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
29 config THREAD_INFO_IN_TASK
32 Select this to move thread_info off the stack into task_struct. To
33 make this work, an arch will need to remove all thread_info fields
34 except flags and fix any runtime bugs.
36 One subtle change that will be needed is to use try_get_task_stack()
37 and put_task_stack() in save_thread_stack_tsk() and get_wchan().
46 depends on BROKEN || !SMP
49 config INIT_ENV_ARG_LIMIT
54 Maximum of each of the number of arguments and environment
55 variables passed to init from the kernel command line.
59 string "Cross-compiler tool prefix"
61 Same as running 'make CROSS_COMPILE=prefix-' but stored for
62 default make runs in this kernel build directory. You don't
63 need to set this unless you want the configured kernel build
64 directory to select the cross-compiler automatically.
67 bool "Compile also drivers which will not load"
71 Some drivers can be compiled on a different platform than they are
72 intended to be run on. Despite they cannot be loaded there (or even
73 when they load they cannot be used due to missing HW support),
74 developers still, opposing to distributors, might want to build such
75 drivers to compile-test them.
77 If you are a developer and want to build everything available, say Y
78 here. If you are a user/distributor, say N here to exclude useless
79 drivers to be distributed.
82 string "Local version - append to kernel release"
84 Append an extra string to the end of your kernel version.
85 This will show up when you type uname, for example.
86 The string you set here will be appended after the contents of
87 any files with a filename matching localversion* in your
88 object and source tree, in that order. Your total string can
89 be a maximum of 64 characters.
91 config LOCALVERSION_AUTO
92 bool "Automatically append version information to the version string"
94 depends on !COMPILE_TEST
96 This will try to automatically determine if the current tree is a
97 release tree by looking for git tags that belong to the current
100 A string of the format -gxxxxxxxx will be added to the localversion
101 if a git-based tree is found. The string generated by this will be
102 appended after any matching localversion* files, and after the value
103 set in CONFIG_LOCALVERSION.
105 (The actual string used here is the first eight characters produced
106 by running the command:
108 $ git rev-parse --verify HEAD
110 which is done within the script "scripts/setlocalversion".)
112 config HAVE_KERNEL_GZIP
115 config HAVE_KERNEL_BZIP2
118 config HAVE_KERNEL_LZMA
121 config HAVE_KERNEL_XZ
124 config HAVE_KERNEL_LZO
127 config HAVE_KERNEL_LZ4
131 prompt "Kernel compression mode"
133 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4
135 The linux kernel is a kind of self-extracting executable.
136 Several compression algorithms are available, which differ
137 in efficiency, compression and decompression speed.
138 Compression speed is only relevant when building a kernel.
139 Decompression speed is relevant at each boot.
141 If you have any problems with bzip2 or lzma compressed
142 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
143 version of this functionality (bzip2 only), for 2.4, was
144 supplied by Christian Ludwig)
146 High compression options are mostly useful for users, who
147 are low on disk space (embedded systems), but for whom ram
150 If in doubt, select 'gzip'
154 depends on HAVE_KERNEL_GZIP
156 The old and tried gzip compression. It provides a good balance
157 between compression ratio and decompression speed.
161 depends on HAVE_KERNEL_BZIP2
163 Its compression ratio and speed is intermediate.
164 Decompression speed is slowest among the choices. The kernel
165 size is about 10% smaller with bzip2, in comparison to gzip.
166 Bzip2 uses a large amount of memory. For modern kernels you
167 will need at least 8MB RAM or more for booting.
171 depends on HAVE_KERNEL_LZMA
173 This compression algorithm's ratio is best. Decompression speed
174 is between gzip and bzip2. Compression is slowest.
175 The kernel size is about 33% smaller with LZMA in comparison to gzip.
179 depends on HAVE_KERNEL_XZ
181 XZ uses the LZMA2 algorithm and instruction set specific
182 BCJ filters which can improve compression ratio of executable
183 code. The size of the kernel is about 30% smaller with XZ in
184 comparison to gzip. On architectures for which there is a BCJ
185 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ
186 will create a few percent smaller kernel than plain LZMA.
188 The speed is about the same as with LZMA: The decompression
189 speed of XZ is better than that of bzip2 but worse than gzip
190 and LZO. Compression is slow.
194 depends on HAVE_KERNEL_LZO
196 Its compression ratio is the poorest among the choices. The kernel
197 size is about 10% bigger than gzip; however its speed
198 (both compression and decompression) is the fastest.
202 depends on HAVE_KERNEL_LZ4
204 LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding.
205 A preliminary version of LZ4 de/compression tool is available at
206 <https://code.google.com/p/lz4/>.
208 Its compression ratio is worse than LZO. The size of the kernel
209 is about 8% bigger than LZO. But the decompression speed is
214 config DEFAULT_HOSTNAME
215 string "Default hostname"
218 This option determines the default system hostname before userspace
219 calls sethostname(2). The kernel traditionally uses "(none)" here,
220 but you may wish to use a different default here to make a minimal
221 system more usable with less configuration.
223 config VERSION_SIGNATURE
224 string "Arbitrary version signature"
226 This string will be created in a file, /proc/version_signature. It
227 is useful in determining arbitrary data about your kernel. For instance,
228 if you have several kernels of the same version, but need to keep track
229 of a revision of the same kernel, but not affect it's ability to load
230 compatible modules, this is the easiest way to do that.
233 bool "Support for paging of anonymous memory (swap)"
234 depends on MMU && BLOCK
237 This option allows you to choose whether you want to have support
238 for so called swap devices or swap files in your kernel that are
239 used to provide more virtual memory than the actual RAM present
240 in your computer. If unsure say Y.
245 Inter Process Communication is a suite of library functions and
246 system calls which let processes (running programs) synchronize and
247 exchange information. It is generally considered to be a good thing,
248 and some programs won't run unless you say Y here. In particular, if
249 you want to run the DOS emulator dosemu under Linux (read the
250 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
251 you'll need to say Y here.
253 You can find documentation about IPC with "info ipc" and also in
254 section 6.4 of the Linux Programmer's Guide, available from
255 <http://www.tldp.org/guides.html>.
257 config SYSVIPC_SYSCTL
264 bool "POSIX Message Queues"
267 POSIX variant of message queues is a part of IPC. In POSIX message
268 queues every message has a priority which decides about succession
269 of receiving it by a process. If you want to compile and run
270 programs written e.g. for Solaris with use of its POSIX message
271 queues (functions mq_*) say Y here.
273 POSIX message queues are visible as a filesystem called 'mqueue'
274 and can be mounted somewhere if you want to do filesystem
275 operations on message queues.
279 config POSIX_MQUEUE_SYSCTL
281 depends on POSIX_MQUEUE
285 config CROSS_MEMORY_ATTACH
286 bool "Enable process_vm_readv/writev syscalls"
290 Enabling this option adds the system calls process_vm_readv and
291 process_vm_writev which allow a process with the correct privileges
292 to directly read from or write to another process' address space.
293 See the man page for more details.
296 bool "open by fhandle syscalls" if EXPERT
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 "uselib syscall"
310 def_bool ALPHA || M68K || SPARC || X86_32 || IA32_EMULATION
312 This option enables the uselib syscall, a system call used in the
313 dynamic linker from libc5 and earlier. glibc does not use this
314 system call. If you intend to run programs built on libc5 or
315 earlier, you may need to enable this syscall. Current systems
316 running glibc can safely disable this.
319 bool "Auditing support"
322 Enable auditing infrastructure that can be used with another
323 kernel subsystem, such as SELinux (which requires this for
324 logging of avc messages output). System call auditing is included
325 on architectures which support it.
327 config HAVE_ARCH_AUDITSYSCALL
332 depends on AUDIT && HAVE_ARCH_AUDITSYSCALL
336 depends on AUDITSYSCALL
341 depends on AUDITSYSCALL
344 source "kernel/irq/Kconfig"
345 source "kernel/time/Kconfig"
347 menu "CPU/Task time and stats accounting"
349 config VIRT_CPU_ACCOUNTING
353 prompt "Cputime accounting"
354 default TICK_CPU_ACCOUNTING if !PPC64
355 default VIRT_CPU_ACCOUNTING_NATIVE if PPC64
357 # Kind of a stub config for the pure tick based cputime accounting
358 config TICK_CPU_ACCOUNTING
359 bool "Simple tick based cputime accounting"
360 depends on !S390 && !NO_HZ_FULL
362 This is the basic tick based cputime accounting that maintains
363 statistics about user, system and idle time spent on per jiffies
368 config VIRT_CPU_ACCOUNTING_NATIVE
369 bool "Deterministic task and CPU time accounting"
370 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL
371 select VIRT_CPU_ACCOUNTING
373 Select this option to enable more accurate task and CPU time
374 accounting. This is done by reading a CPU counter on each
375 kernel entry and exit and on transitions within the kernel
376 between system, softirq and hardirq state, so there is a
377 small performance impact. In the case of s390 or IBM POWER > 5,
378 this also enables accounting of stolen time on logically-partitioned
381 config VIRT_CPU_ACCOUNTING_GEN
382 bool "Full dynticks CPU time accounting"
383 depends on HAVE_CONTEXT_TRACKING
384 depends on HAVE_VIRT_CPU_ACCOUNTING_GEN
385 select VIRT_CPU_ACCOUNTING
386 select CONTEXT_TRACKING
388 Select this option to enable task and CPU time accounting on full
389 dynticks systems. This accounting is implemented by watching every
390 kernel-user boundaries using the context tracking subsystem.
391 The accounting is thus performed at the expense of some significant
394 For now this is only useful if you are working on the full
395 dynticks subsystem development.
401 config IRQ_TIME_ACCOUNTING
402 bool "Fine granularity task level IRQ time accounting"
403 depends on HAVE_IRQ_TIME_ACCOUNTING && !VIRT_CPU_ACCOUNTING_NATIVE
405 Select this option to enable fine granularity task irq time
406 accounting. This is done by reading a timestamp on each
407 transitions between softirq and hardirq state, so there can be a
408 small performance impact.
410 If in doubt, say N here.
412 config BSD_PROCESS_ACCT
413 bool "BSD Process Accounting"
416 If you say Y here, a user level program will be able to instruct the
417 kernel (via a special system call) to write process accounting
418 information to a file: whenever a process exits, information about
419 that process will be appended to the file by the kernel. The
420 information includes things such as creation time, owning user,
421 command name, memory usage, controlling terminal etc. (the complete
422 list is in the struct acct in <file:include/linux/acct.h>). It is
423 up to the user level program to do useful things with this
424 information. This is generally a good idea, so say Y.
426 config BSD_PROCESS_ACCT_V3
427 bool "BSD Process Accounting version 3 file format"
428 depends on BSD_PROCESS_ACCT
431 If you say Y here, the process accounting information is written
432 in a new file format that also logs the process IDs of each
433 process and it's parent. Note that this file format is incompatible
434 with previous v0/v1/v2 file formats, so you will need updated tools
435 for processing it. A preliminary version of these tools is available
436 at <http://www.gnu.org/software/acct/>.
439 bool "Export task/process statistics through netlink"
444 Export selected statistics for tasks/processes through the
445 generic netlink interface. Unlike BSD process accounting, the
446 statistics are available during the lifetime of tasks/processes as
447 responses to commands. Like BSD accounting, they are sent to user
452 config TASK_DELAY_ACCT
453 bool "Enable per-task delay accounting"
457 Collect information on time spent by a task waiting for system
458 resources like cpu, synchronous block I/O completion and swapping
459 in pages. Such statistics can help in setting a task's priorities
460 relative to other tasks for cpu, io, rss limits etc.
465 bool "Enable extended accounting over taskstats"
468 Collect extended task accounting data and send the data
469 to userland for processing over the taskstats interface.
473 config TASK_IO_ACCOUNTING
474 bool "Enable per-task storage I/O accounting"
475 depends on TASK_XACCT
477 Collect information on the number of bytes of storage I/O which this
482 endmenu # "CPU/Task time and stats accounting"
488 default y if !PREEMPT && SMP
490 This option selects the RCU implementation that is
491 designed for very large SMP system with hundreds or
492 thousands of CPUs. It also scales down nicely to
499 This option selects the RCU implementation that is
500 designed for very large SMP systems with hundreds or
501 thousands of CPUs, but for which real-time response
502 is also required. It also scales down nicely to
505 Select this option if you are unsure.
509 default y if !PREEMPT && !SMP
511 This option selects the RCU implementation that is
512 designed for UP systems from which real-time response
513 is not required. This option greatly reduces the
514 memory footprint of RCU.
517 bool "Make expert-level adjustments to RCU configuration"
520 This option needs to be enabled if you wish to make
521 expert-level adjustments to RCU configuration. By default,
522 no such adjustments can be made, which has the often-beneficial
523 side-effect of preventing "make oldconfig" from asking you all
524 sorts of detailed questions about how you would like numerous
525 obscure RCU options to be set up.
527 Say Y if you need to make expert-level adjustments to RCU.
529 Say N if you are unsure.
534 This option selects the sleepable version of RCU. This version
535 permits arbitrary sleeping or blocking within RCU read-side critical
544 This option enables a task-based RCU implementation that uses
545 only voluntary context switch (not preemption!), idle, and
546 user-mode execution as quiescent states.
548 config RCU_STALL_COMMON
549 def_bool ( TREE_RCU || PREEMPT_RCU || RCU_TRACE )
551 This option enables RCU CPU stall code that is common between
552 the TINY and TREE variants of RCU. The purpose is to allow
553 the tiny variants to disable RCU CPU stall warnings, while
554 making these warnings mandatory for the tree variants.
556 config CONTEXT_TRACKING
559 config CONTEXT_TRACKING_FORCE
560 bool "Force context tracking"
561 depends on CONTEXT_TRACKING
562 default y if !NO_HZ_FULL
564 The major pre-requirement for full dynticks to work is to
565 support the context tracking subsystem. But there are also
566 other dependencies to provide in order to make the full
569 This option stands for testing when an arch implements the
570 context tracking backend but doesn't yet fullfill all the
571 requirements to make the full dynticks feature working.
572 Without the full dynticks, there is no way to test the support
573 for context tracking and the subsystems that rely on it: RCU
574 userspace extended quiescent state and tickless cputime
575 accounting. This option copes with the absence of the full
576 dynticks subsystem by forcing the context tracking on all
579 Say Y only if you're working on the development of an
580 architecture backend for the context tracking.
582 Say N otherwise, this option brings an overhead that you
583 don't want in production.
587 int "Tree-based hierarchical RCU fanout value"
590 depends on (TREE_RCU || PREEMPT_RCU) && RCU_EXPERT
594 This option controls the fanout of hierarchical implementations
595 of RCU, allowing RCU to work efficiently on machines with
596 large numbers of CPUs. This value must be at least the fourth
597 root of NR_CPUS, which allows NR_CPUS to be insanely large.
598 The default value of RCU_FANOUT should be used for production
599 systems, but if you are stress-testing the RCU implementation
600 itself, small RCU_FANOUT values allow you to test large-system
601 code paths on small(er) systems.
603 Select a specific number if testing RCU itself.
604 Take the default if unsure.
606 config RCU_FANOUT_LEAF
607 int "Tree-based hierarchical RCU leaf-level fanout value"
610 depends on (TREE_RCU || PREEMPT_RCU) && RCU_EXPERT
613 This option controls the leaf-level fanout of hierarchical
614 implementations of RCU, and allows trading off cache misses
615 against lock contention. Systems that synchronize their
616 scheduling-clock interrupts for energy-efficiency reasons will
617 want the default because the smaller leaf-level fanout keeps
618 lock contention levels acceptably low. Very large systems
619 (hundreds or thousands of CPUs) will instead want to set this
620 value to the maximum value possible in order to reduce the
621 number of cache misses incurred during RCU's grace-period
622 initialization. These systems tend to run CPU-bound, and thus
623 are not helped by synchronized interrupts, and thus tend to
624 skew them, which reduces lock contention enough that large
625 leaf-level fanouts work well.
627 Select a specific number if testing RCU itself.
629 Select the maximum permissible value for large systems.
631 Take the default if unsure.
633 config RCU_FAST_NO_HZ
634 bool "Accelerate last non-dyntick-idle CPU's grace periods"
635 depends on NO_HZ_COMMON && SMP && RCU_EXPERT
638 This option permits CPUs to enter dynticks-idle state even if
639 they have RCU callbacks queued, and prevents RCU from waking
640 these CPUs up more than roughly once every four jiffies (by
641 default, you can adjust this using the rcutree.rcu_idle_gp_delay
642 parameter), thus improving energy efficiency. On the other
643 hand, this option increases the duration of RCU grace periods,
644 for example, slowing down synchronize_rcu().
646 Say Y if energy efficiency is critically important, and you
647 don't care about increased grace-period durations.
649 Say N if you are unsure.
651 config TREE_RCU_TRACE
652 def_bool RCU_TRACE && ( TREE_RCU || PREEMPT_RCU )
655 This option provides tracing for the TREE_RCU and
656 PREEMPT_RCU implementations, permitting Makefile to
657 trivially select kernel/rcutree_trace.c.
660 bool "Enable RCU priority boosting"
661 depends on RT_MUTEXES && PREEMPT_RCU && RCU_EXPERT
664 This option boosts the priority of preempted RCU readers that
665 block the current preemptible RCU grace period for too long.
666 This option also prevents heavy loads from blocking RCU
667 callback invocation for all flavors of RCU.
669 Say Y here if you are working with real-time apps or heavy loads
670 Say N here if you are unsure.
672 config RCU_KTHREAD_PRIO
673 int "Real-time priority to use for RCU worker threads"
674 range 1 99 if RCU_BOOST
675 range 0 99 if !RCU_BOOST
676 default 1 if RCU_BOOST
677 default 0 if !RCU_BOOST
678 depends on RCU_EXPERT
680 This option specifies the SCHED_FIFO priority value that will be
681 assigned to the rcuc/n and rcub/n threads and is also the value
682 used for RCU_BOOST (if enabled). If you are working with a
683 real-time application that has one or more CPU-bound threads
684 running at a real-time priority level, you should set
685 RCU_KTHREAD_PRIO to a priority higher than the highest-priority
686 real-time CPU-bound application thread. The default RCU_KTHREAD_PRIO
687 value of 1 is appropriate in the common case, which is real-time
688 applications that do not have any CPU-bound threads.
690 Some real-time applications might not have a single real-time
691 thread that saturates a given CPU, but instead might have
692 multiple real-time threads that, taken together, fully utilize
693 that CPU. In this case, you should set RCU_KTHREAD_PRIO to
694 a priority higher than the lowest-priority thread that is
695 conspiring to prevent the CPU from running any non-real-time
696 tasks. For example, if one thread at priority 10 and another
697 thread at priority 5 are between themselves fully consuming
698 the CPU time on a given CPU, then RCU_KTHREAD_PRIO should be
699 set to priority 6 or higher.
701 Specify the real-time priority, or take the default if unsure.
703 config RCU_BOOST_DELAY
704 int "Milliseconds to delay boosting after RCU grace-period start"
709 This option specifies the time to wait after the beginning of
710 a given grace period before priority-boosting preempted RCU
711 readers blocking that grace period. Note that any RCU reader
712 blocking an expedited RCU grace period is boosted immediately.
714 Accept the default if unsure.
717 bool "Offload RCU callback processing from boot-selected CPUs"
718 depends on TREE_RCU || PREEMPT_RCU
719 depends on RCU_EXPERT || NO_HZ_FULL
722 Use this option to reduce OS jitter for aggressive HPC or
723 real-time workloads. It can also be used to offload RCU
724 callback invocation to energy-efficient CPUs in battery-powered
725 asymmetric multiprocessors.
727 This option offloads callback invocation from the set of
728 CPUs specified at boot time by the rcu_nocbs parameter.
729 For each such CPU, a kthread ("rcuox/N") will be created to
730 invoke callbacks, where the "N" is the CPU being offloaded,
731 and where the "x" is "b" for RCU-bh, "p" for RCU-preempt, and
732 "s" for RCU-sched. Nothing prevents this kthread from running
733 on the specified CPUs, but (1) the kthreads may be preempted
734 between each callback, and (2) affinity or cgroups can be used
735 to force the kthreads to run on whatever set of CPUs is desired.
737 Say Y here if you want to help to debug reduced OS jitter.
738 Say N here if you are unsure.
741 prompt "Build-forced no-CBs CPUs"
742 default RCU_NOCB_CPU_NONE
743 depends on RCU_NOCB_CPU
745 This option allows no-CBs CPUs (whose RCU callbacks are invoked
746 from kthreads rather than from softirq context) to be specified
747 at build time. Additional no-CBs CPUs may be specified by
748 the rcu_nocbs= boot parameter.
750 config RCU_NOCB_CPU_NONE
751 bool "No build_forced no-CBs CPUs"
753 This option does not force any of the CPUs to be no-CBs CPUs.
754 Only CPUs designated by the rcu_nocbs= boot parameter will be
755 no-CBs CPUs, whose RCU callbacks will be invoked by per-CPU
756 kthreads whose names begin with "rcuo". All other CPUs will
757 invoke their own RCU callbacks in softirq context.
759 Select this option if you want to choose no-CBs CPUs at
760 boot time, for example, to allow testing of different no-CBs
761 configurations without having to rebuild the kernel each time.
763 config RCU_NOCB_CPU_ZERO
764 bool "CPU 0 is a build_forced no-CBs CPU"
766 This option forces CPU 0 to be a no-CBs CPU, so that its RCU
767 callbacks are invoked by a per-CPU kthread whose name begins
768 with "rcuo". Additional CPUs may be designated as no-CBs
769 CPUs using the rcu_nocbs= boot parameter will be no-CBs CPUs.
770 All other CPUs will invoke their own RCU callbacks in softirq
773 Select this if CPU 0 needs to be a no-CBs CPU for real-time
774 or energy-efficiency reasons, but the real reason it exists
775 is to ensure that randconfig testing covers mixed systems.
777 config RCU_NOCB_CPU_ALL
778 bool "All CPUs are build_forced no-CBs CPUs"
780 This option forces all CPUs to be no-CBs CPUs. The rcu_nocbs=
781 boot parameter will be ignored. All CPUs' RCU callbacks will
782 be executed in the context of per-CPU rcuo kthreads created for
783 this purpose. Assuming that the kthreads whose names start with
784 "rcuo" are bound to "housekeeping" CPUs, this reduces OS jitter
785 on the remaining CPUs, but might decrease memory locality during
786 RCU-callback invocation, thus potentially degrading throughput.
788 Select this if all CPUs need to be no-CBs CPUs for real-time
789 or energy-efficiency reasons.
793 config RCU_EXPEDITE_BOOT
797 This option enables expedited grace periods at boot time,
798 as if rcu_expedite_gp() had been invoked early in boot.
799 The corresponding rcu_unexpedite_gp() is invoked from
800 rcu_end_inkernel_boot(), which is intended to be invoked
801 at the end of the kernel-only boot sequence, just before
804 Accept the default if unsure.
806 endmenu # "RCU Subsystem"
813 tristate "Kernel .config support"
816 This option enables the complete Linux kernel ".config" file
817 contents to be saved in the kernel. It provides documentation
818 of which kernel options are used in a running kernel or in an
819 on-disk kernel. This information can be extracted from the kernel
820 image file with the script scripts/extract-ikconfig and used as
821 input to rebuild the current kernel or to build another kernel.
822 It can also be extracted from a running kernel by reading
823 /proc/config.gz if enabled (below).
826 bool "Enable access to .config through /proc/config.gz"
827 depends on IKCONFIG && PROC_FS
829 This option enables access to the kernel configuration file
830 through /proc/config.gz.
833 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
838 Select the minimal kernel log buffer size as a power of 2.
839 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config
840 parameter, see below. Any higher size also might be forced
841 by "log_buf_len" boot parameter.
851 config LOG_CPU_MAX_BUF_SHIFT
852 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)"
855 default 12 if !BASE_SMALL
856 default 0 if BASE_SMALL
859 This option allows to increase the default ring buffer size
860 according to the number of CPUs. The value defines the contribution
861 of each CPU as a power of 2. The used space is typically only few
862 lines however it might be much more when problems are reported,
865 The increased size means that a new buffer has to be allocated and
866 the original static one is unused. It makes sense only on systems
867 with more CPUs. Therefore this value is used only when the sum of
868 contributions is greater than the half of the default kernel ring
869 buffer as defined by LOG_BUF_SHIFT. The default values are set
870 so that more than 64 CPUs are needed to trigger the allocation.
872 Also this option is ignored when "log_buf_len" kernel parameter is
873 used as it forces an exact (power of two) size of the ring buffer.
875 The number of possible CPUs is used for this computation ignoring
876 hotplugging making the computation optimal for the worst case
877 scenario while allowing a simple algorithm to be used from bootup.
879 Examples shift values and their meaning:
880 17 => 128 KB for each CPU
881 16 => 64 KB for each CPU
882 15 => 32 KB for each CPU
883 14 => 16 KB for each CPU
884 13 => 8 KB for each CPU
885 12 => 4 KB for each CPU
887 config NMI_LOG_BUF_SHIFT
888 int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)"
891 depends on PRINTK_NMI
893 Select the size of a per-CPU buffer where NMI messages are temporary
894 stored. They are copied to the main log buffer in a safe context
895 to avoid a deadlock. The value defines the size as a power of 2.
897 NMI messages are rare and limited. The largest one is when
898 a backtrace is printed. It usually fits into 4KB. Select
899 8KB if you want to be on the safe side.
902 17 => 128 KB for each CPU
903 16 => 64 KB for each CPU
904 15 => 32 KB for each CPU
905 14 => 16 KB for each CPU
906 13 => 8 KB for each CPU
907 12 => 4 KB for each CPU
910 # Architectures with an unreliable sched_clock() should select this:
912 config HAVE_UNSTABLE_SCHED_CLOCK
915 config GENERIC_SCHED_CLOCK
919 # For architectures that want to enable the support for NUMA-affine scheduler
922 config ARCH_SUPPORTS_NUMA_BALANCING
926 # For architectures that prefer to flush all TLBs after a number of pages
927 # are unmapped instead of sending one IPI per page to flush. The architecture
928 # must provide guarantees on what happens if a clean TLB cache entry is
929 # written after the unmap. Details are in mm/rmap.c near the check for
930 # should_defer_flush. The architecture should also consider if the full flush
931 # and the refill costs are offset by the savings of sending fewer IPIs.
932 config ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
936 # For architectures that know their GCC __int128 support is sound
938 config ARCH_SUPPORTS_INT128
941 # For architectures that (ab)use NUMA to represent different memory regions
942 # all cpu-local but of different latencies, such as SuperH.
944 config ARCH_WANT_NUMA_VARIABLE_LOCALITY
947 config NUMA_BALANCING
948 bool "Memory placement aware NUMA scheduler"
949 depends on ARCH_SUPPORTS_NUMA_BALANCING
950 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
951 depends on SMP && NUMA && MIGRATION
953 This option adds support for automatic NUMA aware memory/task placement.
954 The mechanism is quite primitive and is based on migrating memory when
955 it has references to the node the task is running on.
957 This system will be inactive on UMA systems.
959 config NUMA_BALANCING_DEFAULT_ENABLED
960 bool "Automatically enable NUMA aware memory/task placement"
962 depends on NUMA_BALANCING
964 If set, automatic NUMA balancing will be enabled if running on a NUMA
968 bool "Control Group support"
971 This option adds support for grouping sets of processes together, for
972 use with process control subsystems such as Cpusets, CFS, memory
973 controls or device isolation.
975 - Documentation/scheduler/sched-design-CFS.txt (CFS)
976 - Documentation/cgroup-v1/ (features for grouping, isolation
977 and resource control)
987 bool "Memory controller"
991 Provides control over the memory footprint of tasks in a cgroup.
994 bool "Swap controller"
995 depends on MEMCG && SWAP
997 Provides control over the swap space consumed by tasks in a cgroup.
999 config MEMCG_SWAP_ENABLED
1000 bool "Swap controller enabled by default"
1001 depends on MEMCG_SWAP
1004 Memory Resource Controller Swap Extension comes with its price in
1005 a bigger memory consumption. General purpose distribution kernels
1006 which want to enable the feature but keep it disabled by default
1007 and let the user enable it by swapaccount=1 boot command line
1008 parameter should have this option unselected.
1009 For those who want to have the feature enabled by default should
1010 select this option (if, for some reason, they need to disable it
1011 then swapaccount=0 does the trick).
1014 bool "IO controller"
1018 Generic block IO controller cgroup interface. This is the common
1019 cgroup interface which should be used by various IO controlling
1022 Currently, CFQ IO scheduler uses it to recognize task groups and
1023 control disk bandwidth allocation (proportional time slice allocation)
1024 to such task groups. It is also used by bio throttling logic in
1025 block layer to implement upper limit in IO rates on a device.
1027 This option only enables generic Block IO controller infrastructure.
1028 One needs to also enable actual IO controlling logic/policy. For
1029 enabling proportional weight division of disk bandwidth in CFQ, set
1030 CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set
1031 CONFIG_BLK_DEV_THROTTLING=y.
1033 See Documentation/cgroup-v1/blkio-controller.txt for more information.
1035 config DEBUG_BLK_CGROUP
1036 bool "IO controller debugging"
1037 depends on BLK_CGROUP
1040 Enable some debugging help. Currently it exports additional stat
1041 files in a cgroup which can be useful for debugging.
1043 config CGROUP_WRITEBACK
1045 depends on MEMCG && BLK_CGROUP
1048 menuconfig CGROUP_SCHED
1049 bool "CPU controller"
1052 This feature lets CPU scheduler recognize task groups and control CPU
1053 bandwidth allocation to such task groups. It uses cgroups to group
1057 config FAIR_GROUP_SCHED
1058 bool "Group scheduling for SCHED_OTHER"
1059 depends on CGROUP_SCHED
1060 default CGROUP_SCHED
1062 config CFS_BANDWIDTH
1063 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED"
1064 depends on FAIR_GROUP_SCHED
1067 This option allows users to define CPU bandwidth rates (limits) for
1068 tasks running within the fair group scheduler. Groups with no limit
1069 set are considered to be unconstrained and will run with no
1071 See tip/Documentation/scheduler/sched-bwc.txt for more information.
1073 config RT_GROUP_SCHED
1074 bool "Group scheduling for SCHED_RR/FIFO"
1075 depends on CGROUP_SCHED
1078 This feature lets you explicitly allocate real CPU bandwidth
1079 to task groups. If enabled, it will also make it impossible to
1080 schedule realtime tasks for non-root users until you allocate
1081 realtime bandwidth for them.
1082 See Documentation/scheduler/sched-rt-group.txt for more information.
1087 bool "PIDs controller"
1089 Provides enforcement of process number limits in the scope of a
1090 cgroup. Any attempt to fork more processes than is allowed in the
1091 cgroup will fail. PIDs are fundamentally a global resource because it
1092 is fairly trivial to reach PID exhaustion before you reach even a
1093 conservative kmemcg limit. As a result, it is possible to grind a
1094 system to halt without being limited by other cgroup policies. The
1095 PIDs controller is designed to stop this from happening.
1097 It should be noted that organisational operations (such as attaching
1098 to a cgroup hierarchy will *not* be blocked by the PIDs controller),
1099 since the PIDs limit only affects a process's ability to fork, not to
1102 config CGROUP_FREEZER
1103 bool "Freezer controller"
1105 Provides a way to freeze and unfreeze all tasks in a
1108 This option affects the ORIGINAL cgroup interface. The cgroup2 memory
1109 controller includes important in-kernel memory consumers per default.
1111 If you're using cgroup2, say N.
1113 config CGROUP_HUGETLB
1114 bool "HugeTLB controller"
1115 depends on HUGETLB_PAGE
1119 Provides a cgroup controller for HugeTLB pages.
1120 When you enable this, you can put a per cgroup limit on HugeTLB usage.
1121 The limit is enforced during page fault. Since HugeTLB doesn't
1122 support page reclaim, enforcing the limit at page fault time implies
1123 that, the application will get SIGBUS signal if it tries to access
1124 HugeTLB pages beyond its limit. This requires the application to know
1125 beforehand how much HugeTLB pages it would require for its use. The
1126 control group is tracked in the third page lru pointer. This means
1127 that we cannot use the controller with huge page less than 3 pages.
1130 bool "Cpuset controller"
1132 This option will let you create and manage CPUSETs which
1133 allow dynamically partitioning a system into sets of CPUs and
1134 Memory Nodes and assigning tasks to run only within those sets.
1135 This is primarily useful on large SMP or NUMA systems.
1139 config PROC_PID_CPUSET
1140 bool "Include legacy /proc/<pid>/cpuset file"
1144 config CGROUP_DEVICE
1145 bool "Device controller"
1147 Provides a cgroup controller implementing whitelists for
1148 devices which a process in the cgroup can mknod or open.
1150 config CGROUP_CPUACCT
1151 bool "Simple CPU accounting controller"
1153 Provides a simple controller for monitoring the
1154 total CPU consumed by the tasks in a cgroup.
1157 bool "Perf controller"
1158 depends on PERF_EVENTS
1160 This option extends the perf per-cpu mode to restrict monitoring
1161 to threads which belong to the cgroup specified and run on the
1167 bool "Support for eBPF programs attached to cgroups"
1168 depends on BPF_SYSCALL
1169 select SOCK_CGROUP_DATA
1171 Allow attaching eBPF programs to a cgroup using the bpf(2)
1172 syscall command BPF_PROG_ATTACH.
1174 In which context these programs are accessed depends on the type
1175 of attachment. For instance, programs that are attached using
1176 BPF_CGROUP_INET_INGRESS will be executed on the ingress path of
1180 bool "Example controller"
1183 This option enables a simple controller that exports
1184 debugging information about the cgroups framework.
1188 config SOCK_CGROUP_DATA
1194 config CHECKPOINT_RESTORE
1195 bool "Checkpoint/restore support" if EXPERT
1196 select PROC_CHILDREN
1199 Enables additional kernel features in a sake of checkpoint/restore.
1200 In particular it adds auxiliary prctl codes to setup process text,
1201 data and heap segment sizes, and a few additional /proc filesystem
1204 If unsure, say N here.
1206 menuconfig NAMESPACES
1207 bool "Namespaces support" if EXPERT
1208 depends on MULTIUSER
1211 Provides the way to make tasks work with different objects using
1212 the same id. For example same IPC id may refer to different objects
1213 or same user id or pid may refer to different tasks when used in
1214 different namespaces.
1219 bool "UTS namespace"
1222 In this namespace tasks see different info provided with the
1226 bool "IPC namespace"
1227 depends on (SYSVIPC || POSIX_MQUEUE)
1230 In this namespace tasks work with IPC ids which correspond to
1231 different IPC objects in different namespaces.
1234 bool "User namespace"
1237 This allows containers, i.e. vservers, to use user namespaces
1238 to provide different user info for different servers.
1240 When user namespaces are enabled in the kernel it is
1241 recommended that the MEMCG option also be enabled and that
1242 user-space use the memory control groups to limit the amount
1243 of memory a memory unprivileged users can use.
1248 bool "PID Namespaces"
1251 Support process id namespaces. This allows having multiple
1252 processes with the same pid as long as they are in different
1253 pid namespaces. This is a building block of containers.
1256 bool "Network namespace"
1260 Allow user space to create what appear to be multiple instances
1261 of the network stack.
1265 config SCHED_AUTOGROUP
1266 bool "Automatic process group scheduling"
1269 select FAIR_GROUP_SCHED
1271 This option optimizes the scheduler for common desktop workloads by
1272 automatically creating and populating task groups. This separation
1273 of workloads isolates aggressive CPU burners (like build jobs) from
1274 desktop applications. Task group autogeneration is currently based
1277 config SYSFS_DEPRECATED
1278 bool "Enable deprecated sysfs features to support old userspace tools"
1282 This option adds code that switches the layout of the "block" class
1283 devices, to not show up in /sys/class/block/, but only in
1286 This switch is only active when the sysfs.deprecated=1 boot option is
1287 passed or the SYSFS_DEPRECATED_V2 option is set.
1289 This option allows new kernels to run on old distributions and tools,
1290 which might get confused by /sys/class/block/. Since 2007/2008 all
1291 major distributions and tools handle this just fine.
1293 Recent distributions and userspace tools after 2009/2010 depend on
1294 the existence of /sys/class/block/, and will not work with this
1297 Only if you are using a new kernel on an old distribution, you might
1300 config SYSFS_DEPRECATED_V2
1301 bool "Enable deprecated sysfs features by default"
1304 depends on SYSFS_DEPRECATED
1306 Enable deprecated sysfs by default.
1308 See the CONFIG_SYSFS_DEPRECATED option for more details about this
1311 Only if you are using a new kernel on an old distribution, you might
1312 need to say Y here. Even then, odds are you would not need it
1313 enabled, you can always pass the boot option if absolutely necessary.
1316 bool "Kernel->user space relay support (formerly relayfs)"
1319 This option enables support for relay interface support in
1320 certain file systems (such as debugfs).
1321 It is designed to provide an efficient mechanism for tools and
1322 facilities to relay large amounts of data from kernel space to
1327 config BLK_DEV_INITRD
1328 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
1329 depends on BROKEN || !FRV
1331 The initial RAM filesystem is a ramfs which is loaded by the
1332 boot loader (loadlin or lilo) and that is mounted as root
1333 before the normal boot procedure. It is typically used to
1334 load modules needed to mount the "real" root file system,
1335 etc. See <file:Documentation/admin-guide/initrd.rst> for details.
1337 If RAM disk support (BLK_DEV_RAM) is also included, this
1338 also enables initial RAM disk (initrd) support and adds
1339 15 Kbytes (more on some other architectures) to the kernel size.
1345 source "usr/Kconfig"
1350 prompt "Compiler optimization level"
1351 default CONFIG_CC_OPTIMIZE_FOR_PERFORMANCE
1353 config CC_OPTIMIZE_FOR_PERFORMANCE
1354 bool "Optimize for performance"
1356 This is the default optimization level for the kernel, building
1357 with the "-O2" compiler flag for best performance and most
1358 helpful compile-time warnings.
1360 config CC_OPTIMIZE_FOR_SIZE
1361 bool "Optimize for size"
1363 Enabling this option will pass "-Os" instead of "-O2" to
1364 your compiler resulting in a smaller kernel.
1379 config SYSCTL_EXCEPTION_TRACE
1382 Enable support for /proc/sys/debug/exception-trace.
1384 config SYSCTL_ARCH_UNALIGN_NO_WARN
1387 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap
1388 Allows arch to define/use @no_unaligned_warning to possibly warn
1389 about unaligned access emulation going on under the hood.
1391 config SYSCTL_ARCH_UNALIGN_ALLOW
1394 Enable support for /proc/sys/kernel/unaligned-trap
1395 Allows arches to define/use @unaligned_enabled to runtime toggle
1396 the unaligned access emulation.
1397 see arch/parisc/kernel/unaligned.c for reference
1399 config HAVE_PCSPKR_PLATFORM
1402 # interpreter that classic socket filters depend on
1407 bool "Configure standard kernel features (expert users)"
1408 # Unhide debug options, to make the on-by-default options visible
1411 This option allows certain base kernel options and settings
1412 to be disabled or tweaked. This is for specialized
1413 environments which can tolerate a "non-standard" kernel.
1414 Only use this if you really know what you are doing.
1417 bool "Enable 16-bit UID system calls" if EXPERT
1418 depends on HAVE_UID16 && MULTIUSER
1421 This enables the legacy 16-bit UID syscall wrappers.
1424 bool "Multiple users, groups and capabilities support" if EXPERT
1427 This option enables support for non-root users, groups and
1430 If you say N here, all processes will run with UID 0, GID 0, and all
1431 possible capabilities. Saying N here also compiles out support for
1432 system calls related to UIDs, GIDs, and capabilities, such as setuid,
1435 If unsure, say Y here.
1437 config SGETMASK_SYSCALL
1438 bool "sgetmask/ssetmask syscalls support" if EXPERT
1439 def_bool PARISC || MN10300 || BLACKFIN || M68K || PPC || MIPS || X86 || SPARC || CRIS || MICROBLAZE || SUPERH
1441 sys_sgetmask and sys_ssetmask are obsolete system calls
1442 no longer supported in libc but still enabled by default in some
1445 If unsure, leave the default option here.
1447 config SYSFS_SYSCALL
1448 bool "Sysfs syscall support" if EXPERT
1451 sys_sysfs is an obsolete system call no longer supported in libc.
1452 Note that disabling this option is more secure but might break
1453 compatibility with some systems.
1455 If unsure say Y here.
1457 config SYSCTL_SYSCALL
1458 bool "Sysctl syscall support" if EXPERT
1459 depends on PROC_SYSCTL
1463 sys_sysctl uses binary paths that have been found challenging
1464 to properly maintain and use. The interface in /proc/sys
1465 using paths with ascii names is now the primary path to this
1468 Almost nothing using the binary sysctl interface so if you are
1469 trying to save some space it is probably safe to disable this,
1470 making your kernel marginally smaller.
1472 If unsure say N here.
1475 bool "Posix Clocks & timers" if EXPERT
1478 This includes native support for POSIX timers to the kernel.
1479 Some embedded systems have no use for them and therefore they
1480 can be configured out to reduce the size of the kernel image.
1482 When this option is disabled, the following syscalls won't be
1483 available: timer_create, timer_gettime: timer_getoverrun,
1484 timer_settime, timer_delete, clock_adjtime, getitimer,
1485 setitimer, alarm. Furthermore, the clock_settime, clock_gettime,
1486 clock_getres and clock_nanosleep syscalls will be limited to
1487 CLOCK_REALTIME, CLOCK_MONOTONIC and CLOCK_BOOTTIME only.
1492 bool "Load all symbols for debugging/ksymoops" if EXPERT
1495 Say Y here to let the kernel print out symbolic crash information and
1496 symbolic stack backtraces. This increases the size of the kernel
1497 somewhat, as all symbols have to be loaded into the kernel image.
1500 bool "Include all symbols in kallsyms"
1501 depends on DEBUG_KERNEL && KALLSYMS
1503 Normally kallsyms only contains the symbols of functions for nicer
1504 OOPS messages and backtraces (i.e., symbols from the text and inittext
1505 sections). This is sufficient for most cases. And only in very rare
1506 cases (e.g., when a debugger is used) all symbols are required (e.g.,
1507 names of variables from the data sections, etc).
1509 This option makes sure that all symbols are loaded into the kernel
1510 image (i.e., symbols from all sections) in cost of increased kernel
1511 size (depending on the kernel configuration, it may be 300KiB or
1512 something like this).
1514 Say N unless you really need all symbols.
1516 config KALLSYMS_ABSOLUTE_PERCPU
1519 default X86_64 && SMP
1521 config KALLSYMS_BASE_RELATIVE
1524 default !IA64 && !(TILE && 64BIT)
1526 Instead of emitting them as absolute values in the native word size,
1527 emit the symbol references in the kallsyms table as 32-bit entries,
1528 each containing a relative value in the range [base, base + U32_MAX]
1529 or, when KALLSYMS_ABSOLUTE_PERCPU is in effect, each containing either
1530 an absolute value in the range [0, S32_MAX] or a relative value in the
1531 range [base, base + S32_MAX], where base is the lowest relative symbol
1532 address encountered in the image.
1534 On 64-bit builds, this reduces the size of the address table by 50%,
1535 but more importantly, it results in entries whose values are build
1536 time constants, and no relocation pass is required at runtime to fix
1537 up the entries based on the runtime load address of the kernel.
1541 bool "Enable support for printk" if EXPERT
1544 This option enables normal printk support. Removing it
1545 eliminates most of the message strings from the kernel image
1546 and makes the kernel more or less silent. As this makes it
1547 very difficult to diagnose system problems, saying N here is
1548 strongly discouraged.
1556 bool "BUG() support" if EXPERT
1559 Disabling this option eliminates support for BUG and WARN, reducing
1560 the size of your kernel image and potentially quietly ignoring
1561 numerous fatal conditions. You should only consider disabling this
1562 option for embedded systems with no facilities for reporting errors.
1568 bool "Enable ELF core dumps" if EXPERT
1570 Enable support for generating core dumps. Disabling saves about 4k.
1573 config PCSPKR_PLATFORM
1574 bool "Enable PC-Speaker support" if EXPERT
1575 depends on HAVE_PCSPKR_PLATFORM
1579 This option allows to disable the internal PC-Speaker
1580 support, saving some memory.
1584 bool "Enable full-sized data structures for core" if EXPERT
1586 Disabling this option reduces the size of miscellaneous core
1587 kernel data structures. This saves memory on small machines,
1588 but may reduce performance.
1591 bool "Enable futex support" if EXPERT
1595 Disabling this option will cause the kernel to be built without
1596 support for "fast userspace mutexes". The resulting kernel may not
1597 run glibc-based applications correctly.
1599 config HAVE_FUTEX_CMPXCHG
1603 Architectures should select this if futex_atomic_cmpxchg_inatomic()
1604 is implemented and always working. This removes a couple of runtime
1608 bool "Enable eventpoll support" if EXPERT
1612 Disabling this option will cause the kernel to be built without
1613 support for epoll family of system calls.
1616 bool "Enable signalfd() system call" if EXPERT
1620 Enable the signalfd() system call that allows to receive signals
1621 on a file descriptor.
1626 bool "Enable timerfd() system call" if EXPERT
1630 Enable the timerfd() system call that allows to receive timer
1631 events on a file descriptor.
1636 bool "Enable eventfd() system call" if EXPERT
1640 Enable the eventfd() system call that allows to receive both
1641 kernel notification (ie. KAIO) or userspace notifications.
1645 # syscall, maps, verifier
1647 bool "Enable bpf() system call"
1652 Enable the bpf() system call that allows to manipulate eBPF
1653 programs and maps via file descriptors.
1656 bool "Use full shmem filesystem" if EXPERT
1660 The shmem is an internal filesystem used to manage shared memory.
1661 It is backed by swap and manages resource limits. It is also exported
1662 to userspace as tmpfs if TMPFS is enabled. Disabling this
1663 option replaces shmem and tmpfs with the much simpler ramfs code,
1664 which may be appropriate on small systems without swap.
1667 bool "Enable AIO support" if EXPERT
1670 This option enables POSIX asynchronous I/O which may by used
1671 by some high performance threaded applications. Disabling
1672 this option saves about 7k.
1674 config ADVISE_SYSCALLS
1675 bool "Enable madvise/fadvise syscalls" if EXPERT
1678 This option enables the madvise and fadvise syscalls, used by
1679 applications to advise the kernel about their future memory or file
1680 usage, improving performance. If building an embedded system where no
1681 applications use these syscalls, you can disable this option to save
1685 bool "Enable userfaultfd() system call"
1689 Enable the userfaultfd() system call that allows to intercept and
1690 handle page faults in userland.
1694 bool "Enable PCI quirk workarounds" if EXPERT
1697 This enables workarounds for various PCI chipset
1698 bugs/quirks. Disable this only if your target machine is
1699 unaffected by PCI quirks.
1702 bool "Enable membarrier() system call" if EXPERT
1705 Enable the membarrier() system call that allows issuing memory
1706 barriers across all running threads, which can be used to distribute
1707 the cost of user-space memory barriers asymmetrically by transforming
1708 pairs of memory barriers into pairs consisting of membarrier() and a
1714 bool "Embedded system"
1715 option allnoconfig_y
1718 This option should be enabled if compiling the kernel for
1719 an embedded system so certain expert options are available
1722 config HAVE_PERF_EVENTS
1725 See tools/perf/design.txt for details.
1727 config PERF_USE_VMALLOC
1730 See tools/perf/design.txt for details
1732 menu "Kernel Performance Events And Counters"
1735 bool "Kernel performance events and counters"
1736 default y if PROFILING
1737 depends on HAVE_PERF_EVENTS
1742 Enable kernel support for various performance events provided
1743 by software and hardware.
1745 Software events are supported either built-in or via the
1746 use of generic tracepoints.
1748 Most modern CPUs support performance events via performance
1749 counter registers. These registers count the number of certain
1750 types of hw events: such as instructions executed, cachemisses
1751 suffered, or branches mis-predicted - without slowing down the
1752 kernel or applications. These registers can also trigger interrupts
1753 when a threshold number of events have passed - and can thus be
1754 used to profile the code that runs on that CPU.
1756 The Linux Performance Event subsystem provides an abstraction of
1757 these software and hardware event capabilities, available via a
1758 system call and used by the "perf" utility in tools/perf/. It
1759 provides per task and per CPU counters, and it provides event
1760 capabilities on top of those.
1764 config DEBUG_PERF_USE_VMALLOC
1766 bool "Debug: use vmalloc to back perf mmap() buffers"
1767 depends on PERF_EVENTS && DEBUG_KERNEL && !PPC
1768 select PERF_USE_VMALLOC
1770 Use vmalloc memory to back perf mmap() buffers.
1772 Mostly useful for debugging the vmalloc code on platforms
1773 that don't require it.
1779 config VM_EVENT_COUNTERS
1781 bool "Enable VM event counters for /proc/vmstat" if EXPERT
1783 VM event counters are needed for event counts to be shown.
1784 This option allows the disabling of the VM event counters
1785 on EXPERT systems. /proc/vmstat will only show page counts
1786 if VM event counters are disabled.
1790 bool "Enable SLUB debugging support" if EXPERT
1791 depends on SLUB && SYSFS
1793 SLUB has extensive debug support features. Disabling these can
1794 result in significant savings in code size. This also disables
1795 SLUB sysfs support. /sys/slab will not exist and there will be
1796 no support for cache validation etc.
1799 bool "Disable heap randomization"
1802 Randomizing heap placement makes heap exploits harder, but it
1803 also breaks ancient binaries (including anything libc5 based).
1804 This option changes the bootup default to heap randomization
1805 disabled, and can be overridden at runtime by setting
1806 /proc/sys/kernel/randomize_va_space to 2.
1808 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1811 prompt "Choose SLAB allocator"
1814 This option allows to select a slab allocator.
1818 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1820 The regular slab allocator that is established and known to work
1821 well in all environments. It organizes cache hot objects in
1822 per cpu and per node queues.
1825 bool "SLUB (Unqueued Allocator)"
1826 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1828 SLUB is a slab allocator that minimizes cache line usage
1829 instead of managing queues of cached objects (SLAB approach).
1830 Per cpu caching is realized using slabs of objects instead
1831 of queues of objects. SLUB can use memory efficiently
1832 and has enhanced diagnostics. SLUB is the default choice for
1837 bool "SLOB (Simple Allocator)"
1839 SLOB replaces the stock allocator with a drastically simpler
1840 allocator. SLOB is generally more space efficient but
1841 does not perform as well on large systems.
1845 config SLAB_FREELIST_RANDOM
1847 depends on SLAB || SLUB
1848 bool "SLAB freelist randomization"
1850 Randomizes the freelist order used on creating new pages. This
1851 security feature reduces the predictability of the kernel slab
1852 allocator against heap overflows.
1854 config SLUB_CPU_PARTIAL
1856 depends on SLUB && SMP
1857 bool "SLUB per cpu partial cache"
1859 Per cpu partial caches accellerate objects allocation and freeing
1860 that is local to a processor at the price of more indeterminism
1861 in the latency of the free. On overflow these caches will be cleared
1862 which requires the taking of locks that may cause latency spikes.
1863 Typically one would choose no for a realtime system.
1865 config MMAP_ALLOW_UNINITIALIZED
1866 bool "Allow mmapped anonymous memory to be uninitialized"
1867 depends on EXPERT && !MMU
1870 Normally, and according to the Linux spec, anonymous memory obtained
1871 from mmap() has it's contents cleared before it is passed to
1872 userspace. Enabling this config option allows you to request that
1873 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
1874 providing a huge performance boost. If this option is not enabled,
1875 then the flag will be ignored.
1877 This is taken advantage of by uClibc's malloc(), and also by
1878 ELF-FDPIC binfmt's brk and stack allocator.
1880 Because of the obvious security issues, this option should only be
1881 enabled on embedded devices where you control what is run in
1882 userspace. Since that isn't generally a problem on no-MMU systems,
1883 it is normally safe to say Y here.
1885 See Documentation/nommu-mmap.txt for more information.
1887 config SYSTEM_DATA_VERIFICATION
1889 select SYSTEM_TRUSTED_KEYRING
1893 select ASYMMETRIC_KEY_TYPE
1894 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE
1897 select X509_CERTIFICATE_PARSER
1898 select PKCS7_MESSAGE_PARSER
1900 Provide PKCS#7 message verification using the contents of the system
1901 trusted keyring to provide public keys. This then can be used for
1902 module verification, kexec image verification and firmware blob
1905 config SYSTEM_BLACKLIST_KEYRING
1906 bool "Provide system-wide ring of blacklisted keys"
1909 Provide a system keyring to which blacklisted keys can be added.
1910 Keys in the keyring are considered entirely untrusted. Keys in this
1911 keyring are used by the module signature checking to reject loading
1912 of modules signed with a blacklisted key.
1915 bool "Profiling support"
1917 Say Y here to enable the extended profiling support mechanisms used
1918 by profilers such as OProfile.
1921 # Place an empty function call at each tracepoint site. Can be
1922 # dynamically changed for a probe function.
1927 source "arch/Kconfig"
1929 endmenu # General setup
1931 config HAVE_GENERIC_DMA_COHERENT
1938 depends on SLAB || SLUB_DEBUG
1946 default 0 if BASE_FULL
1947 default 1 if !BASE_FULL
1950 bool "Enable loadable module support"
1953 Kernel modules are small pieces of compiled code which can
1954 be inserted in the running kernel, rather than being
1955 permanently built into the kernel. You use the "modprobe"
1956 tool to add (and sometimes remove) them. If you say Y here,
1957 many parts of the kernel can be built as modules (by
1958 answering M instead of Y where indicated): this is most
1959 useful for infrequently used options which are not required
1960 for booting. For more information, see the man pages for
1961 modprobe, lsmod, modinfo, insmod and rmmod.
1963 If you say Y here, you will need to run "make
1964 modules_install" to put the modules under /lib/modules/
1965 where modprobe can find them (you may need to be root to do
1972 config MODULE_FORCE_LOAD
1973 bool "Forced module loading"
1976 Allow loading of modules without version information (ie. modprobe
1977 --force). Forced module loading sets the 'F' (forced) taint flag and
1978 is usually a really bad idea.
1980 config MODULE_UNLOAD
1981 bool "Module unloading"
1983 Without this option you will not be able to unload any
1984 modules (note that some modules may not be unloadable
1985 anyway), which makes your kernel smaller, faster
1986 and simpler. If unsure, say Y.
1988 config MODULE_FORCE_UNLOAD
1989 bool "Forced module unloading"
1990 depends on MODULE_UNLOAD
1992 This option allows you to force a module to unload, even if the
1993 kernel believes it is unsafe: the kernel will remove the module
1994 without waiting for anyone to stop using it (using the -f option to
1995 rmmod). This is mainly for kernel developers and desperate users.
1999 bool "Module versioning support"
2001 Usually, you have to use modules compiled with your kernel.
2002 Saying Y here makes it sometimes possible to use modules
2003 compiled for different kernels, by adding enough information
2004 to the modules to (hopefully) spot any changes which would
2005 make them incompatible with the kernel you are running. If
2008 config MODULE_REL_CRCS
2010 depends on MODVERSIONS
2012 config MODULE_SRCVERSION_ALL
2013 bool "Source checksum for all modules"
2015 Modules which contain a MODULE_VERSION get an extra "srcversion"
2016 field inserted into their modinfo section, which contains a
2017 sum of the source files which made it. This helps maintainers
2018 see exactly which source was used to build a module (since
2019 others sometimes change the module source without updating
2020 the version). With this option, such a "srcversion" field
2021 will be created for all modules. If unsure, say N.
2024 bool "Module signature verification"
2026 select SYSTEM_DATA_VERIFICATION
2028 Check modules for valid signatures upon load: the signature
2029 is simply appended to the module. For more information see
2030 Documentation/module-signing.txt.
2032 Note that this option adds the OpenSSL development packages as a
2033 kernel build dependency so that the signing tool can use its crypto
2036 !!!WARNING!!! If you enable this option, you MUST make sure that the
2037 module DOES NOT get stripped after being signed. This includes the
2038 debuginfo strip done by some packagers (such as rpmbuild) and
2039 inclusion into an initramfs that wants the module size reduced.
2041 config MODULE_SIG_FORCE
2042 bool "Require modules to be validly signed"
2043 depends on MODULE_SIG
2045 Reject unsigned modules or signed modules for which we don't have a
2046 key. Without this, such modules will simply taint the kernel.
2048 config MODULE_SIG_ALL
2049 bool "Automatically sign all modules"
2051 depends on MODULE_SIG
2053 Sign all modules during make modules_install. Without this option,
2054 modules must be signed manually, using the scripts/sign-file tool.
2056 comment "Do not forget to sign required modules with scripts/sign-file"
2057 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL
2059 config MODULE_SIG_UEFI
2060 bool "Allow modules signed with certs stored in UEFI"
2061 depends on MODULE_SIG && SYSTEM_BLACKLIST_KEYRING && EFI
2062 select EFI_SIGNATURE_LIST_PARSER
2064 This will import certificates stored in UEFI and allow modules
2065 signed with those to be loaded. It will also disallow loading
2066 of modules stored in the UEFI dbx variable.
2069 prompt "Which hash algorithm should modules be signed with?"
2070 depends on MODULE_SIG
2072 This determines which sort of hashing algorithm will be used during
2073 signature generation. This algorithm _must_ be built into the kernel
2074 directly so that signature verification can take place. It is not
2075 possible to load a signed module containing the algorithm to check
2076 the signature on that module.
2078 config MODULE_SIG_SHA1
2079 bool "Sign modules with SHA-1"
2082 config MODULE_SIG_SHA224
2083 bool "Sign modules with SHA-224"
2084 select CRYPTO_SHA256
2086 config MODULE_SIG_SHA256
2087 bool "Sign modules with SHA-256"
2088 select CRYPTO_SHA256
2090 config MODULE_SIG_SHA384
2091 bool "Sign modules with SHA-384"
2092 select CRYPTO_SHA512
2094 config MODULE_SIG_SHA512
2095 bool "Sign modules with SHA-512"
2096 select CRYPTO_SHA512
2100 config MODULE_SIG_HASH
2102 depends on MODULE_SIG
2103 default "sha1" if MODULE_SIG_SHA1
2104 default "sha224" if MODULE_SIG_SHA224
2105 default "sha256" if MODULE_SIG_SHA256
2106 default "sha384" if MODULE_SIG_SHA384
2107 default "sha512" if MODULE_SIG_SHA512
2109 config MODULE_COMPRESS
2110 bool "Compress modules on installation"
2114 Compresses kernel modules when 'make modules_install' is run; gzip or
2115 xz depending on "Compression algorithm" below.
2117 module-init-tools MAY support gzip, and kmod MAY support gzip and xz.
2119 Out-of-tree kernel modules installed using Kbuild will also be
2120 compressed upon installation.
2122 Note: for modules inside an initrd or initramfs, it's more efficient
2123 to compress the whole initrd or initramfs instead.
2125 Note: This is fully compatible with signed modules.
2130 prompt "Compression algorithm"
2131 depends on MODULE_COMPRESS
2132 default MODULE_COMPRESS_GZIP
2134 This determines which sort of compression will be used during
2135 'make modules_install'.
2137 GZIP (default) and XZ are supported.
2139 config MODULE_COMPRESS_GZIP
2142 config MODULE_COMPRESS_XZ
2147 config TRIM_UNUSED_KSYMS
2148 bool "Trim unused exported kernel symbols"
2149 depends on MODULES && !UNUSED_SYMBOLS
2151 The kernel and some modules make many symbols available for
2152 other modules to use via EXPORT_SYMBOL() and variants. Depending
2153 on the set of modules being selected in your kernel configuration,
2154 many of those exported symbols might never be used.
2156 This option allows for unused exported symbols to be dropped from
2157 the build. In turn, this provides the compiler more opportunities
2158 (especially when using LTO) for optimizing the code and reducing
2159 binary size. This might have some security advantages as well.
2161 If unsure, or if you need to build out-of-tree modules, say N.
2165 config MODULES_TREE_LOOKUP
2167 depends on PERF_EVENTS || TRACING
2169 config INIT_ALL_POSSIBLE
2172 Back when each arch used to define their own cpu_online_mask and
2173 cpu_possible_mask, some of them chose to initialize cpu_possible_mask
2174 with all 1s, and others with all 0s. When they were centralised,
2175 it was better to provide this option than to break all the archs
2176 and have several arch maintainers pursuing me down dark alleys.
2178 source "block/Kconfig"
2180 config PREEMPT_NOTIFIERS
2190 Build a simple ASN.1 grammar compiler that produces a bytecode output
2191 that can be interpreted by the ASN.1 stream decoder and used to
2192 inform it as to what tags are to be expected in a stream and what
2193 functions to call on what tags.
2195 source "kernel/Kconfig.locks"
projects / mirror_ubuntu-zesty-kernel.git / blob