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47 drivers that have been replaced by something else, and/or are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
127 prompt "Kernel compression mode"
129 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA
131 The linux kernel is a kind of self-extracting executable.
132 Several compression algorithms are available, which differ
133 in efficiency, compression and decompression speed.
134 Compression speed is only relevant when building a kernel.
135 Decompression speed is relevant at each boot.
137 If you have any problems with bzip2 or lzma compressed
138 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
139 version of this functionality (bzip2 only), for 2.4, was
140 supplied by Christian Ludwig)
142 High compression options are mostly useful for users, who
143 are low on disk space (embedded systems), but for whom ram
146 If in doubt, select 'gzip'
150 depends on HAVE_KERNEL_GZIP
152 The old and tried gzip compression. Its compression ratio is
153 the poorest among the 3 choices; however its speed (both
154 compression and decompression) is the fastest.
158 depends on HAVE_KERNEL_BZIP2
160 Its compression ratio and speed is intermediate.
161 Decompression speed is slowest among the three. The kernel
162 size is about 10% smaller with bzip2, in comparison to gzip.
163 Bzip2 uses a large amount of memory. For modern kernels you
164 will need at least 8MB RAM or more for booting.
168 depends on HAVE_KERNEL_LZMA
170 The most recent compression algorithm.
171 Its ratio is best, decompression speed is between the other
172 two. Compression is slowest. The kernel size is about 33%
173 smaller with LZMA in comparison to gzip.
178 bool "Support for paging of anonymous memory (swap)"
179 depends on MMU && BLOCK
182 This option allows you to choose whether you want to have support
183 for so called swap devices or swap files in your kernel that are
184 used to provide more virtual memory than the actual RAM present
185 in your computer. If unsure say Y.
190 Inter Process Communication is a suite of library functions and
191 system calls which let processes (running programs) synchronize and
192 exchange information. It is generally considered to be a good thing,
193 and some programs won't run unless you say Y here. In particular, if
194 you want to run the DOS emulator dosemu under Linux (read the
195 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
196 you'll need to say Y here.
198 You can find documentation about IPC with "info ipc" and also in
199 section 6.4 of the Linux Programmer's Guide, available from
200 <http://www.tldp.org/guides.html>.
202 config SYSVIPC_SYSCTL
209 bool "POSIX Message Queues"
210 depends on NET && EXPERIMENTAL
212 POSIX variant of message queues is a part of IPC. In POSIX message
213 queues every message has a priority which decides about succession
214 of receiving it by a process. If you want to compile and run
215 programs written e.g. for Solaris with use of its POSIX message
216 queues (functions mq_*) say Y here.
218 POSIX message queues are visible as a filesystem called 'mqueue'
219 and can be mounted somewhere if you want to do filesystem
220 operations on message queues.
224 config POSIX_MQUEUE_SYSCTL
226 depends on POSIX_MQUEUE
230 config BSD_PROCESS_ACCT
231 bool "BSD Process Accounting"
233 If you say Y here, a user level program will be able to instruct the
234 kernel (via a special system call) to write process accounting
235 information to a file: whenever a process exits, information about
236 that process will be appended to the file by the kernel. The
237 information includes things such as creation time, owning user,
238 command name, memory usage, controlling terminal etc. (the complete
239 list is in the struct acct in <file:include/linux/acct.h>). It is
240 up to the user level program to do useful things with this
241 information. This is generally a good idea, so say Y.
243 config BSD_PROCESS_ACCT_V3
244 bool "BSD Process Accounting version 3 file format"
245 depends on BSD_PROCESS_ACCT
248 If you say Y here, the process accounting information is written
249 in a new file format that also logs the process IDs of each
250 process and it's parent. Note that this file format is incompatible
251 with previous v0/v1/v2 file formats, so you will need updated tools
252 for processing it. A preliminary version of these tools is available
253 at <http://www.gnu.org/software/acct/>.
256 bool "Export task/process statistics through netlink (EXPERIMENTAL)"
260 Export selected statistics for tasks/processes through the
261 generic netlink interface. Unlike BSD process accounting, the
262 statistics are available during the lifetime of tasks/processes as
263 responses to commands. Like BSD accounting, they are sent to user
268 config TASK_DELAY_ACCT
269 bool "Enable per-task delay accounting (EXPERIMENTAL)"
272 Collect information on time spent by a task waiting for system
273 resources like cpu, synchronous block I/O completion and swapping
274 in pages. Such statistics can help in setting a task's priorities
275 relative to other tasks for cpu, io, rss limits etc.
280 bool "Enable extended accounting over taskstats (EXPERIMENTAL)"
283 Collect extended task accounting data and send the data
284 to userland for processing over the taskstats interface.
288 config TASK_IO_ACCOUNTING
289 bool "Enable per-task storage I/O accounting (EXPERIMENTAL)"
290 depends on TASK_XACCT
292 Collect information on the number of bytes of storage I/O which this
298 bool "Auditing support"
301 Enable auditing infrastructure that can be used with another
302 kernel subsystem, such as SELinux (which requires this for
303 logging of avc messages output). Does not do system-call
304 auditing without CONFIG_AUDITSYSCALL.
307 bool "Enable system-call auditing support"
308 depends on AUDIT && (X86 || PPC || PPC64 || S390 || IA64 || UML || SPARC64|| SUPERH)
309 default y if SECURITY_SELINUX
311 Enable low-overhead system-call auditing infrastructure that
312 can be used independently or with another kernel subsystem,
317 depends on AUDITSYSCALL
322 depends on AUDITSYSCALL
328 prompt "RCU Implementation"
332 bool "Tree-based hierarchical RCU"
334 This option selects the RCU implementation that is
335 designed for very large SMP system with hundreds or
336 thousands of CPUs. It also scales down nicely to
339 config TREE_PREEMPT_RCU
340 bool "Preemptable tree-based hierarchical RCU"
343 This option selects the RCU implementation that is
344 designed for very large SMP systems with hundreds or
345 thousands of CPUs, but for which real-time response
346 is also required. It also scales down nicely to
352 bool "Enable tracing for RCU"
353 depends on TREE_RCU || TREE_PREEMPT_RCU
355 This option provides tracing in RCU which presents stats
356 in debugfs for debugging RCU implementation.
358 Say Y here if you want to enable RCU tracing
359 Say N if you are unsure.
362 int "Tree-based hierarchical RCU fanout value"
365 depends on TREE_RCU || TREE_PREEMPT_RCU
369 This option controls the fanout of hierarchical implementations
370 of RCU, allowing RCU to work efficiently on machines with
371 large numbers of CPUs. This value must be at least the cube
372 root of NR_CPUS, which allows NR_CPUS up to 32,768 for 32-bit
373 systems and up to 262,144 for 64-bit systems.
375 Select a specific number if testing RCU itself.
376 Take the default if unsure.
378 config RCU_FANOUT_EXACT
379 bool "Disable tree-based hierarchical RCU auto-balancing"
380 depends on TREE_RCU || TREE_PREEMPT_RCU
383 This option forces use of the exact RCU_FANOUT value specified,
384 regardless of imbalances in the hierarchy. This is useful for
385 testing RCU itself, and might one day be useful on systems with
386 strong NUMA behavior.
388 Without RCU_FANOUT_EXACT, the code will balance the hierarchy.
392 config TREE_RCU_TRACE
393 def_bool RCU_TRACE && ( TREE_RCU || TREE_PREEMPT_RCU )
396 This option provides tracing for the TREE_RCU and
397 TREE_PREEMPT_RCU implementations, permitting Makefile to
398 trivially select kernel/rcutree_trace.c.
400 endmenu # "RCU Subsystem"
403 tristate "Kernel .config support"
405 This option enables the complete Linux kernel ".config" file
406 contents to be saved in the kernel. It provides documentation
407 of which kernel options are used in a running kernel or in an
408 on-disk kernel. This information can be extracted from the kernel
409 image file with the script scripts/extract-ikconfig and used as
410 input to rebuild the current kernel or to build another kernel.
411 It can also be extracted from a running kernel by reading
412 /proc/config.gz if enabled (below).
415 bool "Enable access to .config through /proc/config.gz"
416 depends on IKCONFIG && PROC_FS
418 This option enables access to the kernel configuration file
419 through /proc/config.gz.
422 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
426 Select kernel log buffer size as a power of 2.
436 # Architectures with an unreliable sched_clock() should select this:
438 config HAVE_UNSTABLE_SCHED_CLOCK
442 bool "Group CPU scheduler"
443 depends on EXPERIMENTAL
446 This feature lets CPU scheduler recognize task groups and control CPU
447 bandwidth allocation to such task groups.
448 In order to create a group from arbitrary set of processes, use
449 CONFIG_CGROUPS. (See Control Group support.)
451 config FAIR_GROUP_SCHED
452 bool "Group scheduling for SCHED_OTHER"
453 depends on GROUP_SCHED
456 config RT_GROUP_SCHED
457 bool "Group scheduling for SCHED_RR/FIFO"
458 depends on EXPERIMENTAL
459 depends on GROUP_SCHED
462 This feature lets you explicitly allocate real CPU bandwidth
463 to users or control groups (depending on the "Basis for grouping tasks"
464 setting below. If enabled, it will also make it impossible to
465 schedule realtime tasks for non-root users until you allocate
466 realtime bandwidth for them.
467 See Documentation/scheduler/sched-rt-group.txt for more information.
470 depends on GROUP_SCHED
471 prompt "Basis for grouping tasks"
477 This option will choose userid as the basis for grouping
478 tasks, thus providing equal CPU bandwidth to each user.
481 bool "Control groups"
484 This option allows you to create arbitrary task groups
485 using the "cgroup" pseudo filesystem and control
486 the cpu bandwidth allocated to each such task group.
487 Refer to Documentation/cgroups/cgroups.txt for more
488 information on "cgroup" pseudo filesystem.
493 boolean "Control Group support"
495 This option adds support for grouping sets of processes together, for
496 use with process control subsystems such as Cpusets, CFS, memory
497 controls or device isolation.
499 - Documentation/scheduler/sched-design-CFS.txt (CFS)
500 - Documentation/cgroups/ (features for grouping, isolation
501 and resource control)
508 bool "Example debug cgroup subsystem"
512 This option enables a simple cgroup subsystem that
513 exports useful debugging information about the cgroups
519 bool "Namespace cgroup subsystem"
522 Provides a simple namespace cgroup subsystem to
523 provide hierarchical naming of sets of namespaces,
524 for instance virtual servers and checkpoint/restart
527 config CGROUP_FREEZER
528 bool "Freezer cgroup subsystem"
531 Provides a way to freeze and unfreeze all tasks in a
535 bool "Device controller for cgroups"
536 depends on CGROUPS && EXPERIMENTAL
538 Provides a cgroup implementing whitelists for devices which
539 a process in the cgroup can mknod or open.
542 bool "Cpuset support"
545 This option will let you create and manage CPUSETs which
546 allow dynamically partitioning a system into sets of CPUs and
547 Memory Nodes and assigning tasks to run only within those sets.
548 This is primarily useful on large SMP or NUMA systems.
552 config PROC_PID_CPUSET
553 bool "Include legacy /proc/<pid>/cpuset file"
557 config CGROUP_CPUACCT
558 bool "Simple CPU accounting cgroup subsystem"
561 Provides a simple Resource Controller for monitoring the
562 total CPU consumed by the tasks in a cgroup.
564 config RESOURCE_COUNTERS
565 bool "Resource counters"
567 This option enables controller independent resource accounting
568 infrastructure that works with cgroups.
571 config CGROUP_MEM_RES_CTLR
572 bool "Memory Resource Controller for Control Groups"
573 depends on CGROUPS && RESOURCE_COUNTERS
576 Provides a memory resource controller that manages both anonymous
577 memory and page cache. (See Documentation/cgroups/memory.txt)
579 Note that setting this option increases fixed memory overhead
580 associated with each page of memory in the system. By this,
581 20(40)bytes/PAGE_SIZE on 32(64)bit system will be occupied by memory
582 usage tracking struct at boot. Total amount of this is printed out
585 Only enable when you're ok with these trade offs and really
586 sure you need the memory resource controller. Even when you enable
587 this, you can set "cgroup_disable=memory" at your boot option to
588 disable memory resource controller and you can avoid overheads.
589 (and lose benefits of memory resource controller)
591 This config option also selects MM_OWNER config option, which
592 could in turn add some fork/exit overhead.
594 config CGROUP_MEM_RES_CTLR_SWAP
595 bool "Memory Resource Controller Swap Extension(EXPERIMENTAL)"
596 depends on CGROUP_MEM_RES_CTLR && SWAP && EXPERIMENTAL
598 Add swap management feature to memory resource controller. When you
599 enable this, you can limit mem+swap usage per cgroup. In other words,
600 when you disable this, memory resource controller has no cares to
601 usage of swap...a process can exhaust all of the swap. This extension
602 is useful when you want to avoid exhaustion swap but this itself
603 adds more overheads and consumes memory for remembering information.
604 Especially if you use 32bit system or small memory system, please
605 be careful about enabling this. When memory resource controller
606 is disabled by boot option, this will be automatically disabled and
607 there will be no overhead from this. Even when you set this config=y,
608 if boot option "noswapaccount" is set, swap will not be accounted.
609 Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page
610 size is 4096bytes, 512k per 1Gbytes of swap.
617 config SYSFS_DEPRECATED
620 config SYSFS_DEPRECATED_V2
621 bool "remove sysfs features which may confuse old userspace tools"
624 select SYSFS_DEPRECATED
626 This option switches the layout of sysfs to the deprecated
627 version. Do not use it on recent distributions.
629 The current sysfs layout features a unified device tree at
630 /sys/devices/, which is able to express a hierarchy between
631 class devices. If the deprecated option is set to Y, the
632 unified device tree is split into a bus device tree at
633 /sys/devices/ and several individual class device trees at
634 /sys/class/. The class and bus devices will be connected by
635 "<subsystem>:<name>" and the "device" links. The "block"
636 class devices, will not show up in /sys/class/block/. Some
637 subsystems will suppress the creation of some devices which
638 depend on the unified device tree.
640 This option is not a pure compatibility option that can
641 be safely enabled on newer distributions. It will change the
642 layout of sysfs to the non-extensible deprecated version,
643 and disable some features, which can not be exported without
644 confusing older userspace tools. Since 2007/2008 all major
645 distributions do not enable this option, and ship no tools which
646 depend on the deprecated layout or this option.
648 If you are using a new kernel on an older distribution, or use
649 older userspace tools, you might need to say Y here. Do not say Y,
650 if the original kernel, that came with your distribution, has
651 this option set to N.
654 bool "Kernel->user space relay support (formerly relayfs)"
656 This option enables support for relay interface support in
657 certain file systems (such as debugfs).
658 It is designed to provide an efficient mechanism for tools and
659 facilities to relay large amounts of data from kernel space to
665 bool "Namespaces support" if EMBEDDED
668 Provides the way to make tasks work with different objects using
669 the same id. For example same IPC id may refer to different objects
670 or same user id or pid may refer to different tasks when used in
671 different namespaces.
675 depends on NAMESPACES
677 In this namespace tasks see different info provided with the
682 depends on NAMESPACES && (SYSVIPC || POSIX_MQUEUE)
684 In this namespace tasks work with IPC ids which correspond to
685 different IPC objects in different namespaces.
688 bool "User namespace (EXPERIMENTAL)"
689 depends on NAMESPACES && EXPERIMENTAL
691 This allows containers, i.e. vservers, to use user namespaces
692 to provide different user info for different servers.
696 bool "PID Namespaces (EXPERIMENTAL)"
698 depends on NAMESPACES && EXPERIMENTAL
700 Support process id namespaces. This allows having multiple
701 processes with the same pid as long as they are in different
702 pid namespaces. This is a building block of containers.
704 Unless you want to work with an experimental feature
708 bool "Network namespace"
710 depends on NAMESPACES && EXPERIMENTAL && NET
712 Allow user space to create what appear to be multiple instances
713 of the network stack.
715 config BLK_DEV_INITRD
716 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
717 depends on BROKEN || !FRV
719 The initial RAM filesystem is a ramfs which is loaded by the
720 boot loader (loadlin or lilo) and that is mounted as root
721 before the normal boot procedure. It is typically used to
722 load modules needed to mount the "real" root file system,
723 etc. See <file:Documentation/initrd.txt> for details.
725 If RAM disk support (BLK_DEV_RAM) is also included, this
726 also enables initial RAM disk (initrd) support and adds
727 15 Kbytes (more on some other architectures) to the kernel size.
737 config CC_OPTIMIZE_FOR_SIZE
738 bool "Optimize for size"
741 Enabling this option will pass "-Os" instead of "-O2" to gcc
742 resulting in a smaller kernel.
753 bool "Configure standard kernel features (for small systems)"
755 This option allows certain base kernel options and settings
756 to be disabled or tweaked. This is for specialized
757 environments which can tolerate a "non-standard" kernel.
758 Only use this if you really know what you are doing.
761 bool "Enable 16-bit UID system calls" if EMBEDDED
762 depends on ARM || BLACKFIN || CRIS || FRV || H8300 || X86_32 || M68K || (S390 && !64BIT) || SUPERH || SPARC32 || (SPARC64 && COMPAT) || UML || (X86_64 && IA32_EMULATION)
765 This enables the legacy 16-bit UID syscall wrappers.
767 config SYSCTL_SYSCALL
768 bool "Sysctl syscall support" if EMBEDDED
772 sys_sysctl uses binary paths that have been found challenging
773 to properly maintain and use. The interface in /proc/sys
774 using paths with ascii names is now the primary path to this
777 Almost nothing using the binary sysctl interface so if you are
778 trying to save some space it is probably safe to disable this,
779 making your kernel marginally smaller.
781 If unsure say Y here.
784 bool "Load all symbols for debugging/ksymoops" if EMBEDDED
787 Say Y here to let the kernel print out symbolic crash information and
788 symbolic stack backtraces. This increases the size of the kernel
789 somewhat, as all symbols have to be loaded into the kernel image.
792 bool "Include all symbols in kallsyms"
793 depends on DEBUG_KERNEL && KALLSYMS
795 Normally kallsyms only contains the symbols of functions, for nicer
796 OOPS messages. Some debuggers can use kallsyms for other
797 symbols too: say Y here to include all symbols, if you need them
798 and you don't care about adding 300k to the size of your kernel.
802 config KALLSYMS_EXTRA_PASS
803 bool "Do an extra kallsyms pass"
806 If kallsyms is not working correctly, the build will fail with
807 inconsistent kallsyms data. If that occurs, log a bug report and
808 turn on KALLSYMS_EXTRA_PASS which should result in a stable build.
809 Always say N here unless you find a bug in kallsyms, which must be
810 reported. KALLSYMS_EXTRA_PASS is only a temporary workaround while
811 you wait for kallsyms to be fixed.
815 bool "Support for hot-pluggable devices" if EMBEDDED
818 This option is provided for the case where no hotplug or uevent
819 capabilities is wanted by the kernel. You should only consider
820 disabling this option for embedded systems that do not use modules, a
821 dynamic /dev tree, or dynamic device discovery. Just say Y.
825 bool "Enable support for printk" if EMBEDDED
827 This option enables normal printk support. Removing it
828 eliminates most of the message strings from the kernel image
829 and makes the kernel more or less silent. As this makes it
830 very difficult to diagnose system problems, saying N here is
831 strongly discouraged.
834 bool "BUG() support" if EMBEDDED
837 Disabling this option eliminates support for BUG and WARN, reducing
838 the size of your kernel image and potentially quietly ignoring
839 numerous fatal conditions. You should only consider disabling this
840 option for embedded systems with no facilities for reporting errors.
845 bool "Enable ELF core dumps" if EMBEDDED
847 Enable support for generating core dumps. Disabling saves about 4k.
849 config PCSPKR_PLATFORM
850 bool "Enable PC-Speaker support" if EMBEDDED
851 depends on ALPHA || X86 || MIPS || PPC_PREP || PPC_CHRP || PPC_PSERIES
854 This option allows to disable the internal PC-Speaker
855 support, saving some memory.
859 bool "Enable full-sized data structures for core" if EMBEDDED
861 Disabling this option reduces the size of miscellaneous core
862 kernel data structures. This saves memory on small machines,
863 but may reduce performance.
866 bool "Enable futex support" if EMBEDDED
870 Disabling this option will cause the kernel to be built without
871 support for "fast userspace mutexes". The resulting kernel may not
872 run glibc-based applications correctly.
875 bool "Enable eventpoll support" if EMBEDDED
879 Disabling this option will cause the kernel to be built without
880 support for epoll family of system calls.
883 bool "Enable signalfd() system call" if EMBEDDED
887 Enable the signalfd() system call that allows to receive signals
888 on a file descriptor.
893 bool "Enable timerfd() system call" if EMBEDDED
897 Enable the timerfd() system call that allows to receive timer
898 events on a file descriptor.
903 bool "Enable eventfd() system call" if EMBEDDED
907 Enable the eventfd() system call that allows to receive both
908 kernel notification (ie. KAIO) or userspace notifications.
913 bool "Use full shmem filesystem" if EMBEDDED
917 The shmem is an internal filesystem used to manage shared memory.
918 It is backed by swap and manages resource limits. It is also exported
919 to userspace as tmpfs if TMPFS is enabled. Disabling this
920 option replaces shmem and tmpfs with the much simpler ramfs code,
921 which may be appropriate on small systems without swap.
924 bool "Enable AIO support" if EMBEDDED
927 This option enables POSIX asynchronous I/O which may by used
928 by some high performance threaded applications. Disabling
929 this option saves about 7k.
931 config HAVE_PERF_EVENTS
934 See tools/perf/design.txt for details.
936 config PERF_USE_VMALLOC
939 See tools/perf/design.txt for details
941 menu "Kernel Performance Events And Counters"
944 bool "Kernel performance events and counters"
945 default y if (PROFILING || PERF_COUNTERS)
946 depends on HAVE_PERF_EVENTS
949 Enable kernel support for various performance events provided
950 by software and hardware.
952 Software events are supported either build-in or via the
953 use of generic tracepoints.
955 Most modern CPUs support performance events via performance
956 counter registers. These registers count the number of certain
957 types of hw events: such as instructions executed, cachemisses
958 suffered, or branches mis-predicted - without slowing down the
959 kernel or applications. These registers can also trigger interrupts
960 when a threshold number of events have passed - and can thus be
961 used to profile the code that runs on that CPU.
963 The Linux Performance Event subsystem provides an abstraction of
964 these software and hardware cevent apabilities, available via a
965 system call and used by the "perf" utility in tools/perf/. It
966 provides per task and per CPU counters, and it provides event
967 capabilities on top of those.
972 bool "Tracepoint profiling sources"
973 depends on PERF_EVENTS && EVENT_TRACING
976 Allow the use of tracepoints as software performance events.
978 When this is enabled, you can create perf events based on
979 tracepoints using PERF_TYPE_TRACEPOINT and the tracepoint ID
980 found in debugfs://tracing/events/*/*/id. (The -e/--events
981 option to the perf tool can parse and interpret symbolic
982 tracepoints, in the subsystem:tracepoint_name format.)
985 bool "Kernel performance counters (old config option)"
986 depends on HAVE_PERF_EVENTS
988 This config has been obsoleted by the PERF_EVENTS
989 config option - please see that one for details.
991 It has no effect on the kernel whether you enable
992 it or not, it is a compatibility placeholder.
996 config DEBUG_PERF_USE_VMALLOC
998 bool "Debug: use vmalloc to back perf mmap() buffers"
999 depends on PERF_EVENTS && DEBUG_KERNEL
1000 select PERF_USE_VMALLOC
1002 Use vmalloc memory to back perf mmap() buffers.
1004 Mostly useful for debugging the vmalloc code on platforms
1005 that don't require it.
1011 config VM_EVENT_COUNTERS
1013 bool "Enable VM event counters for /proc/vmstat" if EMBEDDED
1015 VM event counters are needed for event counts to be shown.
1016 This option allows the disabling of the VM event counters
1017 on EMBEDDED systems. /proc/vmstat will only show page counts
1018 if VM event counters are disabled.
1022 bool "Enable PCI quirk workarounds" if EMBEDDED
1025 This enables workarounds for various PCI chipset
1026 bugs/quirks. Disable this only if your target machine is
1027 unaffected by PCI quirks.
1031 bool "Enable SLUB debugging support" if EMBEDDED
1032 depends on SLUB && SYSFS
1034 SLUB has extensive debug support features. Disabling these can
1035 result in significant savings in code size. This also disables
1036 SLUB sysfs support. /sys/slab will not exist and there will be
1037 no support for cache validation etc.
1040 bool "Disable heap randomization"
1043 Randomizing heap placement makes heap exploits harder, but it
1044 also breaks ancient binaries (including anything libc5 based).
1045 This option changes the bootup default to heap randomization
1046 disabled, and can be overridden at runtime by setting
1047 /proc/sys/kernel/randomize_va_space to 2.
1049 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1052 prompt "Choose SLAB allocator"
1055 This option allows to select a slab allocator.
1060 The regular slab allocator that is established and known to work
1061 well in all environments. It organizes cache hot objects in
1062 per cpu and per node queues.
1065 bool "SLUB (Unqueued Allocator)"
1067 SLUB is a slab allocator that minimizes cache line usage
1068 instead of managing queues of cached objects (SLAB approach).
1069 Per cpu caching is realized using slabs of objects instead
1070 of queues of objects. SLUB can use memory efficiently
1071 and has enhanced diagnostics. SLUB is the default choice for
1075 bool "SLQB (Queued allocator)"
1077 SLQB is a proposed new slab allocator.
1081 bool "SLOB (Simple Allocator)"
1083 SLOB replaces the stock allocator with a drastically simpler
1084 allocator. SLOB is generally more space efficient but
1085 does not perform as well on large systems.
1090 bool "Profiling support (EXPERIMENTAL)"
1092 Say Y here to enable the extended profiling support mechanisms used
1093 by profilers such as OProfile.
1096 # Place an empty function call at each tracepoint site. Can be
1097 # dynamically changed for a probe function.
1102 source "arch/Kconfig"
1108 The slow work thread pool provides a number of dynamically allocated
1109 threads that can be used by the kernel to perform operations that
1110 take a relatively long time.
1112 An example of this would be CacheFiles doing a path lookup followed
1113 by a series of mkdirs and a create call, all of which have to touch
1116 See Documentation/slow-work.txt.
1118 endmenu # General setup
1120 config HAVE_GENERIC_DMA_COHERENT
1127 depends on SLAB || SLUB_DEBUG || SLQB
1135 default 0 if BASE_FULL
1136 default 1 if !BASE_FULL
1139 bool "Enable loadable module support"
1141 Kernel modules are small pieces of compiled code which can
1142 be inserted in the running kernel, rather than being
1143 permanently built into the kernel. You use the "modprobe"
1144 tool to add (and sometimes remove) them. If you say Y here,
1145 many parts of the kernel can be built as modules (by
1146 answering M instead of Y where indicated): this is most
1147 useful for infrequently used options which are not required
1148 for booting. For more information, see the man pages for
1149 modprobe, lsmod, modinfo, insmod and rmmod.
1151 If you say Y here, you will need to run "make
1152 modules_install" to put the modules under /lib/modules/
1153 where modprobe can find them (you may need to be root to do
1160 config MODULE_FORCE_LOAD
1161 bool "Forced module loading"
1164 Allow loading of modules without version information (ie. modprobe
1165 --force). Forced module loading sets the 'F' (forced) taint flag and
1166 is usually a really bad idea.
1168 config MODULE_UNLOAD
1169 bool "Module unloading"
1171 Without this option you will not be able to unload any
1172 modules (note that some modules may not be unloadable
1173 anyway), which makes your kernel smaller, faster
1174 and simpler. If unsure, say Y.
1176 config MODULE_FORCE_UNLOAD
1177 bool "Forced module unloading"
1178 depends on MODULE_UNLOAD && EXPERIMENTAL
1180 This option allows you to force a module to unload, even if the
1181 kernel believes it is unsafe: the kernel will remove the module
1182 without waiting for anyone to stop using it (using the -f option to
1183 rmmod). This is mainly for kernel developers and desperate users.
1187 bool "Module versioning support"
1189 Usually, you have to use modules compiled with your kernel.
1190 Saying Y here makes it sometimes possible to use modules
1191 compiled for different kernels, by adding enough information
1192 to the modules to (hopefully) spot any changes which would
1193 make them incompatible with the kernel you are running. If
1196 config MODULE_SRCVERSION_ALL
1197 bool "Source checksum for all modules"
1199 Modules which contain a MODULE_VERSION get an extra "srcversion"
1200 field inserted into their modinfo section, which contains a
1201 sum of the source files which made it. This helps maintainers
1202 see exactly which source was used to build a module (since
1203 others sometimes change the module source without updating
1204 the version). With this option, such a "srcversion" field
1205 will be created for all modules. If unsure, say N.
1209 config INIT_ALL_POSSIBLE
1212 Back when each arch used to define their own cpu_online_map and
1213 cpu_possible_map, some of them chose to initialize cpu_possible_map
1214 with all 1s, and others with all 0s. When they were centralised,
1215 it was better to provide this option than to break all the archs
1216 and have several arch maintainers pursuing me down dark alleys.
1221 depends on (SMP && MODULE_UNLOAD) || HOTPLUG_CPU
1223 Need stop_machine() primitive.
1226 bool "Infrastructure for tracing and debugging user processes"
1227 depends on EXPERIMENTAL
1228 depends on HAVE_ARCH_TRACEHOOK
1230 Enable the utrace process tracing interface. This is an internal
1231 kernel interface exported to kernel modules, to track events in
1232 user threads, extract and change user thread state.
1234 source "block/Kconfig"
1236 config PREEMPT_NOTIFIERS