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"
28 depends on HAVE_IRQ_WORK
30 config BUILDTIME_EXTABLE_SORT
36 bool "Prompt for development and/or incomplete code/drivers"
38 Some of the various things that Linux supports (such as network
39 drivers, file systems, network protocols, etc.) can be in a state
40 of development where the functionality, stability, or the level of
41 testing is not yet high enough for general use. This is usually
42 known as the "alpha-test" phase among developers. If a feature is
43 currently in alpha-test, then the developers usually discourage
44 uninformed widespread use of this feature by the general public to
45 avoid "Why doesn't this work?" type mail messages. However, active
46 testing and use of these systems is welcomed. Just be aware that it
47 may not meet the normal level of reliability or it may fail to work
48 in some special cases. Detailed bug reports from people familiar
49 with the kernel internals are usually welcomed by the developers
50 (before submitting bug reports, please read the documents
51 <file:README>, <file:MAINTAINERS>, <file:REPORTING-BUGS>,
52 <file:Documentation/BUG-HUNTING>, and
53 <file:Documentation/oops-tracing.txt> in the kernel source).
55 This option will also make obsoleted drivers available. These are
56 drivers that have been replaced by something else, and/or are
57 scheduled to be removed in a future kernel release.
59 Unless you intend to help test and develop a feature or driver that
60 falls into this category, or you have a situation that requires
61 using these features, you should probably say N here, which will
62 cause the configurator to present you with fewer choices. If
63 you say Y here, you will be offered the choice of using features or
64 drivers that are currently considered to be in the alpha-test phase.
71 depends on BROKEN || !SMP
74 config INIT_ENV_ARG_LIMIT
79 Maximum of each of the number of arguments and environment
80 variables passed to init from the kernel command line.
84 string "Cross-compiler tool prefix"
86 Same as running 'make CROSS_COMPILE=prefix-' but stored for
87 default make runs in this kernel build directory. You don't
88 need to set this unless you want the configured kernel build
89 directory to select the cross-compiler automatically.
92 string "Local version - append to kernel release"
94 Append an extra string to the end of your kernel version.
95 This will show up when you type uname, for example.
96 The string you set here will be appended after the contents of
97 any files with a filename matching localversion* in your
98 object and source tree, in that order. Your total string can
99 be a maximum of 64 characters.
101 config LOCALVERSION_AUTO
102 bool "Automatically append version information to the version string"
105 This will try to automatically determine if the current tree is a
106 release tree by looking for git tags that belong to the current
107 top of tree revision.
109 A string of the format -gxxxxxxxx will be added to the localversion
110 if a git-based tree is found. The string generated by this will be
111 appended after any matching localversion* files, and after the value
112 set in CONFIG_LOCALVERSION.
114 (The actual string used here is the first eight characters produced
115 by running the command:
117 $ git rev-parse --verify HEAD
119 which is done within the script "scripts/setlocalversion".)
121 config HAVE_KERNEL_GZIP
124 config HAVE_KERNEL_BZIP2
127 config HAVE_KERNEL_LZMA
130 config HAVE_KERNEL_XZ
133 config HAVE_KERNEL_LZO
137 prompt "Kernel compression mode"
139 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO
141 The linux kernel is a kind of self-extracting executable.
142 Several compression algorithms are available, which differ
143 in efficiency, compression and decompression speed.
144 Compression speed is only relevant when building a kernel.
145 Decompression speed is relevant at each boot.
147 If you have any problems with bzip2 or lzma compressed
148 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
149 version of this functionality (bzip2 only), for 2.4, was
150 supplied by Christian Ludwig)
152 High compression options are mostly useful for users, who
153 are low on disk space (embedded systems), but for whom ram
156 If in doubt, select 'gzip'
160 depends on HAVE_KERNEL_GZIP
162 The old and tried gzip compression. It provides a good balance
163 between compression ratio and decompression speed.
167 depends on HAVE_KERNEL_BZIP2
169 Its compression ratio and speed is intermediate.
170 Decompression speed is slowest among the three. The kernel
171 size is about 10% smaller with bzip2, in comparison to gzip.
172 Bzip2 uses a large amount of memory. For modern kernels you
173 will need at least 8MB RAM or more for booting.
177 depends on HAVE_KERNEL_LZMA
179 The most recent compression algorithm.
180 Its ratio is best, decompression speed is between the other
181 two. Compression is slowest. The kernel size is about 33%
182 smaller with LZMA in comparison to gzip.
186 depends on HAVE_KERNEL_XZ
188 XZ uses the LZMA2 algorithm and instruction set specific
189 BCJ filters which can improve compression ratio of executable
190 code. The size of the kernel is about 30% smaller with XZ in
191 comparison to gzip. On architectures for which there is a BCJ
192 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ
193 will create a few percent smaller kernel than plain LZMA.
195 The speed is about the same as with LZMA: The decompression
196 speed of XZ is better than that of bzip2 but worse than gzip
197 and LZO. Compression is slow.
201 depends on HAVE_KERNEL_LZO
203 Its compression ratio is the poorest among the 4. The kernel
204 size is about 10% bigger than gzip; however its speed
205 (both compression and decompression) is the fastest.
209 config DEFAULT_HOSTNAME
210 string "Default hostname"
213 This option determines the default system hostname before userspace
214 calls sethostname(2). The kernel traditionally uses "(none)" here,
215 but you may wish to use a different default here to make a minimal
216 system more usable with less configuration.
219 bool "Support for paging of anonymous memory (swap)"
220 depends on MMU && BLOCK
223 This option allows you to choose whether you want to have support
224 for so called swap devices or swap files in your kernel that are
225 used to provide more virtual memory than the actual RAM present
226 in your computer. If unsure say Y.
231 Inter Process Communication is a suite of library functions and
232 system calls which let processes (running programs) synchronize and
233 exchange information. It is generally considered to be a good thing,
234 and some programs won't run unless you say Y here. In particular, if
235 you want to run the DOS emulator dosemu under Linux (read the
236 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
237 you'll need to say Y here.
239 You can find documentation about IPC with "info ipc" and also in
240 section 6.4 of the Linux Programmer's Guide, available from
241 <http://www.tldp.org/guides.html>.
243 config SYSVIPC_SYSCTL
250 bool "POSIX Message Queues"
251 depends on NET && EXPERIMENTAL
253 POSIX variant of message queues is a part of IPC. In POSIX message
254 queues every message has a priority which decides about succession
255 of receiving it by a process. If you want to compile and run
256 programs written e.g. for Solaris with use of its POSIX message
257 queues (functions mq_*) say Y here.
259 POSIX message queues are visible as a filesystem called 'mqueue'
260 and can be mounted somewhere if you want to do filesystem
261 operations on message queues.
265 config POSIX_MQUEUE_SYSCTL
267 depends on POSIX_MQUEUE
271 config BSD_PROCESS_ACCT
272 bool "BSD Process Accounting"
274 If you say Y here, a user level program will be able to instruct the
275 kernel (via a special system call) to write process accounting
276 information to a file: whenever a process exits, information about
277 that process will be appended to the file by the kernel. The
278 information includes things such as creation time, owning user,
279 command name, memory usage, controlling terminal etc. (the complete
280 list is in the struct acct in <file:include/linux/acct.h>). It is
281 up to the user level program to do useful things with this
282 information. This is generally a good idea, so say Y.
284 config BSD_PROCESS_ACCT_V3
285 bool "BSD Process Accounting version 3 file format"
286 depends on BSD_PROCESS_ACCT
289 If you say Y here, the process accounting information is written
290 in a new file format that also logs the process IDs of each
291 process and it's parent. Note that this file format is incompatible
292 with previous v0/v1/v2 file formats, so you will need updated tools
293 for processing it. A preliminary version of these tools is available
294 at <http://www.gnu.org/software/acct/>.
297 bool "open by fhandle syscalls"
300 If you say Y here, a user level program will be able to map
301 file names to handle and then later use the handle for
302 different file system operations. This is useful in implementing
303 userspace file servers, which now track files using handles instead
304 of names. The handle would remain the same even if file names
305 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2)
309 bool "Export task/process statistics through netlink (EXPERIMENTAL)"
313 Export selected statistics for tasks/processes through the
314 generic netlink interface. Unlike BSD process accounting, the
315 statistics are available during the lifetime of tasks/processes as
316 responses to commands. Like BSD accounting, they are sent to user
321 config TASK_DELAY_ACCT
322 bool "Enable per-task delay accounting (EXPERIMENTAL)"
325 Collect information on time spent by a task waiting for system
326 resources like cpu, synchronous block I/O completion and swapping
327 in pages. Such statistics can help in setting a task's priorities
328 relative to other tasks for cpu, io, rss limits etc.
333 bool "Enable extended accounting over taskstats (EXPERIMENTAL)"
336 Collect extended task accounting data and send the data
337 to userland for processing over the taskstats interface.
341 config TASK_IO_ACCOUNTING
342 bool "Enable per-task storage I/O accounting (EXPERIMENTAL)"
343 depends on TASK_XACCT
345 Collect information on the number of bytes of storage I/O which this
351 bool "Auditing support"
354 Enable auditing infrastructure that can be used with another
355 kernel subsystem, such as SELinux (which requires this for
356 logging of avc messages output). Does not do system-call
357 auditing without CONFIG_AUDITSYSCALL.
360 bool "Enable system-call auditing support"
361 depends on AUDIT && (X86 || PPC || S390 || IA64 || UML || SPARC64 || SUPERH || ARM)
362 default y if SECURITY_SELINUX
364 Enable low-overhead system-call auditing infrastructure that
365 can be used independently or with another kernel subsystem,
370 depends on AUDITSYSCALL
375 depends on AUDITSYSCALL
378 config AUDIT_LOGINUID_IMMUTABLE
379 bool "Make audit loginuid immutable"
382 The config option toggles if a task setting its loginuid requires
383 CAP_SYS_AUDITCONTROL or if that task should require no special permissions
384 but should instead only allow setting its loginuid if it was never
385 previously set. On systems which use systemd or a similar central
386 process to restart login services this should be set to true. On older
387 systems in which an admin would typically have to directly stop and
388 start processes this should be set to false. Setting this to true allows
389 one to drop potentially dangerous capabilites from the login tasks,
390 but may not be backwards compatible with older init systems.
392 source "kernel/irq/Kconfig"
397 prompt "RCU Implementation"
401 bool "Tree-based hierarchical RCU"
402 depends on !PREEMPT && SMP
404 This option selects the RCU implementation that is
405 designed for very large SMP system with hundreds or
406 thousands of CPUs. It also scales down nicely to
409 config TREE_PREEMPT_RCU
410 bool "Preemptible tree-based hierarchical RCU"
411 depends on PREEMPT && SMP
413 This option selects the RCU implementation that is
414 designed for very large SMP systems with hundreds or
415 thousands of CPUs, but for which real-time response
416 is also required. It also scales down nicely to
420 bool "UP-only small-memory-footprint RCU"
421 depends on !PREEMPT && !SMP
423 This option selects the RCU implementation that is
424 designed for UP systems from which real-time response
425 is not required. This option greatly reduces the
426 memory footprint of RCU.
428 config TINY_PREEMPT_RCU
429 bool "Preemptible UP-only small-memory-footprint RCU"
430 depends on PREEMPT && !SMP
432 This option selects the RCU implementation that is designed
433 for real-time UP systems. This option greatly reduces the
434 memory footprint of RCU.
439 def_bool ( TREE_PREEMPT_RCU || TINY_PREEMPT_RCU )
441 This option enables preemptible-RCU code that is common between
442 the TREE_PREEMPT_RCU and TINY_PREEMPT_RCU implementations.
445 int "Tree-based hierarchical RCU fanout value"
448 depends on TREE_RCU || TREE_PREEMPT_RCU
452 This option controls the fanout of hierarchical implementations
453 of RCU, allowing RCU to work efficiently on machines with
454 large numbers of CPUs. This value must be at least the fourth
455 root of NR_CPUS, which allows NR_CPUS to be insanely large.
456 The default value of RCU_FANOUT should be used for production
457 systems, but if you are stress-testing the RCU implementation
458 itself, small RCU_FANOUT values allow you to test large-system
459 code paths on small(er) systems.
461 Select a specific number if testing RCU itself.
462 Take the default if unsure.
464 config RCU_FANOUT_LEAF
465 int "Tree-based hierarchical RCU leaf-level fanout value"
466 range 2 RCU_FANOUT if 64BIT
467 range 2 RCU_FANOUT if !64BIT
468 depends on TREE_RCU || TREE_PREEMPT_RCU
471 This option controls the leaf-level fanout of hierarchical
472 implementations of RCU, and allows trading off cache misses
473 against lock contention. Systems that synchronize their
474 scheduling-clock interrupts for energy-efficiency reasons will
475 want the default because the smaller leaf-level fanout keeps
476 lock contention levels acceptably low. Very large systems
477 (hundreds or thousands of CPUs) will instead want to set this
478 value to the maximum value possible in order to reduce the
479 number of cache misses incurred during RCU's grace-period
480 initialization. These systems tend to run CPU-bound, and thus
481 are not helped by synchronized interrupts, and thus tend to
482 skew them, which reduces lock contention enough that large
483 leaf-level fanouts work well.
485 Select a specific number if testing RCU itself.
487 Select the maximum permissible value for large systems.
489 Take the default if unsure.
491 config RCU_FANOUT_EXACT
492 bool "Disable tree-based hierarchical RCU auto-balancing"
493 depends on TREE_RCU || TREE_PREEMPT_RCU
496 This option forces use of the exact RCU_FANOUT value specified,
497 regardless of imbalances in the hierarchy. This is useful for
498 testing RCU itself, and might one day be useful on systems with
499 strong NUMA behavior.
501 Without RCU_FANOUT_EXACT, the code will balance the hierarchy.
505 config RCU_FAST_NO_HZ
506 bool "Accelerate last non-dyntick-idle CPU's grace periods"
507 depends on NO_HZ && SMP
510 This option causes RCU to attempt to accelerate grace periods
511 in order to allow CPUs to enter dynticks-idle state more
512 quickly. On the other hand, this option increases the overhead
513 of the dynticks-idle checking, particularly on systems with
514 large numbers of CPUs.
516 Say Y if energy efficiency is critically important, particularly
517 if you have relatively few CPUs.
519 Say N if you are unsure.
521 config TREE_RCU_TRACE
522 def_bool RCU_TRACE && ( TREE_RCU || TREE_PREEMPT_RCU )
525 This option provides tracing for the TREE_RCU and
526 TREE_PREEMPT_RCU implementations, permitting Makefile to
527 trivially select kernel/rcutree_trace.c.
530 bool "Enable RCU priority boosting"
531 depends on RT_MUTEXES && PREEMPT_RCU
534 This option boosts the priority of preempted RCU readers that
535 block the current preemptible RCU grace period for too long.
536 This option also prevents heavy loads from blocking RCU
537 callback invocation for all flavors of RCU.
539 Say Y here if you are working with real-time apps or heavy loads
540 Say N here if you are unsure.
542 config RCU_BOOST_PRIO
543 int "Real-time priority to boost RCU readers to"
548 This option specifies the real-time priority to which long-term
549 preempted RCU readers are to be boosted. If you are working
550 with a real-time application that has one or more CPU-bound
551 threads running at a real-time priority level, you should set
552 RCU_BOOST_PRIO to a priority higher then the highest-priority
553 real-time CPU-bound thread. The default RCU_BOOST_PRIO value
554 of 1 is appropriate in the common case, which is real-time
555 applications that do not have any CPU-bound threads.
557 Some real-time applications might not have a single real-time
558 thread that saturates a given CPU, but instead might have
559 multiple real-time threads that, taken together, fully utilize
560 that CPU. In this case, you should set RCU_BOOST_PRIO to
561 a priority higher than the lowest-priority thread that is
562 conspiring to prevent the CPU from running any non-real-time
563 tasks. For example, if one thread at priority 10 and another
564 thread at priority 5 are between themselves fully consuming
565 the CPU time on a given CPU, then RCU_BOOST_PRIO should be
566 set to priority 6 or higher.
568 Specify the real-time priority, or take the default if unsure.
570 config RCU_BOOST_DELAY
571 int "Milliseconds to delay boosting after RCU grace-period start"
576 This option specifies the time to wait after the beginning of
577 a given grace period before priority-boosting preempted RCU
578 readers blocking that grace period. Note that any RCU reader
579 blocking an expedited RCU grace period is boosted immediately.
581 Accept the default if unsure.
583 endmenu # "RCU Subsystem"
586 tristate "Kernel .config support"
588 This option enables the complete Linux kernel ".config" file
589 contents to be saved in the kernel. It provides documentation
590 of which kernel options are used in a running kernel or in an
591 on-disk kernel. This information can be extracted from the kernel
592 image file with the script scripts/extract-ikconfig and used as
593 input to rebuild the current kernel or to build another kernel.
594 It can also be extracted from a running kernel by reading
595 /proc/config.gz if enabled (below).
598 bool "Enable access to .config through /proc/config.gz"
599 depends on IKCONFIG && PROC_FS
601 This option enables access to the kernel configuration file
602 through /proc/config.gz.
605 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
609 Select kernel log buffer size as a power of 2.
619 # Architectures with an unreliable sched_clock() should select this:
621 config HAVE_UNSTABLE_SCHED_CLOCK
625 boolean "Control Group support"
628 This option adds support for grouping sets of processes together, for
629 use with process control subsystems such as Cpusets, CFS, memory
630 controls or device isolation.
632 - Documentation/scheduler/sched-design-CFS.txt (CFS)
633 - Documentation/cgroups/ (features for grouping, isolation
634 and resource control)
641 bool "Example debug cgroup subsystem"
644 This option enables a simple cgroup subsystem that
645 exports useful debugging information about the cgroups
650 config CGROUP_FREEZER
651 bool "Freezer cgroup subsystem"
653 Provides a way to freeze and unfreeze all tasks in a
657 bool "Device controller for cgroups"
659 Provides a cgroup implementing whitelists for devices which
660 a process in the cgroup can mknod or open.
663 bool "Cpuset support"
665 This option will let you create and manage CPUSETs which
666 allow dynamically partitioning a system into sets of CPUs and
667 Memory Nodes and assigning tasks to run only within those sets.
668 This is primarily useful on large SMP or NUMA systems.
672 config PROC_PID_CPUSET
673 bool "Include legacy /proc/<pid>/cpuset file"
677 config CGROUP_CPUACCT
678 bool "Simple CPU accounting cgroup subsystem"
680 Provides a simple Resource Controller for monitoring the
681 total CPU consumed by the tasks in a cgroup.
683 config RESOURCE_COUNTERS
684 bool "Resource counters"
686 This option enables controller independent resource accounting
687 infrastructure that works with cgroups.
689 config CGROUP_MEM_RES_CTLR
690 bool "Memory Resource Controller for Control Groups"
691 depends on RESOURCE_COUNTERS
694 Provides a memory resource controller that manages both anonymous
695 memory and page cache. (See Documentation/cgroups/memory.txt)
697 Note that setting this option increases fixed memory overhead
698 associated with each page of memory in the system. By this,
699 20(40)bytes/PAGE_SIZE on 32(64)bit system will be occupied by memory
700 usage tracking struct at boot. Total amount of this is printed out
703 Only enable when you're ok with these trade offs and really
704 sure you need the memory resource controller. Even when you enable
705 this, you can set "cgroup_disable=memory" at your boot option to
706 disable memory resource controller and you can avoid overheads.
707 (and lose benefits of memory resource controller)
709 This config option also selects MM_OWNER config option, which
710 could in turn add some fork/exit overhead.
712 config CGROUP_MEM_RES_CTLR_SWAP
713 bool "Memory Resource Controller Swap Extension"
714 depends on CGROUP_MEM_RES_CTLR && SWAP
716 Add swap management feature to memory resource controller. When you
717 enable this, you can limit mem+swap usage per cgroup. In other words,
718 when you disable this, memory resource controller has no cares to
719 usage of swap...a process can exhaust all of the swap. This extension
720 is useful when you want to avoid exhaustion swap but this itself
721 adds more overheads and consumes memory for remembering information.
722 Especially if you use 32bit system or small memory system, please
723 be careful about enabling this. When memory resource controller
724 is disabled by boot option, this will be automatically disabled and
725 there will be no overhead from this. Even when you set this config=y,
726 if boot option "swapaccount=0" is set, swap will not be accounted.
727 Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page
728 size is 4096bytes, 512k per 1Gbytes of swap.
729 config CGROUP_MEM_RES_CTLR_SWAP_ENABLED
730 bool "Memory Resource Controller Swap Extension enabled by default"
731 depends on CGROUP_MEM_RES_CTLR_SWAP
734 Memory Resource Controller Swap Extension comes with its price in
735 a bigger memory consumption. General purpose distribution kernels
736 which want to enable the feature but keep it disabled by default
737 and let the user enable it by swapaccount boot command line
738 parameter should have this option unselected.
739 For those who want to have the feature enabled by default should
740 select this option (if, for some reason, they need to disable it
741 then swapaccount=0 does the trick).
742 config CGROUP_MEM_RES_CTLR_KMEM
743 bool "Memory Resource Controller Kernel Memory accounting (EXPERIMENTAL)"
744 depends on CGROUP_MEM_RES_CTLR && EXPERIMENTAL
747 The Kernel Memory extension for Memory Resource Controller can limit
748 the amount of memory used by kernel objects in the system. Those are
749 fundamentally different from the entities handled by the standard
750 Memory Controller, which are page-based, and can be swapped. Users of
751 the kmem extension can use it to guarantee that no group of processes
752 will ever exhaust kernel resources alone.
755 bool "Enable perf_event per-cpu per-container group (cgroup) monitoring"
756 depends on PERF_EVENTS && CGROUPS
758 This option extends the per-cpu mode to restrict monitoring to
759 threads which belong to the cgroup specified and run on the
764 menuconfig CGROUP_SCHED
765 bool "Group CPU scheduler"
768 This feature lets CPU scheduler recognize task groups and control CPU
769 bandwidth allocation to such task groups. It uses cgroups to group
773 config FAIR_GROUP_SCHED
774 bool "Group scheduling for SCHED_OTHER"
775 depends on CGROUP_SCHED
779 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED"
780 depends on EXPERIMENTAL
781 depends on FAIR_GROUP_SCHED
784 This option allows users to define CPU bandwidth rates (limits) for
785 tasks running within the fair group scheduler. Groups with no limit
786 set are considered to be unconstrained and will run with no
788 See tip/Documentation/scheduler/sched-bwc.txt for more information.
790 config RT_GROUP_SCHED
791 bool "Group scheduling for SCHED_RR/FIFO"
792 depends on EXPERIMENTAL
793 depends on CGROUP_SCHED
796 This feature lets you explicitly allocate real CPU bandwidth
797 to task groups. If enabled, it will also make it impossible to
798 schedule realtime tasks for non-root users until you allocate
799 realtime bandwidth for them.
800 See Documentation/scheduler/sched-rt-group.txt for more information.
805 tristate "Block IO controller"
809 Generic block IO controller cgroup interface. This is the common
810 cgroup interface which should be used by various IO controlling
813 Currently, CFQ IO scheduler uses it to recognize task groups and
814 control disk bandwidth allocation (proportional time slice allocation)
815 to such task groups. It is also used by bio throttling logic in
816 block layer to implement upper limit in IO rates on a device.
818 This option only enables generic Block IO controller infrastructure.
819 One needs to also enable actual IO controlling logic/policy. For
820 enabling proportional weight division of disk bandwidth in CFQ, set
821 CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set
822 CONFIG_BLK_DEV_THROTTLING=y.
824 See Documentation/cgroups/blkio-controller.txt for more information.
826 config DEBUG_BLK_CGROUP
827 bool "Enable Block IO controller debugging"
828 depends on BLK_CGROUP
831 Enable some debugging help. Currently it exports additional stat
832 files in a cgroup which can be useful for debugging.
836 config CHECKPOINT_RESTORE
837 bool "Checkpoint/restore support" if EXPERT
840 Enables additional kernel features in a sake of checkpoint/restore.
841 In particular it adds auxiliary prctl codes to setup process text,
842 data and heap segment sizes, and a few additional /proc filesystem
845 If unsure, say N here.
847 menuconfig NAMESPACES
848 bool "Namespaces support" if EXPERT
851 Provides the way to make tasks work with different objects using
852 the same id. For example same IPC id may refer to different objects
853 or same user id or pid may refer to different tasks when used in
854 different namespaces.
862 In this namespace tasks see different info provided with the
867 depends on (SYSVIPC || POSIX_MQUEUE)
870 In this namespace tasks work with IPC ids which correspond to
871 different IPC objects in different namespaces.
874 bool "User namespace (EXPERIMENTAL)"
875 depends on EXPERIMENTAL
878 This allows containers, i.e. vservers, to use user namespaces
879 to provide different user info for different servers.
883 bool "PID Namespaces"
886 Support process id namespaces. This allows having multiple
887 processes with the same pid as long as they are in different
888 pid namespaces. This is a building block of containers.
891 bool "Network namespace"
895 Allow user space to create what appear to be multiple instances
896 of the network stack.
900 config SCHED_AUTOGROUP
901 bool "Automatic process group scheduling"
905 select FAIR_GROUP_SCHED
907 This option optimizes the scheduler for common desktop workloads by
908 automatically creating and populating task groups. This separation
909 of workloads isolates aggressive CPU burners (like build jobs) from
910 desktop applications. Task group autogeneration is currently based
916 config SYSFS_DEPRECATED
917 bool "Enable deprecated sysfs features to support old userspace tools"
921 This option adds code that switches the layout of the "block" class
922 devices, to not show up in /sys/class/block/, but only in
925 This switch is only active when the sysfs.deprecated=1 boot option is
926 passed or the SYSFS_DEPRECATED_V2 option is set.
928 This option allows new kernels to run on old distributions and tools,
929 which might get confused by /sys/class/block/. Since 2007/2008 all
930 major distributions and tools handle this just fine.
932 Recent distributions and userspace tools after 2009/2010 depend on
933 the existence of /sys/class/block/, and will not work with this
936 Only if you are using a new kernel on an old distribution, you might
939 config SYSFS_DEPRECATED_V2
940 bool "Enable deprecated sysfs features by default"
943 depends on SYSFS_DEPRECATED
945 Enable deprecated sysfs by default.
947 See the CONFIG_SYSFS_DEPRECATED option for more details about this
950 Only if you are using a new kernel on an old distribution, you might
951 need to say Y here. Even then, odds are you would not need it
952 enabled, you can always pass the boot option if absolutely necessary.
955 bool "Kernel->user space relay support (formerly relayfs)"
957 This option enables support for relay interface support in
958 certain file systems (such as debugfs).
959 It is designed to provide an efficient mechanism for tools and
960 facilities to relay large amounts of data from kernel space to
965 config BLK_DEV_INITRD
966 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
967 depends on BROKEN || !FRV
969 The initial RAM filesystem is a ramfs which is loaded by the
970 boot loader (loadlin or lilo) and that is mounted as root
971 before the normal boot procedure. It is typically used to
972 load modules needed to mount the "real" root file system,
973 etc. See <file:Documentation/initrd.txt> for details.
975 If RAM disk support (BLK_DEV_RAM) is also included, this
976 also enables initial RAM disk (initrd) support and adds
977 15 Kbytes (more on some other architectures) to the kernel size.
987 config CC_OPTIMIZE_FOR_SIZE
988 bool "Optimize for size"
990 Enabling this option will pass "-Os" instead of "-O2" to gcc
991 resulting in a smaller kernel.
1002 bool "Configure standard kernel features (expert users)"
1003 # Unhide debug options, to make the on-by-default options visible
1006 This option allows certain base kernel options and settings
1007 to be disabled or tweaked. This is for specialized
1008 environments which can tolerate a "non-standard" kernel.
1009 Only use this if you really know what you are doing.
1012 bool "Enable 16-bit UID system calls" if EXPERT
1013 depends on ARM || BLACKFIN || CRIS || FRV || H8300 || X86_32 || M68K || (S390 && !64BIT) || SUPERH || SPARC32 || (SPARC64 && COMPAT) || UML || (X86_64 && IA32_EMULATION)
1016 This enables the legacy 16-bit UID syscall wrappers.
1018 config SYSCTL_SYSCALL
1019 bool "Sysctl syscall support" if EXPERT
1020 depends on PROC_SYSCTL
1024 sys_sysctl uses binary paths that have been found challenging
1025 to properly maintain and use. The interface in /proc/sys
1026 using paths with ascii names is now the primary path to this
1029 Almost nothing using the binary sysctl interface so if you are
1030 trying to save some space it is probably safe to disable this,
1031 making your kernel marginally smaller.
1033 If unsure say N here.
1036 bool "Load all symbols for debugging/ksymoops" if EXPERT
1039 Say Y here to let the kernel print out symbolic crash information and
1040 symbolic stack backtraces. This increases the size of the kernel
1041 somewhat, as all symbols have to be loaded into the kernel image.
1044 bool "Include all symbols in kallsyms"
1045 depends on DEBUG_KERNEL && KALLSYMS
1047 Normally kallsyms only contains the symbols of functions for nicer
1048 OOPS messages and backtraces (i.e., symbols from the text and inittext
1049 sections). This is sufficient for most cases. And only in very rare
1050 cases (e.g., when a debugger is used) all symbols are required (e.g.,
1051 names of variables from the data sections, etc).
1053 This option makes sure that all symbols are loaded into the kernel
1054 image (i.e., symbols from all sections) in cost of increased kernel
1055 size (depending on the kernel configuration, it may be 300KiB or
1056 something like this).
1058 Say N unless you really need all symbols.
1061 bool "Support for hot-pluggable devices" if EXPERT
1064 This option is provided for the case where no hotplug or uevent
1065 capabilities is wanted by the kernel. You should only consider
1066 disabling this option for embedded systems that do not use modules, a
1067 dynamic /dev tree, or dynamic device discovery. Just say Y.
1071 bool "Enable support for printk" if EXPERT
1073 This option enables normal printk support. Removing it
1074 eliminates most of the message strings from the kernel image
1075 and makes the kernel more or less silent. As this makes it
1076 very difficult to diagnose system problems, saying N here is
1077 strongly discouraged.
1080 bool "BUG() support" if EXPERT
1083 Disabling this option eliminates support for BUG and WARN, reducing
1084 the size of your kernel image and potentially quietly ignoring
1085 numerous fatal conditions. You should only consider disabling this
1086 option for embedded systems with no facilities for reporting errors.
1091 bool "Enable ELF core dumps" if EXPERT
1093 Enable support for generating core dumps. Disabling saves about 4k.
1096 config PCSPKR_PLATFORM
1097 bool "Enable PC-Speaker support" if EXPERT
1098 depends on HAVE_PCSPKR_PLATFORM
1102 This option allows to disable the internal PC-Speaker
1103 support, saving some memory.
1105 config HAVE_PCSPKR_PLATFORM
1110 bool "Enable full-sized data structures for core" if EXPERT
1112 Disabling this option reduces the size of miscellaneous core
1113 kernel data structures. This saves memory on small machines,
1114 but may reduce performance.
1117 bool "Enable futex support" if EXPERT
1121 Disabling this option will cause the kernel to be built without
1122 support for "fast userspace mutexes". The resulting kernel may not
1123 run glibc-based applications correctly.
1126 bool "Enable eventpoll support" if EXPERT
1130 Disabling this option will cause the kernel to be built without
1131 support for epoll family of system calls.
1134 bool "Enable signalfd() system call" if EXPERT
1138 Enable the signalfd() system call that allows to receive signals
1139 on a file descriptor.
1144 bool "Enable timerfd() system call" if EXPERT
1148 Enable the timerfd() system call that allows to receive timer
1149 events on a file descriptor.
1154 bool "Enable eventfd() system call" if EXPERT
1158 Enable the eventfd() system call that allows to receive both
1159 kernel notification (ie. KAIO) or userspace notifications.
1164 bool "Use full shmem filesystem" if EXPERT
1168 The shmem is an internal filesystem used to manage shared memory.
1169 It is backed by swap and manages resource limits. It is also exported
1170 to userspace as tmpfs if TMPFS is enabled. Disabling this
1171 option replaces shmem and tmpfs with the much simpler ramfs code,
1172 which may be appropriate on small systems without swap.
1175 bool "Enable AIO support" if EXPERT
1178 This option enables POSIX asynchronous I/O which may by used
1179 by some high performance threaded applications. Disabling
1180 this option saves about 7k.
1183 bool "Embedded system"
1186 This option should be enabled if compiling the kernel for
1187 an embedded system so certain expert options are available
1190 config HAVE_PERF_EVENTS
1193 See tools/perf/design.txt for details.
1195 config PERF_USE_VMALLOC
1198 See tools/perf/design.txt for details
1200 menu "Kernel Performance Events And Counters"
1203 bool "Kernel performance events and counters"
1204 default y if PROFILING
1205 depends on HAVE_PERF_EVENTS
1209 Enable kernel support for various performance events provided
1210 by software and hardware.
1212 Software events are supported either built-in or via the
1213 use of generic tracepoints.
1215 Most modern CPUs support performance events via performance
1216 counter registers. These registers count the number of certain
1217 types of hw events: such as instructions executed, cachemisses
1218 suffered, or branches mis-predicted - without slowing down the
1219 kernel or applications. These registers can also trigger interrupts
1220 when a threshold number of events have passed - and can thus be
1221 used to profile the code that runs on that CPU.
1223 The Linux Performance Event subsystem provides an abstraction of
1224 these software and hardware event capabilities, available via a
1225 system call and used by the "perf" utility in tools/perf/. It
1226 provides per task and per CPU counters, and it provides event
1227 capabilities on top of those.
1231 config DEBUG_PERF_USE_VMALLOC
1233 bool "Debug: use vmalloc to back perf mmap() buffers"
1234 depends on PERF_EVENTS && DEBUG_KERNEL
1235 select PERF_USE_VMALLOC
1237 Use vmalloc memory to back perf mmap() buffers.
1239 Mostly useful for debugging the vmalloc code on platforms
1240 that don't require it.
1246 config VM_EVENT_COUNTERS
1248 bool "Enable VM event counters for /proc/vmstat" if EXPERT
1250 VM event counters are needed for event counts to be shown.
1251 This option allows the disabling of the VM event counters
1252 on EXPERT systems. /proc/vmstat will only show page counts
1253 if VM event counters are disabled.
1257 bool "Enable PCI quirk workarounds" if EXPERT
1260 This enables workarounds for various PCI chipset
1261 bugs/quirks. Disable this only if your target machine is
1262 unaffected by PCI quirks.
1266 bool "Enable SLUB debugging support" if EXPERT
1267 depends on SLUB && SYSFS
1269 SLUB has extensive debug support features. Disabling these can
1270 result in significant savings in code size. This also disables
1271 SLUB sysfs support. /sys/slab will not exist and there will be
1272 no support for cache validation etc.
1275 bool "Disable heap randomization"
1278 Randomizing heap placement makes heap exploits harder, but it
1279 also breaks ancient binaries (including anything libc5 based).
1280 This option changes the bootup default to heap randomization
1281 disabled, and can be overridden at runtime by setting
1282 /proc/sys/kernel/randomize_va_space to 2.
1284 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1287 prompt "Choose SLAB allocator"
1290 This option allows to select a slab allocator.
1295 The regular slab allocator that is established and known to work
1296 well in all environments. It organizes cache hot objects in
1297 per cpu and per node queues.
1300 bool "SLUB (Unqueued Allocator)"
1302 SLUB is a slab allocator that minimizes cache line usage
1303 instead of managing queues of cached objects (SLAB approach).
1304 Per cpu caching is realized using slabs of objects instead
1305 of queues of objects. SLUB can use memory efficiently
1306 and has enhanced diagnostics. SLUB is the default choice for
1311 bool "SLOB (Simple Allocator)"
1313 SLOB replaces the stock allocator with a drastically simpler
1314 allocator. SLOB is generally more space efficient but
1315 does not perform as well on large systems.
1319 config MMAP_ALLOW_UNINITIALIZED
1320 bool "Allow mmapped anonymous memory to be uninitialized"
1321 depends on EXPERT && !MMU
1324 Normally, and according to the Linux spec, anonymous memory obtained
1325 from mmap() has it's contents cleared before it is passed to
1326 userspace. Enabling this config option allows you to request that
1327 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
1328 providing a huge performance boost. If this option is not enabled,
1329 then the flag will be ignored.
1331 This is taken advantage of by uClibc's malloc(), and also by
1332 ELF-FDPIC binfmt's brk and stack allocator.
1334 Because of the obvious security issues, this option should only be
1335 enabled on embedded devices where you control what is run in
1336 userspace. Since that isn't generally a problem on no-MMU systems,
1337 it is normally safe to say Y here.
1339 See Documentation/nommu-mmap.txt for more information.
1342 bool "Profiling support"
1344 Say Y here to enable the extended profiling support mechanisms used
1345 by profilers such as OProfile.
1348 # Place an empty function call at each tracepoint site. Can be
1349 # dynamically changed for a probe function.
1354 source "arch/Kconfig"
1356 endmenu # General setup
1358 config HAVE_GENERIC_DMA_COHERENT
1365 depends on SLAB || SLUB_DEBUG
1373 default 0 if BASE_FULL
1374 default 1 if !BASE_FULL
1377 bool "Enable loadable module support"
1379 Kernel modules are small pieces of compiled code which can
1380 be inserted in the running kernel, rather than being
1381 permanently built into the kernel. You use the "modprobe"
1382 tool to add (and sometimes remove) them. If you say Y here,
1383 many parts of the kernel can be built as modules (by
1384 answering M instead of Y where indicated): this is most
1385 useful for infrequently used options which are not required
1386 for booting. For more information, see the man pages for
1387 modprobe, lsmod, modinfo, insmod and rmmod.
1389 If you say Y here, you will need to run "make
1390 modules_install" to put the modules under /lib/modules/
1391 where modprobe can find them (you may need to be root to do
1398 config MODULE_FORCE_LOAD
1399 bool "Forced module loading"
1402 Allow loading of modules without version information (ie. modprobe
1403 --force). Forced module loading sets the 'F' (forced) taint flag and
1404 is usually a really bad idea.
1406 config MODULE_UNLOAD
1407 bool "Module unloading"
1409 Without this option you will not be able to unload any
1410 modules (note that some modules may not be unloadable
1411 anyway), which makes your kernel smaller, faster
1412 and simpler. If unsure, say Y.
1414 config MODULE_FORCE_UNLOAD
1415 bool "Forced module unloading"
1416 depends on MODULE_UNLOAD && EXPERIMENTAL
1418 This option allows you to force a module to unload, even if the
1419 kernel believes it is unsafe: the kernel will remove the module
1420 without waiting for anyone to stop using it (using the -f option to
1421 rmmod). This is mainly for kernel developers and desperate users.
1425 bool "Module versioning support"
1427 Usually, you have to use modules compiled with your kernel.
1428 Saying Y here makes it sometimes possible to use modules
1429 compiled for different kernels, by adding enough information
1430 to the modules to (hopefully) spot any changes which would
1431 make them incompatible with the kernel you are running. If
1434 config MODULE_SRCVERSION_ALL
1435 bool "Source checksum for all modules"
1437 Modules which contain a MODULE_VERSION get an extra "srcversion"
1438 field inserted into their modinfo section, which contains a
1439 sum of the source files which made it. This helps maintainers
1440 see exactly which source was used to build a module (since
1441 others sometimes change the module source without updating
1442 the version). With this option, such a "srcversion" field
1443 will be created for all modules. If unsure, say N.
1447 config INIT_ALL_POSSIBLE
1450 Back when each arch used to define their own cpu_online_mask and
1451 cpu_possible_mask, some of them chose to initialize cpu_possible_mask
1452 with all 1s, and others with all 0s. When they were centralised,
1453 it was better to provide this option than to break all the archs
1454 and have several arch maintainers pursuing me down dark alleys.
1459 depends on (SMP && MODULE_UNLOAD) || HOTPLUG_CPU
1461 Need stop_machine() primitive.
1463 source "block/Kconfig"
1465 config PREEMPT_NOTIFIERS
1472 source "kernel/Kconfig.locks"