say Y here, a user level program will be able to instruct the
226 kernel (via a special system call) to write process accounting
227 information to a file: whenever a process exits, information about
228 that process will be appended to the file by the kernel. The
229 information includes things such as creation time, owning user,
230 command name, memory usage, controlling terminal etc. (the complete
231 list is in the struct acct in <file:include/linux/acct.h>). It is
232 up to the user level program to do useful things with this
233 information. This is generally a good idea, so say Y.
235 config BSD_PROCESS_ACCT_V3
236 bool "BSD Process Accounting version 3 file format"
237 depends on BSD_PROCESS_ACCT
240 If you say Y here, the process accounting information is written
241 in a new file format that also logs the process IDs of each
242 process and it's parent. Note that this file format is incompatible
243 with previous v0/v1/v2 file formats, so you will need updated tools
244 for processing it. A preliminary version of these tools is available
245 at <http://www.gnu.org/software/acct/>.
248 bool "Export task/process statistics through netlink (EXPERIMENTAL)"
252 Export selected statistics for tasks/processes through the
253 generic netlink interface. Unlike BSD process accounting, the
254 statistics are available during the lifetime of tasks/processes as
255 responses to commands. Like BSD accounting, they are sent to user
260 config TASK_DELAY_ACCT
261 bool "Enable per-task delay accounting (EXPERIMENTAL)"
264 Collect information on time spent by a task waiting for system
265 resources like cpu, synchronous block I/O completion and swapping
266 in pages. Such statistics can help in setting a task's priorities
267 relative to other tasks for cpu, io, rss limits etc.
272 bool "Enable extended accounting over taskstats (EXPERIMENTAL)"
275 Collect extended task accounting data and send the data
276 to userland for processing over the taskstats interface.
280 config TASK_IO_ACCOUNTING
281 bool "Enable per-task storage I/O accounting (EXPERIMENTAL)"
282 depends on TASK_XACCT
284 Collect information on the number of bytes of storage I/O which this
290 bool "Auditing support"
293 Enable auditing infrastructure that can be used with another
294 kernel subsystem, such as SELinux (which requires this for
295 logging of avc messages output). Does not do system-call
296 auditing without CONFIG_AUDITSYSCALL.
299 bool "Enable system-call auditing support"
300 depends on AUDIT && (X86 || PPC || PPC64 || S390 || IA64 || UML || SPARC64|| SUPERH)
301 default y if SECURITY_SELINUX
303 Enable low-overhead system-call auditing infrastructure that
304 can be used independently or with another kernel subsystem,
305 such as SELinux. To use audit's filesystem watch feature, please
306 ensure that INOTIFY is configured.
310 depends on AUDITSYSCALL
316 prompt "RCU Implementation"
320 bool "Tree-based hierarchical RCU"
322 This option selects the RCU implementation that is
323 designed for very large SMP system with hundreds or
324 thousands of CPUs. It also scales down nicely to
328 bool "Preemptible RCU"
331 This option reduces the latency of the kernel by making certain
332 RCU sections preemptible. Normally RCU code is non-preemptible, if
333 this option is selected then read-only RCU sections become
334 preemptible. This helps latency, but may expose bugs due to
335 now-naive assumptions about each RCU read-side critical section
336 remaining on a given CPU through its execution.
341 bool "Enable tracing for RCU"
342 depends on TREE_RCU || PREEMPT_RCU
344 This option provides tracing in RCU which presents stats
345 in debugfs for debugging RCU implementation.
347 Say Y here if you want to enable RCU tracing
348 Say N if you are unsure.
351 int "Tree-based hierarchical RCU fanout value"
358 This option controls the fanout of hierarchical implementations
359 of RCU, allowing RCU to work efficiently on machines with
360 large numbers of CPUs. This value must be at least the cube
361 root of NR_CPUS, which allows NR_CPUS up to 32,768 for 32-bit
362 systems and up to 262,144 for 64-bit systems.
364 Select a specific number if testing RCU itself.
365 Take the default if unsure.
367 config RCU_FANOUT_EXACT
368 bool "Disable tree-based hierarchical RCU auto-balancing"
372 This option forces use of the exact RCU_FANOUT value specified,
373 regardless of imbalances in the hierarchy. This is useful for
374 testing RCU itself, and might one day be useful on systems with
375 strong NUMA behavior.
377 Without RCU_FANOUT_EXACT, the code will balance the hierarchy.
381 config TREE_RCU_TRACE
382 def_bool RCU_TRACE && TREE_RCU
385 This option provides tracing for the TREE_RCU implementation,
386 permitting Makefile to trivially select kernel/rcutree_trace.c.
388 config PREEMPT_RCU_TRACE
389 def_bool RCU_TRACE && PREEMPT_RCU
392 This option provides tracing for the PREEMPT_RCU implementation,
393 permitting Makefile to trivially select kernel/rcupreempt_trace.c.
395 endmenu # "RCU Subsystem"
398 tristate "Kernel .config support"
400 This option enables the complete Linux kernel ".config" file
401 contents to be saved in the kernel. It provides documentation
402 of which kernel options are used in a running kernel or in an
403 on-disk kernel. This information can be extracted from the kernel
404 image file with the script scripts/extract-ikconfig and used as
405 input to rebuild the current kernel or to build another kernel.
406 It can also be extracted from a running kernel by reading
407 /proc/config.gz if enabled (below).
410 bool "Enable access to .config through /proc/config.gz"
411 depends on IKCONFIG && PROC_FS
413 This option enables access to the kernel configuration file
414 through /proc/config.gz.
417 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
421 Select kernel log buffer size as a power of 2.
431 # Architectures with an unreliable sched_clock() should select this:
433 config HAVE_UNSTABLE_SCHED_CLOCK
437 bool "Group CPU scheduler"
438 depends on EXPERIMENTAL
441 This feature lets CPU scheduler recognize task groups and control CPU
442 bandwidth allocation to such task groups.
443 In order to create a group from arbitrary set of processes, use
444 CONFIG_CGROUPS. (See Control Group support.)
446 config FAIR_GROUP_SCHED
447 bool "Group scheduling for SCHED_OTHER"
448 depends on GROUP_SCHED
451 config RT_GROUP_SCHED
452 bool "Group scheduling for SCHED_RR/FIFO"
453 depends on EXPERIMENTAL
454 depends on GROUP_SCHED
457 This feature lets you explicitly allocate real CPU bandwidth
458 to users or control groups (depending on the "Basis for grouping tasks"
459 setting below. If enabled, it will also make it impossible to
460 schedule realtime tasks for non-root users until you allocate
461 realtime bandwidth for them.
462 See Documentation/scheduler/sched-rt-group.txt for more information.
465 depends on GROUP_SCHED
466 prompt "Basis for grouping tasks"
472 This option will choose userid as the basis for grouping
473 tasks, thus providing equal CPU bandwidth to each user.
476 bool "Control groups"
479 This option allows you to create arbitrary task groups
480 using the "cgroup" pseudo filesystem and control
481 the cpu bandwidth allocated to each such task group.
482 Refer to Documentation/cgroups/cgroups.txt for more
483 information on "cgroup" pseudo filesystem.
488 boolean "Control Group support"
490 This option adds support for grouping sets of processes together, for
491 use with process control subsystems such as Cpusets, CFS, memory
492 controls or device isolation.
494 - Documentation/scheduler/sched-design-CFS.txt (CFS)
495 - Documentation/cgroups/ (features for grouping, isolation
496 and resource control)
503 bool "Example debug cgroup subsystem"
507 This option enables a simple cgroup subsystem that
508 exports useful debugging information about the cgroups
514 bool "Namespace cgroup subsystem"
517 Provides a simple namespace cgroup subsystem to
518 provide hierarchical naming of sets of namespaces,
519 for instance virtual servers and checkpoint/restart
522 config CGROUP_FREEZER
523 bool "Freezer cgroup subsystem"
526 Provides a way to freeze and unfreeze all tasks in a
530 bool "Device controller for cgroups"
531 depends on CGROUPS && EXPERIMENTAL
533 Provides a cgroup implementing whitelists for devices which
534 a process in the cgroup can mknod or open.
537 bool "Cpuset support"
540 This option will let you create and manage CPUSETs which
541 allow dynamically partitioning a system into sets of CPUs and
542 Memory Nodes and assigning tasks to run only within those sets.
543 This is primarily useful on large SMP or NUMA systems.
547 config PROC_PID_CPUSET
548 bool "Include legacy /proc/<pid>/cpuset file"
552 config CGROUP_CPUACCT
553 bool "Simple CPU accounting cgroup subsystem"
556 Provides a simple Resource Controller for monitoring the
557 total CPU consumed by the tasks in a cgroup.
559 config RESOURCE_COUNTERS
560 bool "Resource counters"
562 This option enables controller independent resource accounting
563 infrastructure that works with cgroups.
566 config CGROUP_MEM_RES_CTLR
567 bool "Memory Resource Controller for Control Groups"
568 depends on CGROUPS && RESOURCE_COUNTERS
571 Provides a memory resource controller that manages both anonymous
572 memory and page cache. (See Documentation/cgroups/memory.txt)
574 Note that setting this option increases fixed memory overhead
575 associated with each page of memory in the system. By this,
576 20(40)bytes/PAGE_SIZE on 32(64)bit system will be occupied by memory
577 usage tracking struct at boot. Total amount of this is printed out
580 Only enable when you're ok with these trade offs and really
581 sure you need the memory resource controller. Even when you enable
582 this, you can set "cgroup_disable=memory" at your boot option to
583 disable memory resource controller and you can avoid overheads.
584 (and lose benefits of memory resource controller)
586 This config option also selects MM_OWNER config option, which
587 could in turn add some fork/exit overhead.
589 config CGROUP_MEM_RES_CTLR_SWAP
590 bool "Memory Resource Controller Swap Extension(EXPERIMENTAL)"
591 depends on CGROUP_MEM_RES_CTLR && SWAP && EXPERIMENTAL
593 Add swap management feature to memory resource controller. When you
594 enable this, you can limit mem+swap usage per cgroup. In other words,
595 when you disable this, memory resource controller has no cares to
596 usage of swap...a process can exhaust all of the swap. This extension
597 is useful when you want to avoid exhaustion swap but this itself
598 adds more overheads and consumes memory for remembering information.
599 Especially if you use 32bit system or small memory system, please
600 be careful about enabling this. When memory resource controller
601 is disabled by boot option, this will be automatically disabled and
602 there will be no overhead from this. Even when you set this config=y,
603 if boot option "noswapaccount" is set, swap will not be accounted.
604 Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page
605 size is 4096bytes, 512k per 1Gbytes of swap.
612 config SYSFS_DEPRECATED
615 config SYSFS_DEPRECATED_V2
616 bool "remove sysfs features which may confuse old userspace tools"
619 select SYSFS_DEPRECATED
621 This option switches the layout of sysfs to the deprecated
622 version. Do not use it on recent distributions.
624 The current sysfs layout features a unified device tree at
625 /sys/devices/, which is able to express a hierarchy between
626 class devices. If the deprecated option is set to Y, the
627 unified device tree is split into a bus device tree at
628 /sys/devices/ and several individual class device trees at
629 /sys/class/. The class and bus devices will be connected by
630 "<subsystem>:<name>" and the "device" links. The "block"
631 class devices, will not show up in /sys/class/block/. Some
632 subsystems will suppress the creation of some devices which
633 depend on the unified device tree.
635 This option is not a pure compatibility option that can
636 be safely enabled on newer distributions. It will change the
637 layout of sysfs to the non-extensible deprecated version,
638 and disable some features, which can not be exported without
639 confusing older userspace tools. Since 2007/2008 all major
640 distributions do not enable this option, and ship no tools which
641 depend on the deprecated layout or this option.
643 If you are using a new kernel on an older distribution, or use
644 older userspace tools, you might need to say Y here. Do not say Y,
645 if the original kernel, that came with your distribution, has
646 this option set to N.
649 bool "Kernel->user space relay support (formerly relayfs)"
651 This option enables support for relay interface support in
652 certain file systems (such as debugfs).
653 It is designed to provide an efficient mechanism for tools and
654 facilities to relay large amounts of data from kernel space to
660 bool "Namespaces support" if EMBEDDED
663 Provides the way to make tasks work with different objects using
664 the same id. For example same IPC id may refer to different objects
665 or same user id or pid may refer to different tasks when used in
666 different namespaces.
670 depends on NAMESPACES
672 In this namespace tasks see different info provided with the
677 depends on NAMESPACES && (SYSVIPC || POSIX_MQUEUE)
679 In this namespace tasks work with IPC ids which correspond to
680 different IPC objects in different namespaces.
683 bool "User namespace (EXPERIMENTAL)"
684 depends on NAMESPACES && EXPERIMENTAL
686 This allows containers, i.e. vservers, to use user namespaces
687 to provide different user info for different servers.
691 bool "PID Namespaces (EXPERIMENTAL)"
693 depends on NAMESPACES && EXPERIMENTAL
695 Support process id namespaces. This allows having multiple
696 processes with the same pid as long as they are in different
697 pid namespaces. This is a building block of containers.
699 Unless you want to work with an experimental feature
703 bool "Network namespace"
705 depends on NAMESPACES && EXPERIMENTAL && NET
707 Allow user space to create what appear to be multiple instances
708 of the network stack.
710 config BLK_DEV_INITRD
711 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
712 depends on BROKEN || !FRV
714 The initial RAM filesystem is a ramfs which is loaded by the
715 boot loader (loadlin or lilo) and that is mounted as root
716 before the normal boot procedure. It is typically used to
717 load modules needed to mount the "real" root file system,
718 etc. See <file:Documentation/initrd.txt> for details.
720 If RAM disk support (BLK_DEV_RAM) is also included, this
721 also enables initial RAM disk (initrd) support and adds
722 15 Kbytes (more on some other architectures) to the kernel size.
732 config CC_OPTIMIZE_FOR_SIZE
733 bool "Optimize for size"
736 Enabling this option will pass "-Os" instead of "-O2" to gcc
737 resulting in a smaller kernel.
748 bool "Configure standard kernel features (for small systems)"
750 This option allows certain base kernel options and settings
751 to be disabled or tweaked. This is for specialized
752 environments which can tolerate a "non-standard" kernel.
753 Only use this if you really know what you are doing.
756 bool "Enable 16-bit UID system calls" if EMBEDDED
757 depends on ARM || BLACKFIN || CRIS || FRV || H8300 || X86_32 || M68K || (S390 && !64BIT) || SUPERH || SPARC32 || (SPARC64 && COMPAT) || UML || (X86_64 && IA32_EMULATION)
760 This enables the legacy 16-bit UID syscall wrappers.
762 config SYSCTL_SYSCALL
763 bool "Sysctl syscall support" if EMBEDDED
767 sys_sysctl uses binary paths that have been found challenging
768 to properly maintain and use. The interface in /proc/sys
769 using paths with ascii names is now the primary path to this
772 Almost nothing using the binary sysctl interface so if you are
773 trying to save some space it is probably safe to disable this,
774 making your kernel marginally smaller.
776 If unsure say Y here.
779 bool "Load all symbols for debugging/ksymoops" if EMBEDDED
782 Say Y here to let the kernel print out symbolic crash information and
783 symbolic stack backtraces. This increases the size of the kernel
784 somewhat, as all symbols have to be loaded into the kernel image.
787 bool "Include all symbols in kallsyms"
788 depends on DEBUG_KERNEL && KALLSYMS
790 Normally kallsyms only contains the symbols of functions, for nicer
791 OOPS messages. Some debuggers can use kallsyms for other
792 symbols too: say Y here to include all symbols, if you need them
793 and you don't care about adding 300k to the size of your kernel.
797 config KALLSYMS_EXTRA_PASS
798 bool "Do an extra kallsyms pass"
801 If kallsyms is not working correctly, the build will fail with
802 inconsistent kallsyms data. If that occurs, log a bug report and
803 turn on KALLSYMS_EXTRA_PASS which should result in a stable build.
804 Always say N here unless you find a bug in kallsyms, which must be
805 reported. KALLSYMS_EXTRA_PASS is only a temporary workaround while
806 you wait for kallsyms to be fixed.
810 bool "Support for hot-pluggable devices" if EMBEDDED
813 This option is provided for the case where no hotplug or uevent
814 capabilities is wanted by the kernel. You should only consider
815 disabling this option for embedded systems that do not use modules, a
816 dynamic /dev tree, or dynamic device discovery. Just say Y.
820 bool "Enable support for printk" if EMBEDDED
822 This option enables normal printk support. Removing it
823 eliminates most of the message strings from the kernel image
824 and makes the kernel more or less silent. As this makes it
825 very difficult to diagnose system problems, saying N here is
826 strongly discouraged.
829 bool "BUG() support" if EMBEDDED
832 Disabling this option eliminates support for BUG and WARN, reducing
833 the size of your kernel image and potentially quietly ignoring
834 numerous fatal conditions. You should only consider disabling this
835 option for embedded systems with no facilities for reporting errors.
840 bool "Enable ELF core dumps" if EMBEDDED
842 Enable support for generating core dumps. Disabling saves about 4k.
844 config PCSPKR_PLATFORM
845 bool "Enable PC-Speaker support" if EMBEDDED
846 depends on ALPHA || X86 || MIPS || PPC_PREP || PPC_CHRP || PPC_PSERIES
849 This option allows to disable the internal PC-Speaker
850 support, saving some memory.
854 bool "Enable full-sized data structures for core" if EMBEDDED
856 Disabling this option reduces the size of miscellaneous core
857 kernel data structures. This saves memory on small machines,
858 but may reduce performance.
861 bool "Enable futex support" if EMBEDDED
865 Disabling this option will cause the kernel to be built without
866 support for "fast userspace mutexes". The resulting kernel may not
867 run glibc-based applications correctly.
870 bool "Enable eventpoll support" if EMBEDDED
874 Disabling this option will cause the kernel to be built without
875 support for epoll family of system calls.
878 bool "Enable signalfd() system call" if EMBEDDED
882 Enable the signalfd() system call that allows to receive signals
883 on a file descriptor.
888 bool "Enable timerfd() system call" if EMBEDDED
892 Enable the timerfd() system call that allows to receive timer
893 events on a file descriptor.
898 bool "Enable eventfd() system call" if EMBEDDED
902 Enable the eventfd() system call that allows to receive both
903 kernel notification (ie. KAIO) or userspace notifications.
908 bool "Use full shmem filesystem" if EMBEDDED
912 The shmem is an internal filesystem used to manage shared memory.
913 It is backed by swap and manages resource limits. It is also exported
914 to userspace as tmpfs if TMPFS is enabled. Disabling this
915 option replaces shmem and tmpfs with the much simpler ramfs code,
916 which may be appropriate on small systems without swap.
919 bool "Enable AIO support" if EMBEDDED
922 This option enables POSIX asynchronous I/O which may by used
923 by some high performance threaded applications. Disabling
924 this option saves about 7k.
926 config HAVE_PERF_COUNTERS
929 See tools/perf/design.txt for details.
931 menu "Performance Counters"
934 bool "Kernel Performance Counters"
935 default y if PROFILING
936 depends on HAVE_PERF_COUNTERS
939 Enable kernel support for performance counter hardware.
941 Performance counters are special hardware registers available
942 on most modern CPUs. These registers count the number of certain
943 types of hw events: such as instructions executed, cachemisses
944 suffered, or branches mis-predicted - without slowing down the
945 kernel or applications. These registers can also trigger interrupts
946 when a threshold number of events have passed - and can thus be
947 used to profile the code that runs on that CPU.
949 The Linux Performance Counter subsystem provides an abstraction of
950 these hardware capabilities, available via a system call. It
951 provides per task and per CPU counters, and it provides event
952 capabilities on top of those.
957 bool "Tracepoint profiling sources"
958 depends on PERF_COUNTERS && EVENT_TRACING
961 Allow the use of tracepoints as software performance counters.
963 When this is enabled, you can create perf counters based on
964 tracepoints using PERF_TYPE_TRACEPOINT and the tracepoint ID
965 found in debugfs://tracing/events/*/*/id. (The -e/--events
966 option to the perf tool can parse and interpret symbolic
967 tracepoints, in the subsystem:tracepoint_name format.)
971 config VM_EVENT_COUNTERS
973 bool "Enable VM event counters for /proc/vmstat" if EMBEDDED
975 VM event counters are needed for event counts to be shown.
976 This option allows the disabling of the VM event counters
977 on EMBEDDED systems. /proc/vmstat will only show page counts
978 if VM event counters are disabled.
982 bool "Enable PCI quirk workarounds" if EMBEDDED
985 This enables workarounds for various PCI chipset
986 bugs/quirks. Disable this only if your target machine is
987 unaffected by PCI quirks.
991 bool "Enable SLUB debugging support" if EMBEDDED
992 depends on SLUB && SYSFS
994 SLUB has extensive debug support features. Disabling these can
995 result in significant savings in code size. This also disables
996 SLUB sysfs support. /sys/slab will not exist and there will be
997 no support for cache validation etc.
999 config STRIP_ASM_SYMS
1000 bool "Strip assembler-generated symbols during link"
1003 Strip internal assembler-generated symbols during a link (symbols
1004 that look like '.Lxxx') so they don't pollute the output of
1005 get_wchan() and suchlike.
1008 bool "Disable heap randomization"
1011 Randomizing heap placement makes heap exploits harder, but it
1012 also breaks ancient binaries (including anything libc5 based).
1013 This option changes the bootup default to heap randomization
1014 disabled, and can be overridden at runtime by setting
1015 /proc/sys/kernel/randomize_va_space to 2.
1017 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1020 prompt "Choose SLAB allocator"
1023 This option allows to select a slab allocator.
1028 The regular slab allocator that is established and known to work
1029 well in all environments. It organizes cache hot objects in
1030 per cpu and per node queues.
1033 bool "SLUB (Unqueued Allocator)"
1035 SLUB is a slab allocator that minimizes cache line usage
1036 instead of managing queues of cached objects (SLAB approach).
1037 Per cpu caching is realized using slabs of objects instead
1038 of queues of objects. SLUB can use memory efficiently
1039 and has enhanced diagnostics. SLUB is the default choice for
1044 bool "SLOB (Simple Allocator)"
1046 SLOB replaces the stock allocator with a drastically simpler
1047 allocator. SLOB is generally more space efficient but
1048 does not perform as well on large systems.
1053 bool "Profiling support (EXPERIMENTAL)"
1055 Say Y here to enable the extended profiling support mechanisms used
1056 by profilers such as OProfile.
1059 # Place an empty function call at each tracepoint site. Can be
1060 # dynamically changed for a probe function.
1066 bool "Activate markers"
1069 Place an empty function call at each marker site. Can be
1070 dynamically changed for a probe function.
1072 source "arch/Kconfig"
1078 The slow work thread pool provides a number of dynamically allocated
1079 threads that can be used by the kernel to perform operations that
1080 take a relatively long time.
1082 An example of this would be CacheFiles doing a path lookup followed
1083 by a series of mkdirs and a create call, all of which have to touch
1086 See Documentation/slow-work.txt.
1088 endmenu # General setup
1090 config HAVE_GENERIC_DMA_COHERENT
1097 depends on SLAB || SLUB_DEBUG
1105 default 0 if BASE_FULL
1106 default 1 if !BASE_FULL
1109 bool "Enable loadable module support"
1111 Kernel modules are small pieces of compiled code which can
1112 be inserted in the running kernel, rather than being
1113 permanently built into the kernel. You use the "modprobe"
1114 tool to add (and sometimes remove) them. If you say Y here,
1115 many parts of the kernel can be built as modules (by
1116 answering M instead of Y where indicated): this is most
1117 useful for infrequently used options which are not required
1118 for booting. For more information, see the man pages for
1119 modprobe, lsmod, modinfo, insmod and rmmod.
1121 If you say Y here, you will need to run "make
1122 modules_install" to put the modules under /lib/modules/
1123 where modprobe can find them (you may need to be root to do
1130 config MODULE_FORCE_LOAD
1131 bool "Forced module loading"
1134 Allow loading of modules without version information (ie. modprobe
1135 --force). Forced module loading sets the 'F' (forced) taint flag and
1136 is usually a really bad idea.
1138 config MODULE_UNLOAD
1139 bool "Module unloading"
1141 Without this option you will not be able to unload any
1142 modules (note that some modules may not be unloadable
1143 anyway), which makes your kernel smaller, faster
1144 and simpler. If unsure, say Y.
1146 config MODULE_FORCE_UNLOAD
1147 bool "Forced module unloading"
1148 depends on MODULE_UNLOAD && EXPERIMENTAL
1150 This option allows you to force a module to unload, even if the
1151 kernel believes it is unsafe: the kernel will remove the module
1152 without waiting for anyone to stop using it (using the -f option to
1153 rmmod). This is mainly for kernel developers and desperate users.
1157 bool "Module versioning support"
1159 Usually, you have to use modules compiled with your kernel.
1160 Saying Y here makes it sometimes possible to use modules
1161 compiled for different kernels, by adding enough information
1162 to the modules to (hopefully) spot any changes which would
1163 make them incompatible with the kernel you are running. If
1166 config MODULE_SRCVERSION_ALL
1167 bool "Source checksum for all modules"
1169 Modules which contain a MODULE_VERSION get an extra "srcversion"
1170 field inserted into their modinfo section, which contains a
1171 sum of the source files which made it. This helps maintainers
1172 see exactly which source was used to build a module (since
1173 others sometimes change the module source without updating
1174 the version). With this option, such a "srcversion" field
1175 will be created for all modules. If unsure, say N.
1179 config INIT_ALL_POSSIBLE
1182 Back when each arch used to define their own cpu_online_map and
1183 cpu_possible_map, some of them chose to initialize cpu_possible_map
1184 with all 1s, and others with all 0s. When they were centralised,
1185 it was better to provide this option than to break all the archs
1186 and have several arch maintainers pursuing me down dark alleys.
1191 depends on (SMP && MODULE_UNLOAD) || HOTPLUG_CPU
1193 Need stop_machine() primitive.
1195 source "block/Kconfig"
1197 config PREEMPT_NOTIFIERS