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"
26 config BUILDTIME_EXTABLE_SORT
36 depends on BROKEN || !SMP
39 config INIT_ENV_ARG_LIMIT
44 Maximum of each of the number of arguments and environment
45 variables passed to init from the kernel command line.
49 string "Cross-compiler tool prefix"
51 Same as running 'make CROSS_COMPILE=prefix-' but stored for
52 default make runs in this kernel build directory. You don't
53 need to set this unless you want the configured kernel build
54 directory to select the cross-compiler automatically.
57 bool "Compile also drivers which will not load"
60 Some drivers can be compiled on a different platform than they are
61 intended to be run on. Despite they cannot be loaded there (or even
62 when they load they cannot be used due to missing HW support),
63 developers still, opposing to distributors, might want to build such
64 drivers to compile-test them.
66 If you are a developer and want to build everything available, say Y
67 here. If you are a user/distributor, say N here to exclude useless
68 drivers to be distributed.
71 string "Local version - append to kernel release"
73 Append an extra string to the end of your kernel version.
74 This will show up when you type uname, for example.
75 The string you set here will be appended after the contents of
76 any files with a filename matching localversion* in your
77 object and source tree, in that order. Your total string can
78 be a maximum of 64 characters.
80 config LOCALVERSION_AUTO
81 bool "Automatically append version information to the version string"
84 This will try to automatically determine if the current tree is a
85 release tree by looking for git tags that belong to the current
88 A string of the format -gxxxxxxxx will be added to the localversion
89 if a git-based tree is found. The string generated by this will be
90 appended after any matching localversion* files, and after the value
91 set in CONFIG_LOCALVERSION.
93 (The actual string used here is the first eight characters produced
94 by running the command:
96 $ git rev-parse --verify HEAD
98 which is done within the script "scripts/setlocalversion".)
100 config HAVE_KERNEL_GZIP
103 config HAVE_KERNEL_BZIP2
106 config HAVE_KERNEL_LZMA
109 config HAVE_KERNEL_XZ
112 config HAVE_KERNEL_LZO
115 config HAVE_KERNEL_LZ4
119 prompt "Kernel compression mode"
121 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4
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. It provides a good balance
145 between compression ratio and decompression speed.
149 depends on HAVE_KERNEL_BZIP2
151 Its compression ratio and speed is intermediate.
152 Decompression speed is slowest among the choices. The kernel
153 size is about 10% smaller with bzip2, in comparison to gzip.
154 Bzip2 uses a large amount of memory. For modern kernels you
155 will need at least 8MB RAM or more for booting.
159 depends on HAVE_KERNEL_LZMA
161 This compression algorithm's ratio is best. Decompression speed
162 is between gzip and bzip2. Compression is slowest.
163 The kernel size is about 33% smaller with LZMA in comparison to gzip.
167 depends on HAVE_KERNEL_XZ
169 XZ uses the LZMA2 algorithm and instruction set specific
170 BCJ filters which can improve compression ratio of executable
171 code. The size of the kernel is about 30% smaller with XZ in
172 comparison to gzip. On architectures for which there is a BCJ
173 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ
174 will create a few percent smaller kernel than plain LZMA.
176 The speed is about the same as with LZMA: The decompression
177 speed of XZ is better than that of bzip2 but worse than gzip
178 and LZO. Compression is slow.
182 depends on HAVE_KERNEL_LZO
184 Its compression ratio is the poorest among the choices. The kernel
185 size is about 10% bigger than gzip; however its speed
186 (both compression and decompression) is the fastest.
190 depends on HAVE_KERNEL_LZ4
192 LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding.
193 A preliminary version of LZ4 de/compression tool is available at
194 <https://code.google.com/p/lz4/>.
196 Its compression ratio is worse than LZO. The size of the kernel
197 is about 8% bigger than LZO. But the decompression speed is
202 config DEFAULT_HOSTNAME
203 string "Default hostname"
206 This option determines the default system hostname before userspace
207 calls sethostname(2). The kernel traditionally uses "(none)" here,
208 but you may wish to use a different default here to make a minimal
209 system more usable with less configuration.
212 bool "Support for paging of anonymous memory (swap)"
213 depends on MMU && BLOCK
216 This option allows you to choose whether you want to have support
217 for so called swap devices or swap files in your kernel that are
218 used to provide more virtual memory than the actual RAM present
219 in your computer. If unsure say Y.
224 Inter Process Communication is a suite of library functions and
225 system calls which let processes (running programs) synchronize and
226 exchange information. It is generally considered to be a good thing,
227 and some programs won't run unless you say Y here. In particular, if
228 you want to run the DOS emulator dosemu under Linux (read the
229 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
230 you'll need to say Y here.
232 You can find documentation about IPC with "info ipc" and also in
233 section 6.4 of the Linux Programmer's Guide, available from
234 <http://www.tldp.org/guides.html>.
236 config SYSVIPC_SYSCTL
243 bool "POSIX Message Queues"
246 POSIX variant of message queues is a part of IPC. In POSIX message
247 queues every message has a priority which decides about succession
248 of receiving it by a process. If you want to compile and run
249 programs written e.g. for Solaris with use of its POSIX message
250 queues (functions mq_*) say Y here.
252 POSIX message queues are visible as a filesystem called 'mqueue'
253 and can be mounted somewhere if you want to do filesystem
254 operations on message queues.
258 config POSIX_MQUEUE_SYSCTL
260 depends on POSIX_MQUEUE
264 config CROSS_MEMORY_ATTACH
265 bool "Enable process_vm_readv/writev syscalls"
269 Enabling this option adds the system calls process_vm_readv and
270 process_vm_writev which allow a process with the correct privileges
271 to directly read from or write to another process' address space.
272 See the man page for more details.
275 bool "open by fhandle syscalls"
278 If you say Y here, a user level program will be able to map
279 file names to handle and then later use the handle for
280 different file system operations. This is useful in implementing
281 userspace file servers, which now track files using handles instead
282 of names. The handle would remain the same even if file names
283 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2)
287 bool "uselib syscall"
290 This option enables the uselib syscall, a system call used in the
291 dynamic linker from libc5 and earlier. glibc does not use this
292 system call. If you intend to run programs built on libc5 or
293 earlier, you may need to enable this syscall. Current systems
294 running glibc can safely disable this.
297 bool "Auditing support"
300 Enable auditing infrastructure that can be used with another
301 kernel subsystem, such as SELinux (which requires this for
302 logging of avc messages output). Does not do system-call
303 auditing without CONFIG_AUDITSYSCALL.
305 config HAVE_ARCH_AUDITSYSCALL
309 bool "Enable system-call auditing support"
310 depends on AUDIT && HAVE_ARCH_AUDITSYSCALL
311 default y if SECURITY_SELINUX
313 Enable low-overhead system-call auditing infrastructure that
314 can be used independently or with another kernel subsystem,
319 depends on AUDITSYSCALL
324 depends on AUDITSYSCALL
327 source "kernel/irq/Kconfig"
328 source "kernel/time/Kconfig"
330 menu "CPU/Task time and stats accounting"
332 config VIRT_CPU_ACCOUNTING
336 prompt "Cputime accounting"
337 default TICK_CPU_ACCOUNTING if !PPC64
338 default VIRT_CPU_ACCOUNTING_NATIVE if PPC64
340 # Kind of a stub config for the pure tick based cputime accounting
341 config TICK_CPU_ACCOUNTING
342 bool "Simple tick based cputime accounting"
343 depends on !S390 && !NO_HZ_FULL
345 This is the basic tick based cputime accounting that maintains
346 statistics about user, system and idle time spent on per jiffies
351 config VIRT_CPU_ACCOUNTING_NATIVE
352 bool "Deterministic task and CPU time accounting"
353 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL
354 select VIRT_CPU_ACCOUNTING
356 Select this option to enable more accurate task and CPU time
357 accounting. This is done by reading a CPU counter on each
358 kernel entry and exit and on transitions within the kernel
359 between system, softirq and hardirq state, so there is a
360 small performance impact. In the case of s390 or IBM POWER > 5,
361 this also enables accounting of stolen time on logically-partitioned
364 config VIRT_CPU_ACCOUNTING_GEN
365 bool "Full dynticks CPU time accounting"
366 depends on HAVE_CONTEXT_TRACKING
367 depends on HAVE_VIRT_CPU_ACCOUNTING_GEN
368 select VIRT_CPU_ACCOUNTING
369 select CONTEXT_TRACKING
371 Select this option to enable task and CPU time accounting on full
372 dynticks systems. This accounting is implemented by watching every
373 kernel-user boundaries using the context tracking subsystem.
374 The accounting is thus performed at the expense of some significant
377 For now this is only useful if you are working on the full
378 dynticks subsystem development.
382 config IRQ_TIME_ACCOUNTING
383 bool "Fine granularity task level IRQ time accounting"
384 depends on HAVE_IRQ_TIME_ACCOUNTING && !NO_HZ_FULL
386 Select this option to enable fine granularity task irq time
387 accounting. This is done by reading a timestamp on each
388 transitions between softirq and hardirq state, so there can be a
389 small performance impact.
391 If in doubt, say N here.
396 bool "Support window based load tracking"
398 depends on FAIR_GROUP_SCHED
400 This feature will allow the scheduler to maintain a tunable window
401 based set of metrics for tasks and runqueues. These metrics can be
402 used to guide task placement as well as task frequency requirements
403 for cpufreq governors.
405 config BSD_PROCESS_ACCT
406 bool "BSD Process Accounting"
409 If you say Y here, a user level program will be able to instruct the
410 kernel (via a special system call) to write process accounting
411 information to a file: whenever a process exits, information about
412 that process will be appended to the file by the kernel. The
413 information includes things such as creation time, owning user,
414 command name, memory usage, controlling terminal etc. (the complete
415 list is in the struct acct in <file:include/linux/acct.h>). It is
416 up to the user level program to do useful things with this
417 information. This is generally a good idea, so say Y.
419 config BSD_PROCESS_ACCT_V3
420 bool "BSD Process Accounting version 3 file format"
421 depends on BSD_PROCESS_ACCT
424 If you say Y here, the process accounting information is written
425 in a new file format that also logs the process IDs of each
426 process and it's parent. Note that this file format is incompatible
427 with previous v0/v1/v2 file formats, so you will need updated tools
428 for processing it. A preliminary version of these tools is available
429 at <http://www.gnu.org/software/acct/>.
432 bool "Export task/process statistics through netlink"
437 Export selected statistics for tasks/processes through the
438 generic netlink interface. Unlike BSD process accounting, the
439 statistics are available during the lifetime of tasks/processes as
440 responses to commands. Like BSD accounting, they are sent to user
445 config TASK_DELAY_ACCT
446 bool "Enable per-task delay accounting"
450 Collect information on time spent by a task waiting for system
451 resources like cpu, synchronous block I/O completion and swapping
452 in pages. Such statistics can help in setting a task's priorities
453 relative to other tasks for cpu, io, rss limits etc.
458 bool "Enable extended accounting over taskstats"
461 Collect extended task accounting data and send the data
462 to userland for processing over the taskstats interface.
466 config TASK_IO_ACCOUNTING
467 bool "Enable per-task storage I/O accounting"
468 depends on TASK_XACCT
470 Collect information on the number of bytes of storage I/O which this
475 endmenu # "CPU/Task time and stats accounting"
481 default y if !PREEMPT && SMP
483 This option selects the RCU implementation that is
484 designed for very large SMP system with hundreds or
485 thousands of CPUs. It also scales down nicely to
492 This option selects the RCU implementation that is
493 designed for very large SMP systems with hundreds or
494 thousands of CPUs, but for which real-time response
495 is also required. It also scales down nicely to
498 Select this option if you are unsure.
502 default y if !PREEMPT && !SMP
504 This option selects the RCU implementation that is
505 designed for UP systems from which real-time response
506 is not required. This option greatly reduces the
507 memory footprint of RCU.
510 bool "Make expert-level adjustments to RCU configuration"
513 This option needs to be enabled if you wish to make
514 expert-level adjustments to RCU configuration. By default,
515 no such adjustments can be made, which has the often-beneficial
516 side-effect of preventing "make oldconfig" from asking you all
517 sorts of detailed questions about how you would like numerous
518 obscure RCU options to be set up.
520 Say Y if you need to make expert-level adjustments to RCU.
522 Say N if you are unsure.
527 This option selects the sleepable version of RCU. This version
528 permits arbitrary sleeping or blocking within RCU read-side critical
536 This option enables a task-based RCU implementation that uses
537 only voluntary context switch (not preemption!), idle, and
538 user-mode execution as quiescent states.
540 config RCU_STALL_COMMON
541 def_bool ( TREE_RCU || PREEMPT_RCU || RCU_TRACE )
543 This option enables RCU CPU stall code that is common between
544 the TINY and TREE variants of RCU. The purpose is to allow
545 the tiny variants to disable RCU CPU stall warnings, while
546 making these warnings mandatory for the tree variants.
548 config CONTEXT_TRACKING
551 config CONTEXT_TRACKING_FORCE
552 bool "Force context tracking"
553 depends on CONTEXT_TRACKING
554 default y if !NO_HZ_FULL
556 The major pre-requirement for full dynticks to work is to
557 support the context tracking subsystem. But there are also
558 other dependencies to provide in order to make the full
561 This option stands for testing when an arch implements the
562 context tracking backend but doesn't yet fullfill all the
563 requirements to make the full dynticks feature working.
564 Without the full dynticks, there is no way to test the support
565 for context tracking and the subsystems that rely on it: RCU
566 userspace extended quiescent state and tickless cputime
567 accounting. This option copes with the absence of the full
568 dynticks subsystem by forcing the context tracking on all
571 Say Y only if you're working on the development of an
572 architecture backend for the context tracking.
574 Say N otherwise, this option brings an overhead that you
575 don't want in production.
579 int "Tree-based hierarchical RCU fanout value"
582 depends on (TREE_RCU || PREEMPT_RCU) && RCU_EXPERT
586 This option controls the fanout of hierarchical implementations
587 of RCU, allowing RCU to work efficiently on machines with
588 large numbers of CPUs. This value must be at least the fourth
589 root of NR_CPUS, which allows NR_CPUS to be insanely large.
590 The default value of RCU_FANOUT should be used for production
591 systems, but if you are stress-testing the RCU implementation
592 itself, small RCU_FANOUT values allow you to test large-system
593 code paths on small(er) systems.
595 Select a specific number if testing RCU itself.
596 Take the default if unsure.
598 config RCU_FANOUT_LEAF
599 int "Tree-based hierarchical RCU leaf-level fanout value"
602 depends on (TREE_RCU || PREEMPT_RCU) && RCU_EXPERT
605 This option controls the leaf-level fanout of hierarchical
606 implementations of RCU, and allows trading off cache misses
607 against lock contention. Systems that synchronize their
608 scheduling-clock interrupts for energy-efficiency reasons will
609 want the default because the smaller leaf-level fanout keeps
610 lock contention levels acceptably low. Very large systems
611 (hundreds or thousands of CPUs) will instead want to set this
612 value to the maximum value possible in order to reduce the
613 number of cache misses incurred during RCU's grace-period
614 initialization. These systems tend to run CPU-bound, and thus
615 are not helped by synchronized interrupts, and thus tend to
616 skew them, which reduces lock contention enough that large
617 leaf-level fanouts work well.
619 Select a specific number if testing RCU itself.
621 Select the maximum permissible value for large systems.
623 Take the default if unsure.
625 config RCU_FAST_NO_HZ
626 bool "Accelerate last non-dyntick-idle CPU's grace periods"
627 depends on NO_HZ_COMMON && SMP && RCU_EXPERT
630 This option permits CPUs to enter dynticks-idle state even if
631 they have RCU callbacks queued, and prevents RCU from waking
632 these CPUs up more than roughly once every four jiffies (by
633 default, you can adjust this using the rcutree.rcu_idle_gp_delay
634 parameter), thus improving energy efficiency. On the other
635 hand, this option increases the duration of RCU grace periods,
636 for example, slowing down synchronize_rcu().
638 Say Y if energy efficiency is critically important, and you
639 don't care about increased grace-period durations.
641 Say N if you are unsure.
643 config TREE_RCU_TRACE
644 def_bool RCU_TRACE && ( TREE_RCU || PREEMPT_RCU )
647 This option provides tracing for the TREE_RCU and
648 PREEMPT_RCU implementations, permitting Makefile to
649 trivially select kernel/rcutree_trace.c.
652 bool "Enable RCU priority boosting"
653 depends on RT_MUTEXES && PREEMPT_RCU && RCU_EXPERT
656 This option boosts the priority of preempted RCU readers that
657 block the current preemptible RCU grace period for too long.
658 This option also prevents heavy loads from blocking RCU
659 callback invocation for all flavors of RCU.
661 Say Y here if you are working with real-time apps or heavy loads
662 Say N here if you are unsure.
664 config RCU_KTHREAD_PRIO
665 int "Real-time priority to use for RCU worker threads"
666 range 1 99 if RCU_BOOST
667 range 0 99 if !RCU_BOOST
668 default 1 if RCU_BOOST
669 default 0 if !RCU_BOOST
670 depends on RCU_EXPERT
672 This option specifies the SCHED_FIFO priority value that will be
673 assigned to the rcuc/n and rcub/n threads and is also the value
674 used for RCU_BOOST (if enabled). If you are working with a
675 real-time application that has one or more CPU-bound threads
676 running at a real-time priority level, you should set
677 RCU_KTHREAD_PRIO to a priority higher than the highest-priority
678 real-time CPU-bound application thread. The default RCU_KTHREAD_PRIO
679 value of 1 is appropriate in the common case, which is real-time
680 applications that do not have any CPU-bound threads.
682 Some real-time applications might not have a single real-time
683 thread that saturates a given CPU, but instead might have
684 multiple real-time threads that, taken together, fully utilize
685 that CPU. In this case, you should set RCU_KTHREAD_PRIO to
686 a priority higher than the lowest-priority thread that is
687 conspiring to prevent the CPU from running any non-real-time
688 tasks. For example, if one thread at priority 10 and another
689 thread at priority 5 are between themselves fully consuming
690 the CPU time on a given CPU, then RCU_KTHREAD_PRIO should be
691 set to priority 6 or higher.
693 Specify the real-time priority, or take the default if unsure.
695 config RCU_BOOST_DELAY
696 int "Milliseconds to delay boosting after RCU grace-period start"
701 This option specifies the time to wait after the beginning of
702 a given grace period before priority-boosting preempted RCU
703 readers blocking that grace period. Note that any RCU reader
704 blocking an expedited RCU grace period is boosted immediately.
706 Accept the default if unsure.
709 bool "Offload RCU callback processing from boot-selected CPUs"
710 depends on TREE_RCU || PREEMPT_RCU
711 depends on RCU_EXPERT || NO_HZ_FULL
714 Use this option to reduce OS jitter for aggressive HPC or
715 real-time workloads. It can also be used to offload RCU
716 callback invocation to energy-efficient CPUs in battery-powered
717 asymmetric multiprocessors.
719 This option offloads callback invocation from the set of
720 CPUs specified at boot time by the rcu_nocbs parameter.
721 For each such CPU, a kthread ("rcuox/N") will be created to
722 invoke callbacks, where the "N" is the CPU being offloaded,
723 and where the "x" is "b" for RCU-bh, "p" for RCU-preempt, and
724 "s" for RCU-sched. Nothing prevents this kthread from running
725 on the specified CPUs, but (1) the kthreads may be preempted
726 between each callback, and (2) affinity or cgroups can be used
727 to force the kthreads to run on whatever set of CPUs is desired.
729 Say Y here if you want to help to debug reduced OS jitter.
730 Say N here if you are unsure.
733 prompt "Build-forced no-CBs CPUs"
734 default RCU_NOCB_CPU_NONE
735 depends on RCU_NOCB_CPU
737 This option allows no-CBs CPUs (whose RCU callbacks are invoked
738 from kthreads rather than from softirq context) to be specified
739 at build time. Additional no-CBs CPUs may be specified by
740 the rcu_nocbs= boot parameter.
742 config RCU_NOCB_CPU_NONE
743 bool "No build_forced no-CBs CPUs"
745 This option does not force any of the CPUs to be no-CBs CPUs.
746 Only CPUs designated by the rcu_nocbs= boot parameter will be
747 no-CBs CPUs, whose RCU callbacks will be invoked by per-CPU
748 kthreads whose names begin with "rcuo". All other CPUs will
749 invoke their own RCU callbacks in softirq context.
751 Select this option if you want to choose no-CBs CPUs at
752 boot time, for example, to allow testing of different no-CBs
753 configurations without having to rebuild the kernel each time.
755 config RCU_NOCB_CPU_ZERO
756 bool "CPU 0 is a build_forced no-CBs CPU"
758 This option forces CPU 0 to be a no-CBs CPU, so that its RCU
759 callbacks are invoked by a per-CPU kthread whose name begins
760 with "rcuo". Additional CPUs may be designated as no-CBs
761 CPUs using the rcu_nocbs= boot parameter will be no-CBs CPUs.
762 All other CPUs will invoke their own RCU callbacks in softirq
765 Select this if CPU 0 needs to be a no-CBs CPU for real-time
766 or energy-efficiency reasons, but the real reason it exists
767 is to ensure that randconfig testing covers mixed systems.
769 config RCU_NOCB_CPU_ALL
770 bool "All CPUs are build_forced no-CBs CPUs"
772 This option forces all CPUs to be no-CBs CPUs. The rcu_nocbs=
773 boot parameter will be ignored. All CPUs' RCU callbacks will
774 be executed in the context of per-CPU rcuo kthreads created for
775 this purpose. Assuming that the kthreads whose names start with
776 "rcuo" are bound to "housekeeping" CPUs, this reduces OS jitter
777 on the remaining CPUs, but might decrease memory locality during
778 RCU-callback invocation, thus potentially degrading throughput.
780 Select this if all CPUs need to be no-CBs CPUs for real-time
781 or energy-efficiency reasons.
785 config RCU_EXPEDITE_BOOT
789 This option enables expedited grace periods at boot time,
790 as if rcu_expedite_gp() had been invoked early in boot.
791 The corresponding rcu_unexpedite_gp() is invoked from
792 rcu_end_inkernel_boot(), which is intended to be invoked
793 at the end of the kernel-only boot sequence, just before
796 Accept the default if unsure.
798 endmenu # "RCU Subsystem"
805 tristate "Kernel .config support"
808 This option enables the complete Linux kernel ".config" file
809 contents to be saved in the kernel. It provides documentation
810 of which kernel options are used in a running kernel or in an
811 on-disk kernel. This information can be extracted from the kernel
812 image file with the script scripts/extract-ikconfig and used as
813 input to rebuild the current kernel or to build another kernel.
814 It can also be extracted from a running kernel by reading
815 /proc/config.gz if enabled (below).
818 bool "Enable access to .config through /proc/config.gz"
819 depends on IKCONFIG && PROC_FS
821 This option enables access to the kernel configuration file
822 through /proc/config.gz.
825 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
830 Select the minimal kernel log buffer size as a power of 2.
831 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config
832 parameter, see below. Any higher size also might be forced
833 by "log_buf_len" boot parameter.
843 config LOG_CPU_MAX_BUF_SHIFT
844 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)"
847 default 12 if !BASE_SMALL
848 default 0 if BASE_SMALL
851 This option allows to increase the default ring buffer size
852 according to the number of CPUs. The value defines the contribution
853 of each CPU as a power of 2. The used space is typically only few
854 lines however it might be much more when problems are reported,
857 The increased size means that a new buffer has to be allocated and
858 the original static one is unused. It makes sense only on systems
859 with more CPUs. Therefore this value is used only when the sum of
860 contributions is greater than the half of the default kernel ring
861 buffer as defined by LOG_BUF_SHIFT. The default values are set
862 so that more than 64 CPUs are needed to trigger the allocation.
864 Also this option is ignored when "log_buf_len" kernel parameter is
865 used as it forces an exact (power of two) size of the ring buffer.
867 The number of possible CPUs is used for this computation ignoring
868 hotplugging making the compuation optimal for the the worst case
869 scenerio while allowing a simple algorithm to be used from bootup.
871 Examples shift values and their meaning:
872 17 => 128 KB for each CPU
873 16 => 64 KB for each CPU
874 15 => 32 KB for each CPU
875 14 => 16 KB for each CPU
876 13 => 8 KB for each CPU
877 12 => 4 KB for each CPU
880 # Architectures with an unreliable sched_clock() should select this:
882 config HAVE_UNSTABLE_SCHED_CLOCK
885 config GENERIC_SCHED_CLOCK
889 # For architectures that want to enable the support for NUMA-affine scheduler
892 config ARCH_SUPPORTS_NUMA_BALANCING
896 # For architectures that prefer to flush all TLBs after a number of pages
897 # are unmapped instead of sending one IPI per page to flush. The architecture
898 # must provide guarantees on what happens if a clean TLB cache entry is
899 # written after the unmap. Details are in mm/rmap.c near the check for
900 # should_defer_flush. The architecture should also consider if the full flush
901 # and the refill costs are offset by the savings of sending fewer IPIs.
902 config ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
906 # For architectures that know their GCC __int128 support is sound
908 config ARCH_SUPPORTS_INT128
911 # For architectures that (ab)use NUMA to represent different memory regions
912 # all cpu-local but of different latencies, such as SuperH.
914 config ARCH_WANT_NUMA_VARIABLE_LOCALITY
917 config NUMA_BALANCING
918 bool "Memory placement aware NUMA scheduler"
919 depends on ARCH_SUPPORTS_NUMA_BALANCING
920 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
921 depends on SMP && NUMA && MIGRATION
923 This option adds support for automatic NUMA aware memory/task placement.
924 The mechanism is quite primitive and is based on migrating memory when
925 it has references to the node the task is running on.
927 This system will be inactive on UMA systems.
929 config NUMA_BALANCING_DEFAULT_ENABLED
930 bool "Automatically enable NUMA aware memory/task placement"
932 depends on NUMA_BALANCING
934 If set, automatic NUMA balancing will be enabled if running on a NUMA
938 bool "Control Group support"
941 This option adds support for grouping sets of processes together, for
942 use with process control subsystems such as Cpusets, CFS, memory
943 controls or device isolation.
945 - Documentation/scheduler/sched-design-CFS.txt (CFS)
946 - Documentation/cgroups/ (features for grouping, isolation
947 and resource control)
954 bool "Example debug cgroup subsystem"
957 This option enables a simple cgroup subsystem that
958 exports useful debugging information about the cgroups
963 config CGROUP_FREEZER
964 bool "Freezer cgroup subsystem"
966 Provides a way to freeze and unfreeze all tasks in a
970 bool "PIDs cgroup subsystem"
972 Provides enforcement of process number limits in the scope of a
973 cgroup. Any attempt to fork more processes than is allowed in the
974 cgroup will fail. PIDs are fundamentally a global resource because it
975 is fairly trivial to reach PID exhaustion before you reach even a
976 conservative kmemcg limit. As a result, it is possible to grind a
977 system to halt without being limited by other cgroup policies. The
978 PIDs cgroup subsystem is designed to stop this from happening.
980 It should be noted that organisational operations (such as attaching
981 to a cgroup hierarchy will *not* be blocked by the PIDs subsystem),
982 since the PIDs limit only affects a process's ability to fork, not to
986 bool "Device controller for cgroups"
988 Provides a cgroup implementing whitelists for devices which
989 a process in the cgroup can mknod or open.
992 bool "Cpuset support"
994 This option will let you create and manage CPUSETs which
995 allow dynamically partitioning a system into sets of CPUs and
996 Memory Nodes and assigning tasks to run only within those sets.
997 This is primarily useful on large SMP or NUMA systems.
1001 config PROC_PID_CPUSET
1002 bool "Include legacy /proc/<pid>/cpuset file"
1006 config CGROUP_CPUACCT
1007 bool "Simple CPU accounting cgroup subsystem"
1009 Provides a simple Resource Controller for monitoring the
1010 total CPU consumed by the tasks in a cgroup.
1012 config CGROUP_SCHEDTUNE
1013 bool "CFS tasks boosting cgroup subsystem (EXPERIMENTAL)"
1014 depends on SCHED_TUNE
1016 This option provides the "schedtune" controller which improves the
1017 flexibility of the task boosting mechanism by introducing the support
1018 to define "per task" boost values.
1020 This new controller:
1021 1. allows only a two layers hierarchy, where the root defines the
1022 system-wide boost value and its direct childrens define each one a
1023 different "class of tasks" to be boosted with a different value
1024 2. supports up to 16 different task classes, each one which could be
1025 configured with a different boost value
1033 bool "Memory Resource Controller for Control Groups"
1037 Provides a memory resource controller that manages both anonymous
1038 memory and page cache. (See Documentation/cgroups/memory.txt)
1041 bool "Memory Resource Controller Swap Extension"
1042 depends on MEMCG && SWAP
1044 Add swap management feature to memory resource controller. When you
1045 enable this, you can limit mem+swap usage per cgroup. In other words,
1046 when you disable this, memory resource controller has no cares to
1047 usage of swap...a process can exhaust all of the swap. This extension
1048 is useful when you want to avoid exhaustion swap but this itself
1049 adds more overheads and consumes memory for remembering information.
1050 Especially if you use 32bit system or small memory system, please
1051 be careful about enabling this. When memory resource controller
1052 is disabled by boot option, this will be automatically disabled and
1053 there will be no overhead from this. Even when you set this config=y,
1054 if boot option "swapaccount=0" is set, swap will not be accounted.
1055 Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page
1056 size is 4096bytes, 512k per 1Gbytes of swap.
1057 config MEMCG_SWAP_ENABLED
1058 bool "Memory Resource Controller Swap Extension enabled by default"
1059 depends on MEMCG_SWAP
1062 Memory Resource Controller Swap Extension comes with its price in
1063 a bigger memory consumption. General purpose distribution kernels
1064 which want to enable the feature but keep it disabled by default
1065 and let the user enable it by swapaccount=1 boot command line
1066 parameter should have this option unselected.
1067 For those who want to have the feature enabled by default should
1068 select this option (if, for some reason, they need to disable it
1069 then swapaccount=0 does the trick).
1071 bool "Memory Resource Controller Kernel Memory accounting"
1073 depends on SLUB || SLAB
1075 The Kernel Memory extension for Memory Resource Controller can limit
1076 the amount of memory used by kernel objects in the system. Those are
1077 fundamentally different from the entities handled by the standard
1078 Memory Controller, which are page-based, and can be swapped. Users of
1079 the kmem extension can use it to guarantee that no group of processes
1080 will ever exhaust kernel resources alone.
1082 config CGROUP_HUGETLB
1083 bool "HugeTLB Resource Controller for Control Groups"
1084 depends on HUGETLB_PAGE
1088 Provides a cgroup Resource Controller for HugeTLB pages.
1089 When you enable this, you can put a per cgroup limit on HugeTLB usage.
1090 The limit is enforced during page fault. Since HugeTLB doesn't
1091 support page reclaim, enforcing the limit at page fault time implies
1092 that, the application will get SIGBUS signal if it tries to access
1093 HugeTLB pages beyond its limit. This requires the application to know
1094 beforehand how much HugeTLB pages it would require for its use. The
1095 control group is tracked in the third page lru pointer. This means
1096 that we cannot use the controller with huge page less than 3 pages.
1099 bool "Enable perf_event per-cpu per-container group (cgroup) monitoring"
1100 depends on PERF_EVENTS && CGROUPS
1102 This option extends the per-cpu mode to restrict monitoring to
1103 threads which belong to the cgroup specified and run on the
1108 menuconfig CGROUP_SCHED
1109 bool "Group CPU scheduler"
1112 This feature lets CPU scheduler recognize task groups and control CPU
1113 bandwidth allocation to such task groups. It uses cgroups to group
1117 config FAIR_GROUP_SCHED
1118 bool "Group scheduling for SCHED_OTHER"
1119 depends on CGROUP_SCHED
1120 default CGROUP_SCHED
1122 config CFS_BANDWIDTH
1123 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED"
1124 depends on FAIR_GROUP_SCHED
1127 This option allows users to define CPU bandwidth rates (limits) for
1128 tasks running within the fair group scheduler. Groups with no limit
1129 set are considered to be unconstrained and will run with no
1131 See tip/Documentation/scheduler/sched-bwc.txt for more information.
1133 config RT_GROUP_SCHED
1134 bool "Group scheduling for SCHED_RR/FIFO"
1135 depends on CGROUP_SCHED
1138 This feature lets you explicitly allocate real CPU bandwidth
1139 to task groups. If enabled, it will also make it impossible to
1140 schedule realtime tasks for non-root users until you allocate
1141 realtime bandwidth for them.
1142 See Documentation/scheduler/sched-rt-group.txt for more information.
1147 bool "Block IO controller"
1151 Generic block IO controller cgroup interface. This is the common
1152 cgroup interface which should be used by various IO controlling
1155 Currently, CFQ IO scheduler uses it to recognize task groups and
1156 control disk bandwidth allocation (proportional time slice allocation)
1157 to such task groups. It is also used by bio throttling logic in
1158 block layer to implement upper limit in IO rates on a device.
1160 This option only enables generic Block IO controller infrastructure.
1161 One needs to also enable actual IO controlling logic/policy. For
1162 enabling proportional weight division of disk bandwidth in CFQ, set
1163 CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set
1164 CONFIG_BLK_DEV_THROTTLING=y.
1166 See Documentation/cgroups/blkio-controller.txt for more information.
1168 config DEBUG_BLK_CGROUP
1169 bool "Enable Block IO controller debugging"
1170 depends on BLK_CGROUP
1173 Enable some debugging help. Currently it exports additional stat
1174 files in a cgroup which can be useful for debugging.
1176 config CGROUP_WRITEBACK
1178 depends on MEMCG && BLK_CGROUP
1183 config CHECKPOINT_RESTORE
1184 bool "Checkpoint/restore support" if EXPERT
1185 select PROC_CHILDREN
1188 Enables additional kernel features in a sake of checkpoint/restore.
1189 In particular it adds auxiliary prctl codes to setup process text,
1190 data and heap segment sizes, and a few additional /proc filesystem
1193 If unsure, say N here.
1195 menuconfig NAMESPACES
1196 bool "Namespaces support" if EXPERT
1197 depends on MULTIUSER
1200 Provides the way to make tasks work with different objects using
1201 the same id. For example same IPC id may refer to different objects
1202 or same user id or pid may refer to different tasks when used in
1203 different namespaces.
1208 bool "UTS namespace"
1211 In this namespace tasks see different info provided with the
1215 bool "IPC namespace"
1216 depends on (SYSVIPC || POSIX_MQUEUE)
1219 In this namespace tasks work with IPC ids which correspond to
1220 different IPC objects in different namespaces.
1223 bool "User namespace"
1226 This allows containers, i.e. vservers, to use user namespaces
1227 to provide different user info for different servers.
1229 When user namespaces are enabled in the kernel it is
1230 recommended that the MEMCG and MEMCG_KMEM options also be
1231 enabled and that user-space use the memory control groups to
1232 limit the amount of memory a memory unprivileged users can
1238 bool "PID Namespaces"
1241 Support process id namespaces. This allows having multiple
1242 processes with the same pid as long as they are in different
1243 pid namespaces. This is a building block of containers.
1246 bool "Network namespace"
1250 Allow user space to create what appear to be multiple instances
1251 of the network stack.
1255 config SCHED_AUTOGROUP
1256 bool "Automatic process group scheduling"
1259 select FAIR_GROUP_SCHED
1261 This option optimizes the scheduler for common desktop workloads by
1262 automatically creating and populating task groups. This separation
1263 of workloads isolates aggressive CPU burners (like build jobs) from
1264 desktop applications. Task group autogeneration is currently based
1268 bool "Boosting for CFS tasks (EXPERIMENTAL)"
1271 This option enables the system-wide support for task boosting.
1272 When this support is enabled a new sysctl interface is exposed to
1274 /proc/sys/kernel/sched_cfs_boost
1275 which allows to set a system-wide boost value in range [0..100].
1277 The currently boosting strategy is implemented in such a way that:
1278 - a 0% boost value requires to operate in "standard" mode by
1279 scheduling all tasks at the minimum capacities required by their
1281 - a 100% boost value requires to push at maximum the task
1282 performances, "regardless" of the incurred energy consumption
1284 A boost value in between these two boundaries is used to bias the
1285 power/performance trade-off, the higher the boost value the more the
1286 scheduler is biased toward performance boosting instead of energy
1289 Since this support exposes a single system-wide knob, the specified
1290 boost value is applied to all (CFS) tasks in the system.
1294 config DEFAULT_USE_ENERGY_AWARE
1295 bool "Default to enabling the Energy Aware Scheduler feature"
1298 This option defaults the ENERGY_AWARE scheduling feature to true,
1299 as without SCHED_DEBUG set this feature can't be enabled or disabled
1304 config SYSFS_DEPRECATED
1305 bool "Enable deprecated sysfs features to support old userspace tools"
1309 This option adds code that switches the layout of the "block" class
1310 devices, to not show up in /sys/class/block/, but only in
1313 This switch is only active when the sysfs.deprecated=1 boot option is
1314 passed or the SYSFS_DEPRECATED_V2 option is set.
1316 This option allows new kernels to run on old distributions and tools,
1317 which might get confused by /sys/class/block/. Since 2007/2008 all
1318 major distributions and tools handle this just fine.
1320 Recent distributions and userspace tools after 2009/2010 depend on
1321 the existence of /sys/class/block/, and will not work with this
1324 Only if you are using a new kernel on an old distribution, you might
1327 config SYSFS_DEPRECATED_V2
1328 bool "Enable deprecated sysfs features by default"
1331 depends on SYSFS_DEPRECATED
1333 Enable deprecated sysfs by default.
1335 See the CONFIG_SYSFS_DEPRECATED option for more details about this
1338 Only if you are using a new kernel on an old distribution, you might
1339 need to say Y here. Even then, odds are you would not need it
1340 enabled, you can always pass the boot option if absolutely necessary.
1343 bool "Kernel->user space relay support (formerly relayfs)"
1345 This option enables support for relay interface support in
1346 certain file systems (such as debugfs).
1347 It is designed to provide an efficient mechanism for tools and
1348 facilities to relay large amounts of data from kernel space to
1353 config BLK_DEV_INITRD
1354 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
1355 depends on BROKEN || !FRV
1357 The initial RAM filesystem is a ramfs which is loaded by the
1358 boot loader (loadlin or lilo) and that is mounted as root
1359 before the normal boot procedure. It is typically used to
1360 load modules needed to mount the "real" root file system,
1361 etc. See <file:Documentation/initrd.txt> for details.
1363 If RAM disk support (BLK_DEV_RAM) is also included, this
1364 also enables initial RAM disk (initrd) support and adds
1365 15 Kbytes (more on some other architectures) to the kernel size.
1371 source "usr/Kconfig"
1375 config CC_OPTIMIZE_FOR_SIZE
1376 bool "Optimize for size"
1378 Enabling this option will pass "-Os" instead of "-O2" to
1379 your compiler resulting in a smaller kernel.
1392 config SYSCTL_EXCEPTION_TRACE
1395 Enable support for /proc/sys/debug/exception-trace.
1397 config SYSCTL_ARCH_UNALIGN_NO_WARN
1400 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap
1401 Allows arch to define/use @no_unaligned_warning to possibly warn
1402 about unaligned access emulation going on under the hood.
1404 config SYSCTL_ARCH_UNALIGN_ALLOW
1407 Enable support for /proc/sys/kernel/unaligned-trap
1408 Allows arches to define/use @unaligned_enabled to runtime toggle
1409 the unaligned access emulation.
1410 see arch/parisc/kernel/unaligned.c for reference
1412 config HAVE_PCSPKR_PLATFORM
1415 # interpreter that classic socket filters depend on
1420 bool "Configure standard kernel features (expert users)"
1421 # Unhide debug options, to make the on-by-default options visible
1424 This option allows certain base kernel options and settings
1425 to be disabled or tweaked. This is for specialized
1426 environments which can tolerate a "non-standard" kernel.
1427 Only use this if you really know what you are doing.
1430 bool "Enable 16-bit UID system calls" if EXPERT
1431 depends on HAVE_UID16 && MULTIUSER
1434 This enables the legacy 16-bit UID syscall wrappers.
1437 bool "Multiple users, groups and capabilities support" if EXPERT
1440 This option enables support for non-root users, groups and
1443 If you say N here, all processes will run with UID 0, GID 0, and all
1444 possible capabilities. Saying N here also compiles out support for
1445 system calls related to UIDs, GIDs, and capabilities, such as setuid,
1448 If unsure, say Y here.
1450 config SGETMASK_SYSCALL
1451 bool "sgetmask/ssetmask syscalls support" if EXPERT
1452 def_bool PARISC || MN10300 || BLACKFIN || M68K || PPC || MIPS || X86 || SPARC || CRIS || MICROBLAZE || SUPERH
1454 sys_sgetmask and sys_ssetmask are obsolete system calls
1455 no longer supported in libc but still enabled by default in some
1458 If unsure, leave the default option here.
1460 config SYSFS_SYSCALL
1461 bool "Sysfs syscall support" if EXPERT
1464 sys_sysfs is an obsolete system call no longer supported in libc.
1465 Note that disabling this option is more secure but might break
1466 compatibility with some systems.
1468 If unsure say Y here.
1470 config SYSCTL_SYSCALL
1471 bool "Sysctl syscall support" if EXPERT
1472 depends on PROC_SYSCTL
1476 sys_sysctl uses binary paths that have been found challenging
1477 to properly maintain and use. The interface in /proc/sys
1478 using paths with ascii names is now the primary path to this
1481 Almost nothing using the binary sysctl interface so if you are
1482 trying to save some space it is probably safe to disable this,
1483 making your kernel marginally smaller.
1485 If unsure say N here.
1488 bool "Load all symbols for debugging/ksymoops" if EXPERT
1491 Say Y here to let the kernel print out symbolic crash information and
1492 symbolic stack backtraces. This increases the size of the kernel
1493 somewhat, as all symbols have to be loaded into the kernel image.
1496 bool "Include all symbols in kallsyms"
1497 depends on DEBUG_KERNEL && KALLSYMS
1499 Normally kallsyms only contains the symbols of functions for nicer
1500 OOPS messages and backtraces (i.e., symbols from the text and inittext
1501 sections). This is sufficient for most cases. And only in very rare
1502 cases (e.g., when a debugger is used) all symbols are required (e.g.,
1503 names of variables from the data sections, etc).
1505 This option makes sure that all symbols are loaded into the kernel
1506 image (i.e., symbols from all sections) in cost of increased kernel
1507 size (depending on the kernel configuration, it may be 300KiB or
1508 something like this).
1510 Say N unless you really need all symbols.
1514 bool "Enable support for printk" if EXPERT
1517 This option enables normal printk support. Removing it
1518 eliminates most of the message strings from the kernel image
1519 and makes the kernel more or less silent. As this makes it
1520 very difficult to diagnose system problems, saying N here is
1521 strongly discouraged.
1524 bool "BUG() support" if EXPERT
1527 Disabling this option eliminates support for BUG and WARN, reducing
1528 the size of your kernel image and potentially quietly ignoring
1529 numerous fatal conditions. You should only consider disabling this
1530 option for embedded systems with no facilities for reporting errors.
1536 bool "Enable ELF core dumps" if EXPERT
1538 Enable support for generating core dumps. Disabling saves about 4k.
1541 config PCSPKR_PLATFORM
1542 bool "Enable PC-Speaker support" if EXPERT
1543 depends on HAVE_PCSPKR_PLATFORM
1547 This option allows to disable the internal PC-Speaker
1548 support, saving some memory.
1552 bool "Enable full-sized data structures for core" if EXPERT
1554 Disabling this option reduces the size of miscellaneous core
1555 kernel data structures. This saves memory on small machines,
1556 but may reduce performance.
1559 bool "Enable futex support" if EXPERT
1563 Disabling this option will cause the kernel to be built without
1564 support for "fast userspace mutexes". The resulting kernel may not
1565 run glibc-based applications correctly.
1567 config HAVE_FUTEX_CMPXCHG
1571 Architectures should select this if futex_atomic_cmpxchg_inatomic()
1572 is implemented and always working. This removes a couple of runtime
1576 bool "Enable eventpoll support" if EXPERT
1580 Disabling this option will cause the kernel to be built without
1581 support for epoll family of system calls.
1584 bool "Enable signalfd() system call" if EXPERT
1588 Enable the signalfd() system call that allows to receive signals
1589 on a file descriptor.
1594 bool "Enable timerfd() system call" if EXPERT
1598 Enable the timerfd() system call that allows to receive timer
1599 events on a file descriptor.
1604 bool "Enable eventfd() system call" if EXPERT
1608 Enable the eventfd() system call that allows to receive both
1609 kernel notification (ie. KAIO) or userspace notifications.
1613 # syscall, maps, verifier
1615 bool "Enable bpf() system call"
1620 Enable the bpf() system call that allows to manipulate eBPF
1621 programs and maps via file descriptors.
1624 bool "Use full shmem filesystem" if EXPERT
1628 The shmem is an internal filesystem used to manage shared memory.
1629 It is backed by swap and manages resource limits. It is also exported
1630 to userspace as tmpfs if TMPFS is enabled. Disabling this
1631 option replaces shmem and tmpfs with the much simpler ramfs code,
1632 which may be appropriate on small systems without swap.
1635 bool "Enable AIO support" if EXPERT
1638 This option enables POSIX asynchronous I/O which may by used
1639 by some high performance threaded applications. Disabling
1640 this option saves about 7k.
1642 config ADVISE_SYSCALLS
1643 bool "Enable madvise/fadvise syscalls" if EXPERT
1646 This option enables the madvise and fadvise syscalls, used by
1647 applications to advise the kernel about their future memory or file
1648 usage, improving performance. If building an embedded system where no
1649 applications use these syscalls, you can disable this option to save
1653 bool "Enable userfaultfd() system call"
1657 Enable the userfaultfd() system call that allows to intercept and
1658 handle page faults in userland.
1662 bool "Enable PCI quirk workarounds" if EXPERT
1665 This enables workarounds for various PCI chipset
1666 bugs/quirks. Disable this only if your target machine is
1667 unaffected by PCI quirks.
1670 bool "Enable membarrier() system call" if EXPERT
1673 Enable the membarrier() system call that allows issuing memory
1674 barriers across all running threads, which can be used to distribute
1675 the cost of user-space memory barriers asymmetrically by transforming
1676 pairs of memory barriers into pairs consisting of membarrier() and a
1682 bool "Embedded system"
1683 option allnoconfig_y
1686 This option should be enabled if compiling the kernel for
1687 an embedded system so certain expert options are available
1690 config HAVE_PERF_EVENTS
1693 See tools/perf/design.txt for details.
1695 config PERF_USE_VMALLOC
1698 See tools/perf/design.txt for details
1700 menu "Kernel Performance Events And Counters"
1703 bool "Kernel performance events and counters"
1704 default y if PROFILING
1705 depends on HAVE_PERF_EVENTS
1710 Enable kernel support for various performance events provided
1711 by software and hardware.
1713 Software events are supported either built-in or via the
1714 use of generic tracepoints.
1716 Most modern CPUs support performance events via performance
1717 counter registers. These registers count the number of certain
1718 types of hw events: such as instructions executed, cachemisses
1719 suffered, or branches mis-predicted - without slowing down the
1720 kernel or applications. These registers can also trigger interrupts
1721 when a threshold number of events have passed - and can thus be
1722 used to profile the code that runs on that CPU.
1724 The Linux Performance Event subsystem provides an abstraction of
1725 these software and hardware event capabilities, available via a
1726 system call and used by the "perf" utility in tools/perf/. It
1727 provides per task and per CPU counters, and it provides event
1728 capabilities on top of those.
1732 config DEBUG_PERF_USE_VMALLOC
1734 bool "Debug: use vmalloc to back perf mmap() buffers"
1735 depends on PERF_EVENTS && DEBUG_KERNEL && !PPC
1736 select PERF_USE_VMALLOC
1738 Use vmalloc memory to back perf mmap() buffers.
1740 Mostly useful for debugging the vmalloc code on platforms
1741 that don't require it.
1747 config VM_EVENT_COUNTERS
1749 bool "Enable VM event counters for /proc/vmstat" if EXPERT
1751 VM event counters are needed for event counts to be shown.
1752 This option allows the disabling of the VM event counters
1753 on EXPERT systems. /proc/vmstat will only show page counts
1754 if VM event counters are disabled.
1758 bool "Enable SLUB debugging support" if EXPERT
1759 depends on SLUB && SYSFS
1761 SLUB has extensive debug support features. Disabling these can
1762 result in significant savings in code size. This also disables
1763 SLUB sysfs support. /sys/slab will not exist and there will be
1764 no support for cache validation etc.
1767 bool "Disable heap randomization"
1770 Randomizing heap placement makes heap exploits harder, but it
1771 also breaks ancient binaries (including anything libc5 based).
1772 This option changes the bootup default to heap randomization
1773 disabled, and can be overridden at runtime by setting
1774 /proc/sys/kernel/randomize_va_space to 2.
1776 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1779 prompt "Choose SLAB allocator"
1782 This option allows to select a slab allocator.
1786 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1788 The regular slab allocator that is established and known to work
1789 well in all environments. It organizes cache hot objects in
1790 per cpu and per node queues.
1793 bool "SLUB (Unqueued Allocator)"
1794 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1796 SLUB is a slab allocator that minimizes cache line usage
1797 instead of managing queues of cached objects (SLAB approach).
1798 Per cpu caching is realized using slabs of objects instead
1799 of queues of objects. SLUB can use memory efficiently
1800 and has enhanced diagnostics. SLUB is the default choice for
1805 bool "SLOB (Simple Allocator)"
1807 SLOB replaces the stock allocator with a drastically simpler
1808 allocator. SLOB is generally more space efficient but
1809 does not perform as well on large systems.
1813 config SLUB_CPU_PARTIAL
1815 depends on SLUB && SMP
1816 bool "SLUB per cpu partial cache"
1818 Per cpu partial caches accellerate objects allocation and freeing
1819 that is local to a processor at the price of more indeterminism
1820 in the latency of the free. On overflow these caches will be cleared
1821 which requires the taking of locks that may cause latency spikes.
1822 Typically one would choose no for a realtime system.
1824 config MMAP_ALLOW_UNINITIALIZED
1825 bool "Allow mmapped anonymous memory to be uninitialized"
1826 depends on EXPERT && !MMU
1829 Normally, and according to the Linux spec, anonymous memory obtained
1830 from mmap() has it's contents cleared before it is passed to
1831 userspace. Enabling this config option allows you to request that
1832 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
1833 providing a huge performance boost. If this option is not enabled,
1834 then the flag will be ignored.
1836 This is taken advantage of by uClibc's malloc(), and also by
1837 ELF-FDPIC binfmt's brk and stack allocator.
1839 Because of the obvious security issues, this option should only be
1840 enabled on embedded devices where you control what is run in
1841 userspace. Since that isn't generally a problem on no-MMU systems,
1842 it is normally safe to say Y here.
1844 See Documentation/nommu-mmap.txt for more information.
1846 config SYSTEM_DATA_VERIFICATION
1848 select SYSTEM_TRUSTED_KEYRING
1851 select ASYMMETRIC_KEY_TYPE
1852 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE
1853 select PUBLIC_KEY_ALGO_RSA
1856 select X509_CERTIFICATE_PARSER
1857 select PKCS7_MESSAGE_PARSER
1859 Provide PKCS#7 message verification using the contents of the system
1860 trusted keyring to provide public keys. This then can be used for
1861 module verification, kexec image verification and firmware blob
1865 bool "Profiling support"
1867 Say Y here to enable the extended profiling support mechanisms used
1868 by profilers such as OProfile.
1871 # Place an empty function call at each tracepoint site. Can be
1872 # dynamically changed for a probe function.
1877 source "arch/Kconfig"
1879 endmenu # General setup
1881 config HAVE_GENERIC_DMA_COHERENT
1888 depends on SLAB || SLUB_DEBUG
1896 default 0 if BASE_FULL
1897 default 1 if !BASE_FULL
1900 bool "Enable loadable module support"
1903 Kernel modules are small pieces of compiled code which can
1904 be inserted in the running kernel, rather than being
1905 permanently built into the kernel. You use the "modprobe"
1906 tool to add (and sometimes remove) them. If you say Y here,
1907 many parts of the kernel can be built as modules (by
1908 answering M instead of Y where indicated): this is most
1909 useful for infrequently used options which are not required
1910 for booting. For more information, see the man pages for
1911 modprobe, lsmod, modinfo, insmod and rmmod.
1913 If you say Y here, you will need to run "make
1914 modules_install" to put the modules under /lib/modules/
1915 where modprobe can find them (you may need to be root to do
1922 config MODULE_FORCE_LOAD
1923 bool "Forced module loading"
1926 Allow loading of modules without version information (ie. modprobe
1927 --force). Forced module loading sets the 'F' (forced) taint flag and
1928 is usually a really bad idea.
1930 config MODULE_UNLOAD
1931 bool "Module unloading"
1933 Without this option you will not be able to unload any
1934 modules (note that some modules may not be unloadable
1935 anyway), which makes your kernel smaller, faster
1936 and simpler. If unsure, say Y.
1938 config MODULE_FORCE_UNLOAD
1939 bool "Forced module unloading"
1940 depends on MODULE_UNLOAD
1942 This option allows you to force a module to unload, even if the
1943 kernel believes it is unsafe: the kernel will remove the module
1944 without waiting for anyone to stop using it (using the -f option to
1945 rmmod). This is mainly for kernel developers and desperate users.
1949 bool "Module versioning support"
1951 Usually, you have to use modules compiled with your kernel.
1952 Saying Y here makes it sometimes possible to use modules
1953 compiled for different kernels, by adding enough information
1954 to the modules to (hopefully) spot any changes which would
1955 make them incompatible with the kernel you are running. If
1958 config MODULE_SRCVERSION_ALL
1959 bool "Source checksum for all modules"
1961 Modules which contain a MODULE_VERSION get an extra "srcversion"
1962 field inserted into their modinfo section, which contains a
1963 sum of the source files which made it. This helps maintainers
1964 see exactly which source was used to build a module (since
1965 others sometimes change the module source without updating
1966 the version). With this option, such a "srcversion" field
1967 will be created for all modules. If unsure, say N.
1970 bool "Module signature verification"
1972 select SYSTEM_DATA_VERIFICATION
1974 Check modules for valid signatures upon load: the signature
1975 is simply appended to the module. For more information see
1976 Documentation/module-signing.txt.
1978 Note that this option adds the OpenSSL development packages as a
1979 kernel build dependency so that the signing tool can use its crypto
1982 !!!WARNING!!! If you enable this option, you MUST make sure that the
1983 module DOES NOT get stripped after being signed. This includes the
1984 debuginfo strip done by some packagers (such as rpmbuild) and
1985 inclusion into an initramfs that wants the module size reduced.
1987 config MODULE_SIG_FORCE
1988 bool "Require modules to be validly signed"
1989 depends on MODULE_SIG
1991 Reject unsigned modules or signed modules for which we don't have a
1992 key. Without this, such modules will simply taint the kernel.
1994 config MODULE_SIG_ALL
1995 bool "Automatically sign all modules"
1997 depends on MODULE_SIG
1999 Sign all modules during make modules_install. Without this option,
2000 modules must be signed manually, using the scripts/sign-file tool.
2002 comment "Do not forget to sign required modules with scripts/sign-file"
2003 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL
2006 prompt "Which hash algorithm should modules be signed with?"
2007 depends on MODULE_SIG
2009 This determines which sort of hashing algorithm will be used during
2010 signature generation. This algorithm _must_ be built into the kernel
2011 directly so that signature verification can take place. It is not
2012 possible to load a signed module containing the algorithm to check
2013 the signature on that module.
2015 config MODULE_SIG_SHA1
2016 bool "Sign modules with SHA-1"
2019 config MODULE_SIG_SHA224
2020 bool "Sign modules with SHA-224"
2021 select CRYPTO_SHA256
2023 config MODULE_SIG_SHA256
2024 bool "Sign modules with SHA-256"
2025 select CRYPTO_SHA256
2027 config MODULE_SIG_SHA384
2028 bool "Sign modules with SHA-384"
2029 select CRYPTO_SHA512
2031 config MODULE_SIG_SHA512
2032 bool "Sign modules with SHA-512"
2033 select CRYPTO_SHA512
2037 config MODULE_SIG_HASH
2039 depends on MODULE_SIG
2040 default "sha1" if MODULE_SIG_SHA1
2041 default "sha224" if MODULE_SIG_SHA224
2042 default "sha256" if MODULE_SIG_SHA256
2043 default "sha384" if MODULE_SIG_SHA384
2044 default "sha512" if MODULE_SIG_SHA512
2046 config MODULE_COMPRESS
2047 bool "Compress modules on installation"
2051 Compresses kernel modules when 'make modules_install' is run; gzip or
2052 xz depending on "Compression algorithm" below.
2054 module-init-tools MAY support gzip, and kmod MAY support gzip and xz.
2056 Out-of-tree kernel modules installed using Kbuild will also be
2057 compressed upon installation.
2059 Note: for modules inside an initrd or initramfs, it's more efficient
2060 to compress the whole initrd or initramfs instead.
2062 Note: This is fully compatible with signed modules.
2067 prompt "Compression algorithm"
2068 depends on MODULE_COMPRESS
2069 default MODULE_COMPRESS_GZIP
2071 This determines which sort of compression will be used during
2072 'make modules_install'.
2074 GZIP (default) and XZ are supported.
2076 config MODULE_COMPRESS_GZIP
2079 config MODULE_COMPRESS_XZ
2086 config MODULES_TREE_LOOKUP
2088 depends on PERF_EVENTS || TRACING
2090 config INIT_ALL_POSSIBLE
2093 Back when each arch used to define their own cpu_online_mask and
2094 cpu_possible_mask, some of them chose to initialize cpu_possible_mask
2095 with all 1s, and others with all 0s. When they were centralised,
2096 it was better to provide this option than to break all the archs
2097 and have several arch maintainers pursuing me down dark alleys.
2099 source "block/Kconfig"
2101 config PREEMPT_NOTIFIERS
2108 # Can be selected by architectures with broken toolchains
2109 # that get confused by correct const<->read_only section
2111 config BROKEN_RODATA
2117 Build a simple ASN.1 grammar compiler that produces a bytecode output
2118 that can be interpreted by the ASN.1 stream decoder and used to
2119 inform it as to what tags are to be expected in a stream and what
2120 functions to call on what tags.
2122 source "kernel/Kconfig.locks"