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.
224 bool "Support for paging of anonymous memory (swap)"
225 depends on MMU && BLOCK
228 This option allows you to choose whether you want to have support
229 for so called swap devices or swap files in your kernel that are
230 used to provide more virtual memory than the actual RAM present
231 in your computer. If unsure say Y.
236 Inter Process Communication is a suite of library functions and
237 system calls which let processes (running programs) synchronize and
238 exchange information. It is generally considered to be a good thing,
239 and some programs won't run unless you say Y here. In particular, if
240 you want to run the DOS emulator dosemu under Linux (read the
241 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
242 you'll need to say Y here.
244 You can find documentation about IPC with "info ipc" and also in
245 section 6.4 of the Linux Programmer's Guide, available from
246 <http://www.tldp.org/guides.html>.
248 config SYSVIPC_SYSCTL
255 bool "POSIX Message Queues"
258 POSIX variant of message queues is a part of IPC. In POSIX message
259 queues every message has a priority which decides about succession
260 of receiving it by a process. If you want to compile and run
261 programs written e.g. for Solaris with use of its POSIX message
262 queues (functions mq_*) say Y here.
264 POSIX message queues are visible as a filesystem called 'mqueue'
265 and can be mounted somewhere if you want to do filesystem
266 operations on message queues.
270 config POSIX_MQUEUE_SYSCTL
272 depends on POSIX_MQUEUE
276 config CROSS_MEMORY_ATTACH
277 bool "Enable process_vm_readv/writev syscalls"
281 Enabling this option adds the system calls process_vm_readv and
282 process_vm_writev which allow a process with the correct privileges
283 to directly read from or write to another process' address space.
284 See the man page for more details.
287 bool "open by fhandle syscalls" if EXPERT
291 If you say Y here, a user level program will be able to map
292 file names to handle and then later use the handle for
293 different file system operations. This is useful in implementing
294 userspace file servers, which now track files using handles instead
295 of names. The handle would remain the same even if file names
296 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2)
300 bool "uselib syscall"
301 def_bool ALPHA || M68K || SPARC || X86_32 || IA32_EMULATION
303 This option enables the uselib syscall, a system call used in the
304 dynamic linker from libc5 and earlier. glibc does not use this
305 system call. If you intend to run programs built on libc5 or
306 earlier, you may need to enable this syscall. Current systems
307 running glibc can safely disable this.
310 bool "Auditing support"
313 Enable auditing infrastructure that can be used with another
314 kernel subsystem, such as SELinux (which requires this for
315 logging of avc messages output). System call auditing is included
316 on architectures which support it.
318 config HAVE_ARCH_AUDITSYSCALL
323 depends on AUDIT && HAVE_ARCH_AUDITSYSCALL
327 depends on AUDITSYSCALL
332 depends on AUDITSYSCALL
335 source "kernel/irq/Kconfig"
336 source "kernel/time/Kconfig"
338 menu "CPU/Task time and stats accounting"
340 config VIRT_CPU_ACCOUNTING
344 prompt "Cputime accounting"
345 default TICK_CPU_ACCOUNTING if !PPC64
346 default VIRT_CPU_ACCOUNTING_NATIVE if PPC64
348 # Kind of a stub config for the pure tick based cputime accounting
349 config TICK_CPU_ACCOUNTING
350 bool "Simple tick based cputime accounting"
351 depends on !S390 && !NO_HZ_FULL
353 This is the basic tick based cputime accounting that maintains
354 statistics about user, system and idle time spent on per jiffies
359 config VIRT_CPU_ACCOUNTING_NATIVE
360 bool "Deterministic task and CPU time accounting"
361 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL
362 select VIRT_CPU_ACCOUNTING
364 Select this option to enable more accurate task and CPU time
365 accounting. This is done by reading a CPU counter on each
366 kernel entry and exit and on transitions within the kernel
367 between system, softirq and hardirq state, so there is a
368 small performance impact. In the case of s390 or IBM POWER > 5,
369 this also enables accounting of stolen time on logically-partitioned
372 config VIRT_CPU_ACCOUNTING_GEN
373 bool "Full dynticks CPU time accounting"
374 depends on HAVE_CONTEXT_TRACKING
375 depends on HAVE_VIRT_CPU_ACCOUNTING_GEN
376 select VIRT_CPU_ACCOUNTING
377 select CONTEXT_TRACKING
379 Select this option to enable task and CPU time accounting on full
380 dynticks systems. This accounting is implemented by watching every
381 kernel-user boundaries using the context tracking subsystem.
382 The accounting is thus performed at the expense of some significant
385 For now this is only useful if you are working on the full
386 dynticks subsystem development.
392 config IRQ_TIME_ACCOUNTING
393 bool "Fine granularity task level IRQ time accounting"
394 depends on HAVE_IRQ_TIME_ACCOUNTING && !VIRT_CPU_ACCOUNTING_NATIVE
396 Select this option to enable fine granularity task irq time
397 accounting. This is done by reading a timestamp on each
398 transitions between softirq and hardirq state, so there can be a
399 small performance impact.
401 If in doubt, say N here.
403 config BSD_PROCESS_ACCT
404 bool "BSD Process Accounting"
407 If you say Y here, a user level program will be able to instruct the
408 kernel (via a special system call) to write process accounting
409 information to a file: whenever a process exits, information about
410 that process will be appended to the file by the kernel. The
411 information includes things such as creation time, owning user,
412 command name, memory usage, controlling terminal etc. (the complete
413 list is in the struct acct in <file:include/linux/acct.h>). It is
414 up to the user level program to do useful things with this
415 information. This is generally a good idea, so say Y.
417 config BSD_PROCESS_ACCT_V3
418 bool "BSD Process Accounting version 3 file format"
419 depends on BSD_PROCESS_ACCT
422 If you say Y here, the process accounting information is written
423 in a new file format that also logs the process IDs of each
424 process and it's parent. Note that this file format is incompatible
425 with previous v0/v1/v2 file formats, so you will need updated tools
426 for processing it. A preliminary version of these tools is available
427 at <http://www.gnu.org/software/acct/>.
430 bool "Export task/process statistics through netlink"
435 Export selected statistics for tasks/processes through the
436 generic netlink interface. Unlike BSD process accounting, the
437 statistics are available during the lifetime of tasks/processes as
438 responses to commands. Like BSD accounting, they are sent to user
443 config TASK_DELAY_ACCT
444 bool "Enable per-task delay accounting"
448 Collect information on time spent by a task waiting for system
449 resources like cpu, synchronous block I/O completion and swapping
450 in pages. Such statistics can help in setting a task's priorities
451 relative to other tasks for cpu, io, rss limits etc.
456 bool "Enable extended accounting over taskstats"
459 Collect extended task accounting data and send the data
460 to userland for processing over the taskstats interface.
464 config TASK_IO_ACCOUNTING
465 bool "Enable per-task storage I/O accounting"
466 depends on TASK_XACCT
468 Collect information on the number of bytes of storage I/O which this
473 endmenu # "CPU/Task time and stats accounting"
479 default y if !PREEMPT && SMP
481 This option selects the RCU implementation that is
482 designed for very large SMP system with hundreds or
483 thousands of CPUs. It also scales down nicely to
490 This option selects the RCU implementation that is
491 designed for very large SMP systems with hundreds or
492 thousands of CPUs, but for which real-time response
493 is also required. It also scales down nicely to
496 Select this option if you are unsure.
500 default y if !PREEMPT && !SMP
502 This option selects the RCU implementation that is
503 designed for UP systems from which real-time response
504 is not required. This option greatly reduces the
505 memory footprint of RCU.
508 bool "Make expert-level adjustments to RCU configuration"
511 This option needs to be enabled if you wish to make
512 expert-level adjustments to RCU configuration. By default,
513 no such adjustments can be made, which has the often-beneficial
514 side-effect of preventing "make oldconfig" from asking you all
515 sorts of detailed questions about how you would like numerous
516 obscure RCU options to be set up.
518 Say Y if you need to make expert-level adjustments to RCU.
520 Say N if you are unsure.
525 This option selects the sleepable version of RCU. This version
526 permits arbitrary sleeping or blocking within RCU read-side critical
530 bool "Use v4.11 classic SRCU implementation"
532 depends on RCU_EXPERT && SRCU
534 This option selects the traditional well-tested classic SRCU
535 implementation from v4.11, as might be desired for enterprise
536 Linux distributions. Without this option, the shiny new
537 Tiny SRCU and Tree SRCU implementations are used instead.
538 At some point, it is hoped that Tiny SRCU and Tree SRCU
539 will accumulate enough test time and confidence to allow
540 Classic SRCU to be dropped entirely.
542 Say Y if you need a rock-solid SRCU.
544 Say N if you would like help test Tree SRCU.
548 default y if TINY_RCU && !CLASSIC_SRCU
550 This option selects the single-CPU non-preemptible version of SRCU.
554 default y if !TINY_RCU && !CLASSIC_SRCU
556 This option selects the full-fledged version of SRCU.
563 This option enables a task-based RCU implementation that uses
564 only voluntary context switch (not preemption!), idle, and
565 user-mode execution as quiescent states.
567 config RCU_STALL_COMMON
568 def_bool ( TREE_RCU || PREEMPT_RCU || RCU_TRACE )
570 This option enables RCU CPU stall code that is common between
571 the TINY and TREE variants of RCU. The purpose is to allow
572 the tiny variants to disable RCU CPU stall warnings, while
573 making these warnings mandatory for the tree variants.
575 config CONTEXT_TRACKING
578 config CONTEXT_TRACKING_FORCE
579 bool "Force context tracking"
580 depends on CONTEXT_TRACKING
581 default y if !NO_HZ_FULL
583 The major pre-requirement for full dynticks to work is to
584 support the context tracking subsystem. But there are also
585 other dependencies to provide in order to make the full
588 This option stands for testing when an arch implements the
589 context tracking backend but doesn't yet fullfill all the
590 requirements to make the full dynticks feature working.
591 Without the full dynticks, there is no way to test the support
592 for context tracking and the subsystems that rely on it: RCU
593 userspace extended quiescent state and tickless cputime
594 accounting. This option copes with the absence of the full
595 dynticks subsystem by forcing the context tracking on all
598 Say Y only if you're working on the development of an
599 architecture backend for the context tracking.
601 Say N otherwise, this option brings an overhead that you
602 don't want in production.
606 int "Tree-based hierarchical RCU fanout value"
609 depends on (TREE_RCU || PREEMPT_RCU) && RCU_EXPERT
613 This option controls the fanout of hierarchical implementations
614 of RCU, allowing RCU to work efficiently on machines with
615 large numbers of CPUs. This value must be at least the fourth
616 root of NR_CPUS, which allows NR_CPUS to be insanely large.
617 The default value of RCU_FANOUT should be used for production
618 systems, but if you are stress-testing the RCU implementation
619 itself, small RCU_FANOUT values allow you to test large-system
620 code paths on small(er) systems.
622 Select a specific number if testing RCU itself.
623 Take the default if unsure.
625 config RCU_FANOUT_LEAF
626 int "Tree-based hierarchical RCU leaf-level fanout value"
629 depends on (TREE_RCU || PREEMPT_RCU) && RCU_EXPERT
632 This option controls the leaf-level fanout of hierarchical
633 implementations of RCU, and allows trading off cache misses
634 against lock contention. Systems that synchronize their
635 scheduling-clock interrupts for energy-efficiency reasons will
636 want the default because the smaller leaf-level fanout keeps
637 lock contention levels acceptably low. Very large systems
638 (hundreds or thousands of CPUs) will instead want to set this
639 value to the maximum value possible in order to reduce the
640 number of cache misses incurred during RCU's grace-period
641 initialization. These systems tend to run CPU-bound, and thus
642 are not helped by synchronized interrupts, and thus tend to
643 skew them, which reduces lock contention enough that large
644 leaf-level fanouts work well.
646 Select a specific number if testing RCU itself.
648 Select the maximum permissible value for large systems.
650 Take the default if unsure.
652 config RCU_FAST_NO_HZ
653 bool "Accelerate last non-dyntick-idle CPU's grace periods"
654 depends on NO_HZ_COMMON && SMP && RCU_EXPERT
657 This option permits CPUs to enter dynticks-idle state even if
658 they have RCU callbacks queued, and prevents RCU from waking
659 these CPUs up more than roughly once every four jiffies (by
660 default, you can adjust this using the rcutree.rcu_idle_gp_delay
661 parameter), thus improving energy efficiency. On the other
662 hand, this option increases the duration of RCU grace periods,
663 for example, slowing down synchronize_rcu().
665 Say Y if energy efficiency is critically important, and you
666 don't care about increased grace-period durations.
668 Say N if you are unsure.
670 config TREE_RCU_TRACE
671 def_bool RCU_TRACE && ( TREE_RCU || PREEMPT_RCU )
674 This option provides tracing for the TREE_RCU and
675 PREEMPT_RCU implementations, permitting Makefile to
676 trivially select kernel/rcutree_trace.c.
679 bool "Enable RCU priority boosting"
680 depends on RT_MUTEXES && PREEMPT_RCU && RCU_EXPERT
683 This option boosts the priority of preempted RCU readers that
684 block the current preemptible RCU grace period for too long.
685 This option also prevents heavy loads from blocking RCU
686 callback invocation for all flavors of RCU.
688 Say Y here if you are working with real-time apps or heavy loads
689 Say N here if you are unsure.
691 config RCU_KTHREAD_PRIO
692 int "Real-time priority to use for RCU worker threads"
693 range 1 99 if RCU_BOOST
694 range 0 99 if !RCU_BOOST
695 default 1 if RCU_BOOST
696 default 0 if !RCU_BOOST
697 depends on RCU_EXPERT
699 This option specifies the SCHED_FIFO priority value that will be
700 assigned to the rcuc/n and rcub/n threads and is also the value
701 used for RCU_BOOST (if enabled). If you are working with a
702 real-time application that has one or more CPU-bound threads
703 running at a real-time priority level, you should set
704 RCU_KTHREAD_PRIO to a priority higher than the highest-priority
705 real-time CPU-bound application thread. The default RCU_KTHREAD_PRIO
706 value of 1 is appropriate in the common case, which is real-time
707 applications that do not have any CPU-bound threads.
709 Some real-time applications might not have a single real-time
710 thread that saturates a given CPU, but instead might have
711 multiple real-time threads that, taken together, fully utilize
712 that CPU. In this case, you should set RCU_KTHREAD_PRIO to
713 a priority higher than the lowest-priority thread that is
714 conspiring to prevent the CPU from running any non-real-time
715 tasks. For example, if one thread at priority 10 and another
716 thread at priority 5 are between themselves fully consuming
717 the CPU time on a given CPU, then RCU_KTHREAD_PRIO should be
718 set to priority 6 or higher.
720 Specify the real-time priority, or take the default if unsure.
722 config RCU_BOOST_DELAY
723 int "Milliseconds to delay boosting after RCU grace-period start"
728 This option specifies the time to wait after the beginning of
729 a given grace period before priority-boosting preempted RCU
730 readers blocking that grace period. Note that any RCU reader
731 blocking an expedited RCU grace period is boosted immediately.
733 Accept the default if unsure.
736 bool "Offload RCU callback processing from boot-selected CPUs"
737 depends on TREE_RCU || PREEMPT_RCU
738 depends on RCU_EXPERT || NO_HZ_FULL
741 Use this option to reduce OS jitter for aggressive HPC or
742 real-time workloads. It can also be used to offload RCU
743 callback invocation to energy-efficient CPUs in battery-powered
744 asymmetric multiprocessors.
746 This option offloads callback invocation from the set of
747 CPUs specified at boot time by the rcu_nocbs parameter.
748 For each such CPU, a kthread ("rcuox/N") will be created to
749 invoke callbacks, where the "N" is the CPU being offloaded,
750 and where the "x" is "b" for RCU-bh, "p" for RCU-preempt, and
751 "s" for RCU-sched. Nothing prevents this kthread from running
752 on the specified CPUs, but (1) the kthreads may be preempted
753 between each callback, and (2) affinity or cgroups can be used
754 to force the kthreads to run on whatever set of CPUs is desired.
756 Say Y here if you want to help to debug reduced OS jitter.
757 Say N here if you are unsure.
760 prompt "Build-forced no-CBs CPUs"
761 default RCU_NOCB_CPU_NONE
762 depends on RCU_NOCB_CPU
764 This option allows no-CBs CPUs (whose RCU callbacks are invoked
765 from kthreads rather than from softirq context) to be specified
766 at build time. Additional no-CBs CPUs may be specified by
767 the rcu_nocbs= boot parameter.
769 config RCU_NOCB_CPU_NONE
770 bool "No build_forced no-CBs CPUs"
772 This option does not force any of the CPUs to be no-CBs CPUs.
773 Only CPUs designated by the rcu_nocbs= boot parameter will be
774 no-CBs CPUs, whose RCU callbacks will be invoked by per-CPU
775 kthreads whose names begin with "rcuo". All other CPUs will
776 invoke their own RCU callbacks in softirq context.
778 Select this option if you want to choose no-CBs CPUs at
779 boot time, for example, to allow testing of different no-CBs
780 configurations without having to rebuild the kernel each time.
782 config RCU_NOCB_CPU_ZERO
783 bool "CPU 0 is a build_forced no-CBs CPU"
785 This option forces CPU 0 to be a no-CBs CPU, so that its RCU
786 callbacks are invoked by a per-CPU kthread whose name begins
787 with "rcuo". Additional CPUs may be designated as no-CBs
788 CPUs using the rcu_nocbs= boot parameter will be no-CBs CPUs.
789 All other CPUs will invoke their own RCU callbacks in softirq
792 Select this if CPU 0 needs to be a no-CBs CPU for real-time
793 or energy-efficiency reasons, but the real reason it exists
794 is to ensure that randconfig testing covers mixed systems.
796 config RCU_NOCB_CPU_ALL
797 bool "All CPUs are build_forced no-CBs CPUs"
799 This option forces all CPUs to be no-CBs CPUs. The rcu_nocbs=
800 boot parameter will be ignored. All CPUs' RCU callbacks will
801 be executed in the context of per-CPU rcuo kthreads created for
802 this purpose. Assuming that the kthreads whose names start with
803 "rcuo" are bound to "housekeeping" CPUs, this reduces OS jitter
804 on the remaining CPUs, but might decrease memory locality during
805 RCU-callback invocation, thus potentially degrading throughput.
807 Select this if all CPUs need to be no-CBs CPUs for real-time
808 or energy-efficiency reasons.
812 endmenu # "RCU Subsystem"
819 tristate "Kernel .config support"
822 This option enables the complete Linux kernel ".config" file
823 contents to be saved in the kernel. It provides documentation
824 of which kernel options are used in a running kernel or in an
825 on-disk kernel. This information can be extracted from the kernel
826 image file with the script scripts/extract-ikconfig and used as
827 input to rebuild the current kernel or to build another kernel.
828 It can also be extracted from a running kernel by reading
829 /proc/config.gz if enabled (below).
832 bool "Enable access to .config through /proc/config.gz"
833 depends on IKCONFIG && PROC_FS
835 This option enables access to the kernel configuration file
836 through /proc/config.gz.
839 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
844 Select the minimal kernel log buffer size as a power of 2.
845 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config
846 parameter, see below. Any higher size also might be forced
847 by "log_buf_len" boot parameter.
857 config LOG_CPU_MAX_BUF_SHIFT
858 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)"
861 default 12 if !BASE_SMALL
862 default 0 if BASE_SMALL
865 This option allows to increase the default ring buffer size
866 according to the number of CPUs. The value defines the contribution
867 of each CPU as a power of 2. The used space is typically only few
868 lines however it might be much more when problems are reported,
871 The increased size means that a new buffer has to be allocated and
872 the original static one is unused. It makes sense only on systems
873 with more CPUs. Therefore this value is used only when the sum of
874 contributions is greater than the half of the default kernel ring
875 buffer as defined by LOG_BUF_SHIFT. The default values are set
876 so that more than 64 CPUs are needed to trigger the allocation.
878 Also this option is ignored when "log_buf_len" kernel parameter is
879 used as it forces an exact (power of two) size of the ring buffer.
881 The number of possible CPUs is used for this computation ignoring
882 hotplugging making the computation optimal for the worst case
883 scenario while allowing a simple algorithm to be used from bootup.
885 Examples shift values and their meaning:
886 17 => 128 KB for each CPU
887 16 => 64 KB for each CPU
888 15 => 32 KB for each CPU
889 14 => 16 KB for each CPU
890 13 => 8 KB for each CPU
891 12 => 4 KB for each CPU
893 config PRINTK_SAFE_LOG_BUF_SHIFT
894 int "Temporary per-CPU printk log buffer size (12 => 4KB, 13 => 8KB)"
899 Select the size of an alternate printk per-CPU buffer where messages
900 printed from usafe contexts are temporary stored. One example would
901 be NMI messages, another one - printk recursion. The messages are
902 copied to the main log buffer in a safe context to avoid a deadlock.
903 The value defines the size as a power of 2.
905 Those messages are rare and limited. The largest one is when
906 a backtrace is printed. It usually fits into 4KB. Select
907 8KB if you want to be on the safe side.
910 17 => 128 KB for each CPU
911 16 => 64 KB for each CPU
912 15 => 32 KB for each CPU
913 14 => 16 KB for each CPU
914 13 => 8 KB for each CPU
915 12 => 4 KB for each CPU
918 # Architectures with an unreliable sched_clock() should select this:
920 config HAVE_UNSTABLE_SCHED_CLOCK
923 config GENERIC_SCHED_CLOCK
927 # For architectures that want to enable the support for NUMA-affine scheduler
930 config ARCH_SUPPORTS_NUMA_BALANCING
934 # For architectures that prefer to flush all TLBs after a number of pages
935 # are unmapped instead of sending one IPI per page to flush. The architecture
936 # must provide guarantees on what happens if a clean TLB cache entry is
937 # written after the unmap. Details are in mm/rmap.c near the check for
938 # should_defer_flush. The architecture should also consider if the full flush
939 # and the refill costs are offset by the savings of sending fewer IPIs.
940 config ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
944 # For architectures that know their GCC __int128 support is sound
946 config ARCH_SUPPORTS_INT128
949 # For architectures that (ab)use NUMA to represent different memory regions
950 # all cpu-local but of different latencies, such as SuperH.
952 config ARCH_WANT_NUMA_VARIABLE_LOCALITY
955 config NUMA_BALANCING
956 bool "Memory placement aware NUMA scheduler"
957 depends on ARCH_SUPPORTS_NUMA_BALANCING
958 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
959 depends on SMP && NUMA && MIGRATION
961 This option adds support for automatic NUMA aware memory/task placement.
962 The mechanism is quite primitive and is based on migrating memory when
963 it has references to the node the task is running on.
965 This system will be inactive on UMA systems.
967 config NUMA_BALANCING_DEFAULT_ENABLED
968 bool "Automatically enable NUMA aware memory/task placement"
970 depends on NUMA_BALANCING
972 If set, automatic NUMA balancing will be enabled if running on a NUMA
976 bool "Control Group support"
979 This option adds support for grouping sets of processes together, for
980 use with process control subsystems such as Cpusets, CFS, memory
981 controls or device isolation.
983 - Documentation/scheduler/sched-design-CFS.txt (CFS)
984 - Documentation/cgroup-v1/ (features for grouping, isolation
985 and resource control)
995 bool "Memory controller"
999 Provides control over the memory footprint of tasks in a cgroup.
1002 bool "Swap controller"
1003 depends on MEMCG && SWAP
1005 Provides control over the swap space consumed by tasks in a cgroup.
1007 config MEMCG_SWAP_ENABLED
1008 bool "Swap controller enabled by default"
1009 depends on MEMCG_SWAP
1012 Memory Resource Controller Swap Extension comes with its price in
1013 a bigger memory consumption. General purpose distribution kernels
1014 which want to enable the feature but keep it disabled by default
1015 and let the user enable it by swapaccount=1 boot command line
1016 parameter should have this option unselected.
1017 For those who want to have the feature enabled by default should
1018 select this option (if, for some reason, they need to disable it
1019 then swapaccount=0 does the trick).
1022 bool "IO controller"
1026 Generic block IO controller cgroup interface. This is the common
1027 cgroup interface which should be used by various IO controlling
1030 Currently, CFQ IO scheduler uses it to recognize task groups and
1031 control disk bandwidth allocation (proportional time slice allocation)
1032 to such task groups. It is also used by bio throttling logic in
1033 block layer to implement upper limit in IO rates on a device.
1035 This option only enables generic Block IO controller infrastructure.
1036 One needs to also enable actual IO controlling logic/policy. For
1037 enabling proportional weight division of disk bandwidth in CFQ, set
1038 CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set
1039 CONFIG_BLK_DEV_THROTTLING=y.
1041 See Documentation/cgroup-v1/blkio-controller.txt for more information.
1043 config DEBUG_BLK_CGROUP
1044 bool "IO controller debugging"
1045 depends on BLK_CGROUP
1048 Enable some debugging help. Currently it exports additional stat
1049 files in a cgroup which can be useful for debugging.
1051 config CGROUP_WRITEBACK
1053 depends on MEMCG && BLK_CGROUP
1056 menuconfig CGROUP_SCHED
1057 bool "CPU controller"
1060 This feature lets CPU scheduler recognize task groups and control CPU
1061 bandwidth allocation to such task groups. It uses cgroups to group
1065 config FAIR_GROUP_SCHED
1066 bool "Group scheduling for SCHED_OTHER"
1067 depends on CGROUP_SCHED
1068 default CGROUP_SCHED
1070 config CFS_BANDWIDTH
1071 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED"
1072 depends on FAIR_GROUP_SCHED
1075 This option allows users to define CPU bandwidth rates (limits) for
1076 tasks running within the fair group scheduler. Groups with no limit
1077 set are considered to be unconstrained and will run with no
1079 See tip/Documentation/scheduler/sched-bwc.txt for more information.
1081 config RT_GROUP_SCHED
1082 bool "Group scheduling for SCHED_RR/FIFO"
1083 depends on CGROUP_SCHED
1086 This feature lets you explicitly allocate real CPU bandwidth
1087 to task groups. If enabled, it will also make it impossible to
1088 schedule realtime tasks for non-root users until you allocate
1089 realtime bandwidth for them.
1090 See Documentation/scheduler/sched-rt-group.txt for more information.
1095 bool "PIDs controller"
1097 Provides enforcement of process number limits in the scope of a
1098 cgroup. Any attempt to fork more processes than is allowed in the
1099 cgroup will fail. PIDs are fundamentally a global resource because it
1100 is fairly trivial to reach PID exhaustion before you reach even a
1101 conservative kmemcg limit. As a result, it is possible to grind a
1102 system to halt without being limited by other cgroup policies. The
1103 PIDs controller is designed to stop this from happening.
1105 It should be noted that organisational operations (such as attaching
1106 to a cgroup hierarchy will *not* be blocked by the PIDs controller),
1107 since the PIDs limit only affects a process's ability to fork, not to
1111 bool "RDMA controller"
1113 Provides enforcement of RDMA resources defined by IB stack.
1114 It is fairly easy for consumers to exhaust RDMA resources, which
1115 can result into resource unavailability to other consumers.
1116 RDMA controller is designed to stop this from happening.
1117 Attaching processes with active RDMA resources to the cgroup
1118 hierarchy is allowed even if can cross the hierarchy's limit.
1120 config CGROUP_FREEZER
1121 bool "Freezer controller"
1123 Provides a way to freeze and unfreeze all tasks in a
1126 This option affects the ORIGINAL cgroup interface. The cgroup2 memory
1127 controller includes important in-kernel memory consumers per default.
1129 If you're using cgroup2, say N.
1131 config CGROUP_HUGETLB
1132 bool "HugeTLB controller"
1133 depends on HUGETLB_PAGE
1137 Provides a cgroup controller for HugeTLB pages.
1138 When you enable this, you can put a per cgroup limit on HugeTLB usage.
1139 The limit is enforced during page fault. Since HugeTLB doesn't
1140 support page reclaim, enforcing the limit at page fault time implies
1141 that, the application will get SIGBUS signal if it tries to access
1142 HugeTLB pages beyond its limit. This requires the application to know
1143 beforehand how much HugeTLB pages it would require for its use. The
1144 control group is tracked in the third page lru pointer. This means
1145 that we cannot use the controller with huge page less than 3 pages.
1148 bool "Cpuset controller"
1150 This option will let you create and manage CPUSETs which
1151 allow dynamically partitioning a system into sets of CPUs and
1152 Memory Nodes and assigning tasks to run only within those sets.
1153 This is primarily useful on large SMP or NUMA systems.
1157 config PROC_PID_CPUSET
1158 bool "Include legacy /proc/<pid>/cpuset file"
1162 config CGROUP_DEVICE
1163 bool "Device controller"
1165 Provides a cgroup controller implementing whitelists for
1166 devices which a process in the cgroup can mknod or open.
1168 config CGROUP_CPUACCT
1169 bool "Simple CPU accounting controller"
1171 Provides a simple controller for monitoring the
1172 total CPU consumed by the tasks in a cgroup.
1175 bool "Perf controller"
1176 depends on PERF_EVENTS
1178 This option extends the perf per-cpu mode to restrict monitoring
1179 to threads which belong to the cgroup specified and run on the
1185 bool "Support for eBPF programs attached to cgroups"
1186 depends on BPF_SYSCALL
1187 select SOCK_CGROUP_DATA
1189 Allow attaching eBPF programs to a cgroup using the bpf(2)
1190 syscall command BPF_PROG_ATTACH.
1192 In which context these programs are accessed depends on the type
1193 of attachment. For instance, programs that are attached using
1194 BPF_CGROUP_INET_INGRESS will be executed on the ingress path of
1198 bool "Example controller"
1201 This option enables a simple controller that exports
1202 debugging information about the cgroups framework.
1206 config SOCK_CGROUP_DATA
1212 config CHECKPOINT_RESTORE
1213 bool "Checkpoint/restore support" if EXPERT
1214 select PROC_CHILDREN
1217 Enables additional kernel features in a sake of checkpoint/restore.
1218 In particular it adds auxiliary prctl codes to setup process text,
1219 data and heap segment sizes, and a few additional /proc filesystem
1222 If unsure, say N here.
1224 menuconfig NAMESPACES
1225 bool "Namespaces support" if EXPERT
1226 depends on MULTIUSER
1229 Provides the way to make tasks work with different objects using
1230 the same id. For example same IPC id may refer to different objects
1231 or same user id or pid may refer to different tasks when used in
1232 different namespaces.
1237 bool "UTS namespace"
1240 In this namespace tasks see different info provided with the
1244 bool "IPC namespace"
1245 depends on (SYSVIPC || POSIX_MQUEUE)
1248 In this namespace tasks work with IPC ids which correspond to
1249 different IPC objects in different namespaces.
1252 bool "User namespace"
1255 This allows containers, i.e. vservers, to use user namespaces
1256 to provide different user info for different servers.
1258 When user namespaces are enabled in the kernel it is
1259 recommended that the MEMCG option also be enabled and that
1260 user-space use the memory control groups to limit the amount
1261 of memory a memory unprivileged users can use.
1266 bool "PID Namespaces"
1269 Support process id namespaces. This allows having multiple
1270 processes with the same pid as long as they are in different
1271 pid namespaces. This is a building block of containers.
1274 bool "Network namespace"
1278 Allow user space to create what appear to be multiple instances
1279 of the network stack.
1283 config SCHED_AUTOGROUP
1284 bool "Automatic process group scheduling"
1287 select FAIR_GROUP_SCHED
1289 This option optimizes the scheduler for common desktop workloads by
1290 automatically creating and populating task groups. This separation
1291 of workloads isolates aggressive CPU burners (like build jobs) from
1292 desktop applications. Task group autogeneration is currently based
1295 config SYSFS_DEPRECATED
1296 bool "Enable deprecated sysfs features to support old userspace tools"
1300 This option adds code that switches the layout of the "block" class
1301 devices, to not show up in /sys/class/block/, but only in
1304 This switch is only active when the sysfs.deprecated=1 boot option is
1305 passed or the SYSFS_DEPRECATED_V2 option is set.
1307 This option allows new kernels to run on old distributions and tools,
1308 which might get confused by /sys/class/block/. Since 2007/2008 all
1309 major distributions and tools handle this just fine.
1311 Recent distributions and userspace tools after 2009/2010 depend on
1312 the existence of /sys/class/block/, and will not work with this
1315 Only if you are using a new kernel on an old distribution, you might
1318 config SYSFS_DEPRECATED_V2
1319 bool "Enable deprecated sysfs features by default"
1322 depends on SYSFS_DEPRECATED
1324 Enable deprecated sysfs by default.
1326 See the CONFIG_SYSFS_DEPRECATED option for more details about this
1329 Only if you are using a new kernel on an old distribution, you might
1330 need to say Y here. Even then, odds are you would not need it
1331 enabled, you can always pass the boot option if absolutely necessary.
1334 bool "Kernel->user space relay support (formerly relayfs)"
1337 This option enables support for relay interface support in
1338 certain file systems (such as debugfs).
1339 It is designed to provide an efficient mechanism for tools and
1340 facilities to relay large amounts of data from kernel space to
1345 config BLK_DEV_INITRD
1346 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
1347 depends on BROKEN || !FRV
1349 The initial RAM filesystem is a ramfs which is loaded by the
1350 boot loader (loadlin or lilo) and that is mounted as root
1351 before the normal boot procedure. It is typically used to
1352 load modules needed to mount the "real" root file system,
1353 etc. See <file:Documentation/admin-guide/initrd.rst> for details.
1355 If RAM disk support (BLK_DEV_RAM) is also included, this
1356 also enables initial RAM disk (initrd) support and adds
1357 15 Kbytes (more on some other architectures) to the kernel size.
1363 source "usr/Kconfig"
1368 prompt "Compiler optimization level"
1369 default CONFIG_CC_OPTIMIZE_FOR_PERFORMANCE
1371 config CC_OPTIMIZE_FOR_PERFORMANCE
1372 bool "Optimize for performance"
1374 This is the default optimization level for the kernel, building
1375 with the "-O2" compiler flag for best performance and most
1376 helpful compile-time warnings.
1378 config CC_OPTIMIZE_FOR_SIZE
1379 bool "Optimize for size"
1381 Enabling this option will pass "-Os" instead of "-O2" to
1382 your compiler resulting in a smaller kernel.
1397 config SYSCTL_EXCEPTION_TRACE
1400 Enable support for /proc/sys/debug/exception-trace.
1402 config SYSCTL_ARCH_UNALIGN_NO_WARN
1405 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap
1406 Allows arch to define/use @no_unaligned_warning to possibly warn
1407 about unaligned access emulation going on under the hood.
1409 config SYSCTL_ARCH_UNALIGN_ALLOW
1412 Enable support for /proc/sys/kernel/unaligned-trap
1413 Allows arches to define/use @unaligned_enabled to runtime toggle
1414 the unaligned access emulation.
1415 see arch/parisc/kernel/unaligned.c for reference
1417 config HAVE_PCSPKR_PLATFORM
1420 # interpreter that classic socket filters depend on
1425 bool "Configure standard kernel features (expert users)"
1426 # Unhide debug options, to make the on-by-default options visible
1429 This option allows certain base kernel options and settings
1430 to be disabled or tweaked. This is for specialized
1431 environments which can tolerate a "non-standard" kernel.
1432 Only use this if you really know what you are doing.
1435 bool "Enable 16-bit UID system calls" if EXPERT
1436 depends on HAVE_UID16 && MULTIUSER
1439 This enables the legacy 16-bit UID syscall wrappers.
1442 bool "Multiple users, groups and capabilities support" if EXPERT
1445 This option enables support for non-root users, groups and
1448 If you say N here, all processes will run with UID 0, GID 0, and all
1449 possible capabilities. Saying N here also compiles out support for
1450 system calls related to UIDs, GIDs, and capabilities, such as setuid,
1453 If unsure, say Y here.
1455 config SGETMASK_SYSCALL
1456 bool "sgetmask/ssetmask syscalls support" if EXPERT
1457 def_bool PARISC || MN10300 || BLACKFIN || M68K || PPC || MIPS || X86 || SPARC || CRIS || MICROBLAZE || SUPERH
1459 sys_sgetmask and sys_ssetmask are obsolete system calls
1460 no longer supported in libc but still enabled by default in some
1463 If unsure, leave the default option here.
1465 config SYSFS_SYSCALL
1466 bool "Sysfs syscall support" if EXPERT
1469 sys_sysfs is an obsolete system call no longer supported in libc.
1470 Note that disabling this option is more secure but might break
1471 compatibility with some systems.
1473 If unsure say Y here.
1475 config SYSCTL_SYSCALL
1476 bool "Sysctl syscall support" if EXPERT
1477 depends on PROC_SYSCTL
1481 sys_sysctl uses binary paths that have been found challenging
1482 to properly maintain and use. The interface in /proc/sys
1483 using paths with ascii names is now the primary path to this
1486 Almost nothing using the binary sysctl interface so if you are
1487 trying to save some space it is probably safe to disable this,
1488 making your kernel marginally smaller.
1490 If unsure say N here.
1493 bool "Posix Clocks & timers" if EXPERT
1496 This includes native support for POSIX timers to the kernel.
1497 Some embedded systems have no use for them and therefore they
1498 can be configured out to reduce the size of the kernel image.
1500 When this option is disabled, the following syscalls won't be
1501 available: timer_create, timer_gettime: timer_getoverrun,
1502 timer_settime, timer_delete, clock_adjtime, getitimer,
1503 setitimer, alarm. Furthermore, the clock_settime, clock_gettime,
1504 clock_getres and clock_nanosleep syscalls will be limited to
1505 CLOCK_REALTIME, CLOCK_MONOTONIC and CLOCK_BOOTTIME only.
1510 bool "Load all symbols for debugging/ksymoops" if EXPERT
1513 Say Y here to let the kernel print out symbolic crash information and
1514 symbolic stack backtraces. This increases the size of the kernel
1515 somewhat, as all symbols have to be loaded into the kernel image.
1518 bool "Include all symbols in kallsyms"
1519 depends on DEBUG_KERNEL && KALLSYMS
1521 Normally kallsyms only contains the symbols of functions for nicer
1522 OOPS messages and backtraces (i.e., symbols from the text and inittext
1523 sections). This is sufficient for most cases. And only in very rare
1524 cases (e.g., when a debugger is used) all symbols are required (e.g.,
1525 names of variables from the data sections, etc).
1527 This option makes sure that all symbols are loaded into the kernel
1528 image (i.e., symbols from all sections) in cost of increased kernel
1529 size (depending on the kernel configuration, it may be 300KiB or
1530 something like this).
1532 Say N unless you really need all symbols.
1534 config KALLSYMS_ABSOLUTE_PERCPU
1537 default X86_64 && SMP
1539 config KALLSYMS_BASE_RELATIVE
1542 default !IA64 && !(TILE && 64BIT)
1544 Instead of emitting them as absolute values in the native word size,
1545 emit the symbol references in the kallsyms table as 32-bit entries,
1546 each containing a relative value in the range [base, base + U32_MAX]
1547 or, when KALLSYMS_ABSOLUTE_PERCPU is in effect, each containing either
1548 an absolute value in the range [0, S32_MAX] or a relative value in the
1549 range [base, base + S32_MAX], where base is the lowest relative symbol
1550 address encountered in the image.
1552 On 64-bit builds, this reduces the size of the address table by 50%,
1553 but more importantly, it results in entries whose values are build
1554 time constants, and no relocation pass is required at runtime to fix
1555 up the entries based on the runtime load address of the kernel.
1559 bool "Enable support for printk" if EXPERT
1562 This option enables normal printk support. Removing it
1563 eliminates most of the message strings from the kernel image
1564 and makes the kernel more or less silent. As this makes it
1565 very difficult to diagnose system problems, saying N here is
1566 strongly discouraged.
1574 bool "BUG() support" if EXPERT
1577 Disabling this option eliminates support for BUG and WARN, reducing
1578 the size of your kernel image and potentially quietly ignoring
1579 numerous fatal conditions. You should only consider disabling this
1580 option for embedded systems with no facilities for reporting errors.
1586 bool "Enable ELF core dumps" if EXPERT
1588 Enable support for generating core dumps. Disabling saves about 4k.
1591 config PCSPKR_PLATFORM
1592 bool "Enable PC-Speaker support" if EXPERT
1593 depends on HAVE_PCSPKR_PLATFORM
1597 This option allows to disable the internal PC-Speaker
1598 support, saving some memory.
1602 bool "Enable full-sized data structures for core" if EXPERT
1604 Disabling this option reduces the size of miscellaneous core
1605 kernel data structures. This saves memory on small machines,
1606 but may reduce performance.
1609 bool "Enable futex support" if EXPERT
1613 Disabling this option will cause the kernel to be built without
1614 support for "fast userspace mutexes". The resulting kernel may not
1615 run glibc-based applications correctly.
1617 config HAVE_FUTEX_CMPXCHG
1621 Architectures should select this if futex_atomic_cmpxchg_inatomic()
1622 is implemented and always working. This removes a couple of runtime
1626 bool "Enable eventpoll support" if EXPERT
1630 Disabling this option will cause the kernel to be built without
1631 support for epoll family of system calls.
1634 bool "Enable signalfd() system call" if EXPERT
1638 Enable the signalfd() system call that allows to receive signals
1639 on a file descriptor.
1644 bool "Enable timerfd() system call" if EXPERT
1648 Enable the timerfd() system call that allows to receive timer
1649 events on a file descriptor.
1654 bool "Enable eventfd() system call" if EXPERT
1658 Enable the eventfd() system call that allows to receive both
1659 kernel notification (ie. KAIO) or userspace notifications.
1663 # syscall, maps, verifier
1665 bool "Enable bpf() system call"
1670 Enable the bpf() system call that allows to manipulate eBPF
1671 programs and maps via file descriptors.
1674 bool "Use full shmem filesystem" if EXPERT
1678 The shmem is an internal filesystem used to manage shared memory.
1679 It is backed by swap and manages resource limits. It is also exported
1680 to userspace as tmpfs if TMPFS is enabled. Disabling this
1681 option replaces shmem and tmpfs with the much simpler ramfs code,
1682 which may be appropriate on small systems without swap.
1685 bool "Enable AIO support" if EXPERT
1688 This option enables POSIX asynchronous I/O which may by used
1689 by some high performance threaded applications. Disabling
1690 this option saves about 7k.
1692 config ADVISE_SYSCALLS
1693 bool "Enable madvise/fadvise syscalls" if EXPERT
1696 This option enables the madvise and fadvise syscalls, used by
1697 applications to advise the kernel about their future memory or file
1698 usage, improving performance. If building an embedded system where no
1699 applications use these syscalls, you can disable this option to save
1703 bool "Enable userfaultfd() system call"
1707 Enable the userfaultfd() system call that allows to intercept and
1708 handle page faults in userland.
1712 bool "Enable PCI quirk workarounds" if EXPERT
1715 This enables workarounds for various PCI chipset
1716 bugs/quirks. Disable this only if your target machine is
1717 unaffected by PCI quirks.
1720 bool "Enable membarrier() system call" if EXPERT
1723 Enable the membarrier() system call that allows issuing memory
1724 barriers across all running threads, which can be used to distribute
1725 the cost of user-space memory barriers asymmetrically by transforming
1726 pairs of memory barriers into pairs consisting of membarrier() and a
1732 bool "Embedded system"
1733 option allnoconfig_y
1736 This option should be enabled if compiling the kernel for
1737 an embedded system so certain expert options are available
1740 config HAVE_PERF_EVENTS
1743 See tools/perf/design.txt for details.
1745 config PERF_USE_VMALLOC
1748 See tools/perf/design.txt for details
1751 bool "PC/104 support"
1753 Expose PC/104 form factor device drivers and options available for
1754 selection and configuration. Enable this option if your target
1755 machine has a PC/104 bus.
1757 menu "Kernel Performance Events And Counters"
1760 bool "Kernel performance events and counters"
1761 default y if PROFILING
1762 depends on HAVE_PERF_EVENTS
1767 Enable kernel support for various performance events provided
1768 by software and hardware.
1770 Software events are supported either built-in or via the
1771 use of generic tracepoints.
1773 Most modern CPUs support performance events via performance
1774 counter registers. These registers count the number of certain
1775 types of hw events: such as instructions executed, cachemisses
1776 suffered, or branches mis-predicted - without slowing down the
1777 kernel or applications. These registers can also trigger interrupts
1778 when a threshold number of events have passed - and can thus be
1779 used to profile the code that runs on that CPU.
1781 The Linux Performance Event subsystem provides an abstraction of
1782 these software and hardware event capabilities, available via a
1783 system call and used by the "perf" utility in tools/perf/. It
1784 provides per task and per CPU counters, and it provides event
1785 capabilities on top of those.
1789 config DEBUG_PERF_USE_VMALLOC
1791 bool "Debug: use vmalloc to back perf mmap() buffers"
1792 depends on PERF_EVENTS && DEBUG_KERNEL && !PPC
1793 select PERF_USE_VMALLOC
1795 Use vmalloc memory to back perf mmap() buffers.
1797 Mostly useful for debugging the vmalloc code on platforms
1798 that don't require it.
1804 config VM_EVENT_COUNTERS
1806 bool "Enable VM event counters for /proc/vmstat" if EXPERT
1808 VM event counters are needed for event counts to be shown.
1809 This option allows the disabling of the VM event counters
1810 on EXPERT systems. /proc/vmstat will only show page counts
1811 if VM event counters are disabled.
1815 bool "Enable SLUB debugging support" if EXPERT
1816 depends on SLUB && SYSFS
1818 SLUB has extensive debug support features. Disabling these can
1819 result in significant savings in code size. This also disables
1820 SLUB sysfs support. /sys/slab will not exist and there will be
1821 no support for cache validation etc.
1823 config SLUB_MEMCG_SYSFS_ON
1825 bool "Enable memcg SLUB sysfs support by default" if EXPERT
1826 depends on SLUB && SYSFS && MEMCG
1828 SLUB creates a directory under /sys/kernel/slab for each
1829 allocation cache to host info and debug files. If memory
1830 cgroup is enabled, each cache can have per memory cgroup
1831 caches. SLUB can create the same sysfs directories for these
1832 caches under /sys/kernel/slab/CACHE/cgroup but it can lead
1833 to a very high number of debug files being created. This is
1834 controlled by slub_memcg_sysfs boot parameter and this
1835 config option determines the parameter's default value.
1838 bool "Disable heap randomization"
1841 Randomizing heap placement makes heap exploits harder, but it
1842 also breaks ancient binaries (including anything libc5 based).
1843 This option changes the bootup default to heap randomization
1844 disabled, and can be overridden at runtime by setting
1845 /proc/sys/kernel/randomize_va_space to 2.
1847 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1850 prompt "Choose SLAB allocator"
1853 This option allows to select a slab allocator.
1857 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1859 The regular slab allocator that is established and known to work
1860 well in all environments. It organizes cache hot objects in
1861 per cpu and per node queues.
1864 bool "SLUB (Unqueued Allocator)"
1865 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1867 SLUB is a slab allocator that minimizes cache line usage
1868 instead of managing queues of cached objects (SLAB approach).
1869 Per cpu caching is realized using slabs of objects instead
1870 of queues of objects. SLUB can use memory efficiently
1871 and has enhanced diagnostics. SLUB is the default choice for
1876 bool "SLOB (Simple Allocator)"
1878 SLOB replaces the stock allocator with a drastically simpler
1879 allocator. SLOB is generally more space efficient but
1880 does not perform as well on large systems.
1884 config SLAB_FREELIST_RANDOM
1886 depends on SLAB || SLUB
1887 bool "SLAB freelist randomization"
1889 Randomizes the freelist order used on creating new pages. This
1890 security feature reduces the predictability of the kernel slab
1891 allocator against heap overflows.
1893 config SLUB_CPU_PARTIAL
1895 depends on SLUB && SMP
1896 bool "SLUB per cpu partial cache"
1898 Per cpu partial caches accellerate objects allocation and freeing
1899 that is local to a processor at the price of more indeterminism
1900 in the latency of the free. On overflow these caches will be cleared
1901 which requires the taking of locks that may cause latency spikes.
1902 Typically one would choose no for a realtime system.
1904 config MMAP_ALLOW_UNINITIALIZED
1905 bool "Allow mmapped anonymous memory to be uninitialized"
1906 depends on EXPERT && !MMU
1909 Normally, and according to the Linux spec, anonymous memory obtained
1910 from mmap() has it's contents cleared before it is passed to
1911 userspace. Enabling this config option allows you to request that
1912 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
1913 providing a huge performance boost. If this option is not enabled,
1914 then the flag will be ignored.
1916 This is taken advantage of by uClibc's malloc(), and also by
1917 ELF-FDPIC binfmt's brk and stack allocator.
1919 Because of the obvious security issues, this option should only be
1920 enabled on embedded devices where you control what is run in
1921 userspace. Since that isn't generally a problem on no-MMU systems,
1922 it is normally safe to say Y here.
1924 See Documentation/nommu-mmap.txt for more information.
1926 config SYSTEM_DATA_VERIFICATION
1928 select SYSTEM_TRUSTED_KEYRING
1932 select ASYMMETRIC_KEY_TYPE
1933 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE
1936 select X509_CERTIFICATE_PARSER
1937 select PKCS7_MESSAGE_PARSER
1939 Provide PKCS#7 message verification using the contents of the system
1940 trusted keyring to provide public keys. This then can be used for
1941 module verification, kexec image verification and firmware blob
1945 bool "Profiling support"
1947 Say Y here to enable the extended profiling support mechanisms used
1948 by profilers such as OProfile.
1951 # Place an empty function call at each tracepoint site. Can be
1952 # dynamically changed for a probe function.
1957 source "arch/Kconfig"
1959 endmenu # General setup
1961 config HAVE_GENERIC_DMA_COHERENT
1968 depends on SLAB || SLUB_DEBUG
1976 default 0 if BASE_FULL
1977 default 1 if !BASE_FULL
1980 bool "Enable loadable module support"
1983 Kernel modules are small pieces of compiled code which can
1984 be inserted in the running kernel, rather than being
1985 permanently built into the kernel. You use the "modprobe"
1986 tool to add (and sometimes remove) them. If you say Y here,
1987 many parts of the kernel can be built as modules (by
1988 answering M instead of Y where indicated): this is most
1989 useful for infrequently used options which are not required
1990 for booting. For more information, see the man pages for
1991 modprobe, lsmod, modinfo, insmod and rmmod.
1993 If you say Y here, you will need to run "make
1994 modules_install" to put the modules under /lib/modules/
1995 where modprobe can find them (you may need to be root to do
2002 config MODULE_FORCE_LOAD
2003 bool "Forced module loading"
2006 Allow loading of modules without version information (ie. modprobe
2007 --force). Forced module loading sets the 'F' (forced) taint flag and
2008 is usually a really bad idea.
2010 config MODULE_UNLOAD
2011 bool "Module unloading"
2013 Without this option you will not be able to unload any
2014 modules (note that some modules may not be unloadable
2015 anyway), which makes your kernel smaller, faster
2016 and simpler. If unsure, say Y.
2018 config MODULE_FORCE_UNLOAD
2019 bool "Forced module unloading"
2020 depends on MODULE_UNLOAD
2022 This option allows you to force a module to unload, even if the
2023 kernel believes it is unsafe: the kernel will remove the module
2024 without waiting for anyone to stop using it (using the -f option to
2025 rmmod). This is mainly for kernel developers and desperate users.
2029 bool "Module versioning support"
2031 Usually, you have to use modules compiled with your kernel.
2032 Saying Y here makes it sometimes possible to use modules
2033 compiled for different kernels, by adding enough information
2034 to the modules to (hopefully) spot any changes which would
2035 make them incompatible with the kernel you are running. If
2038 config MODULE_REL_CRCS
2040 depends on MODVERSIONS
2042 config MODULE_SRCVERSION_ALL
2043 bool "Source checksum for all modules"
2045 Modules which contain a MODULE_VERSION get an extra "srcversion"
2046 field inserted into their modinfo section, which contains a
2047 sum of the source files which made it. This helps maintainers
2048 see exactly which source was used to build a module (since
2049 others sometimes change the module source without updating
2050 the version). With this option, such a "srcversion" field
2051 will be created for all modules. If unsure, say N.
2054 bool "Module signature verification"
2056 select SYSTEM_DATA_VERIFICATION
2058 Check modules for valid signatures upon load: the signature
2059 is simply appended to the module. For more information see
2060 Documentation/module-signing.txt.
2062 Note that this option adds the OpenSSL development packages as a
2063 kernel build dependency so that the signing tool can use its crypto
2066 !!!WARNING!!! If you enable this option, you MUST make sure that the
2067 module DOES NOT get stripped after being signed. This includes the
2068 debuginfo strip done by some packagers (such as rpmbuild) and
2069 inclusion into an initramfs that wants the module size reduced.
2071 config MODULE_SIG_FORCE
2072 bool "Require modules to be validly signed"
2073 depends on MODULE_SIG
2075 Reject unsigned modules or signed modules for which we don't have a
2076 key. Without this, such modules will simply taint the kernel.
2078 config MODULE_SIG_ALL
2079 bool "Automatically sign all modules"
2081 depends on MODULE_SIG
2083 Sign all modules during make modules_install. Without this option,
2084 modules must be signed manually, using the scripts/sign-file tool.
2086 comment "Do not forget to sign required modules with scripts/sign-file"
2087 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL
2090 prompt "Which hash algorithm should modules be signed with?"
2091 depends on MODULE_SIG
2093 This determines which sort of hashing algorithm will be used during
2094 signature generation. This algorithm _must_ be built into the kernel
2095 directly so that signature verification can take place. It is not
2096 possible to load a signed module containing the algorithm to check
2097 the signature on that module.
2099 config MODULE_SIG_SHA1
2100 bool "Sign modules with SHA-1"
2103 config MODULE_SIG_SHA224
2104 bool "Sign modules with SHA-224"
2105 select CRYPTO_SHA256
2107 config MODULE_SIG_SHA256
2108 bool "Sign modules with SHA-256"
2109 select CRYPTO_SHA256
2111 config MODULE_SIG_SHA384
2112 bool "Sign modules with SHA-384"
2113 select CRYPTO_SHA512
2115 config MODULE_SIG_SHA512
2116 bool "Sign modules with SHA-512"
2117 select CRYPTO_SHA512
2121 config MODULE_SIG_HASH
2123 depends on MODULE_SIG
2124 default "sha1" if MODULE_SIG_SHA1
2125 default "sha224" if MODULE_SIG_SHA224
2126 default "sha256" if MODULE_SIG_SHA256
2127 default "sha384" if MODULE_SIG_SHA384
2128 default "sha512" if MODULE_SIG_SHA512
2130 config MODULE_COMPRESS
2131 bool "Compress modules on installation"
2135 Compresses kernel modules when 'make modules_install' is run; gzip or
2136 xz depending on "Compression algorithm" below.
2138 module-init-tools MAY support gzip, and kmod MAY support gzip and xz.
2140 Out-of-tree kernel modules installed using Kbuild will also be
2141 compressed upon installation.
2143 Note: for modules inside an initrd or initramfs, it's more efficient
2144 to compress the whole initrd or initramfs instead.
2146 Note: This is fully compatible with signed modules.
2151 prompt "Compression algorithm"
2152 depends on MODULE_COMPRESS
2153 default MODULE_COMPRESS_GZIP
2155 This determines which sort of compression will be used during
2156 'make modules_install'.
2158 GZIP (default) and XZ are supported.
2160 config MODULE_COMPRESS_GZIP
2163 config MODULE_COMPRESS_XZ
2168 config TRIM_UNUSED_KSYMS
2169 bool "Trim unused exported kernel symbols"
2170 depends on MODULES && !UNUSED_SYMBOLS
2172 The kernel and some modules make many symbols available for
2173 other modules to use via EXPORT_SYMBOL() and variants. Depending
2174 on the set of modules being selected in your kernel configuration,
2175 many of those exported symbols might never be used.
2177 This option allows for unused exported symbols to be dropped from
2178 the build. In turn, this provides the compiler more opportunities
2179 (especially when using LTO) for optimizing the code and reducing
2180 binary size. This might have some security advantages as well.
2182 If unsure, or if you need to build out-of-tree modules, say N.
2186 config MODULES_TREE_LOOKUP
2188 depends on PERF_EVENTS || TRACING
2190 config INIT_ALL_POSSIBLE
2193 Back when each arch used to define their own cpu_online_mask and
2194 cpu_possible_mask, some of them chose to initialize cpu_possible_mask
2195 with all 1s, and others with all 0s. When they were centralised,
2196 it was better to provide this option than to break all the archs
2197 and have several arch maintainers pursuing me down dark alleys.
2199 source "block/Kconfig"
2201 config PREEMPT_NOTIFIERS
2211 Build a simple ASN.1 grammar compiler that produces a bytecode output
2212 that can be interpreted by the ASN.1 stream decoder and used to
2213 inform it as to what tags are to be expected in a stream and what
2214 functions to call on what tags.
2216 source "kernel/Kconfig.locks"
projects / karo-tx-linux.git / blob