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
535 This option enables a task-based RCU implementation that uses
536 only voluntary context switch (not preemption!), idle, and
537 user-mode execution as quiescent states.
539 config RCU_STALL_COMMON
540 def_bool ( TREE_RCU || PREEMPT_RCU || RCU_TRACE )
542 This option enables RCU CPU stall code that is common between
543 the TINY and TREE variants of RCU. The purpose is to allow
544 the tiny variants to disable RCU CPU stall warnings, while
545 making these warnings mandatory for the tree variants.
547 config CONTEXT_TRACKING
550 config CONTEXT_TRACKING_FORCE
551 bool "Force context tracking"
552 depends on CONTEXT_TRACKING
553 default y if !NO_HZ_FULL
555 The major pre-requirement for full dynticks to work is to
556 support the context tracking subsystem. But there are also
557 other dependencies to provide in order to make the full
560 This option stands for testing when an arch implements the
561 context tracking backend but doesn't yet fullfill all the
562 requirements to make the full dynticks feature working.
563 Without the full dynticks, there is no way to test the support
564 for context tracking and the subsystems that rely on it: RCU
565 userspace extended quiescent state and tickless cputime
566 accounting. This option copes with the absence of the full
567 dynticks subsystem by forcing the context tracking on all
570 Say Y only if you're working on the development of an
571 architecture backend for the context tracking.
573 Say N otherwise, this option brings an overhead that you
574 don't want in production.
578 int "Tree-based hierarchical RCU fanout value"
581 depends on (TREE_RCU || PREEMPT_RCU) && RCU_EXPERT
585 This option controls the fanout of hierarchical implementations
586 of RCU, allowing RCU to work efficiently on machines with
587 large numbers of CPUs. This value must be at least the fourth
588 root of NR_CPUS, which allows NR_CPUS to be insanely large.
589 The default value of RCU_FANOUT should be used for production
590 systems, but if you are stress-testing the RCU implementation
591 itself, small RCU_FANOUT values allow you to test large-system
592 code paths on small(er) systems.
594 Select a specific number if testing RCU itself.
595 Take the default if unsure.
597 config RCU_FANOUT_LEAF
598 int "Tree-based hierarchical RCU leaf-level fanout value"
601 depends on (TREE_RCU || PREEMPT_RCU) && RCU_EXPERT
604 This option controls the leaf-level fanout of hierarchical
605 implementations of RCU, and allows trading off cache misses
606 against lock contention. Systems that synchronize their
607 scheduling-clock interrupts for energy-efficiency reasons will
608 want the default because the smaller leaf-level fanout keeps
609 lock contention levels acceptably low. Very large systems
610 (hundreds or thousands of CPUs) will instead want to set this
611 value to the maximum value possible in order to reduce the
612 number of cache misses incurred during RCU's grace-period
613 initialization. These systems tend to run CPU-bound, and thus
614 are not helped by synchronized interrupts, and thus tend to
615 skew them, which reduces lock contention enough that large
616 leaf-level fanouts work well.
618 Select a specific number if testing RCU itself.
620 Select the maximum permissible value for large systems.
622 Take the default if unsure.
624 config RCU_FAST_NO_HZ
625 bool "Accelerate last non-dyntick-idle CPU's grace periods"
626 depends on NO_HZ_COMMON && SMP && RCU_EXPERT
629 This option permits CPUs to enter dynticks-idle state even if
630 they have RCU callbacks queued, and prevents RCU from waking
631 these CPUs up more than roughly once every four jiffies (by
632 default, you can adjust this using the rcutree.rcu_idle_gp_delay
633 parameter), thus improving energy efficiency. On the other
634 hand, this option increases the duration of RCU grace periods,
635 for example, slowing down synchronize_rcu().
637 Say Y if energy efficiency is critically important, and you
638 don't care about increased grace-period durations.
640 Say N if you are unsure.
642 config TREE_RCU_TRACE
643 def_bool RCU_TRACE && ( TREE_RCU || PREEMPT_RCU )
646 This option provides tracing for the TREE_RCU and
647 PREEMPT_RCU implementations, permitting Makefile to
648 trivially select kernel/rcutree_trace.c.
651 bool "Enable RCU priority boosting"
652 depends on RT_MUTEXES && PREEMPT_RCU && RCU_EXPERT
655 This option boosts the priority of preempted RCU readers that
656 block the current preemptible RCU grace period for too long.
657 This option also prevents heavy loads from blocking RCU
658 callback invocation for all flavors of RCU.
660 Say Y here if you are working with real-time apps or heavy loads
661 Say N here if you are unsure.
663 config RCU_KTHREAD_PRIO
664 int "Real-time priority to use for RCU worker threads"
665 range 1 99 if RCU_BOOST
666 range 0 99 if !RCU_BOOST
667 default 1 if RCU_BOOST
668 default 0 if !RCU_BOOST
669 depends on RCU_EXPERT
671 This option specifies the SCHED_FIFO priority value that will be
672 assigned to the rcuc/n and rcub/n threads and is also the value
673 used for RCU_BOOST (if enabled). If you are working with a
674 real-time application that has one or more CPU-bound threads
675 running at a real-time priority level, you should set
676 RCU_KTHREAD_PRIO to a priority higher than the highest-priority
677 real-time CPU-bound application thread. The default RCU_KTHREAD_PRIO
678 value of 1 is appropriate in the common case, which is real-time
679 applications that do not have any CPU-bound threads.
681 Some real-time applications might not have a single real-time
682 thread that saturates a given CPU, but instead might have
683 multiple real-time threads that, taken together, fully utilize
684 that CPU. In this case, you should set RCU_KTHREAD_PRIO to
685 a priority higher than the lowest-priority thread that is
686 conspiring to prevent the CPU from running any non-real-time
687 tasks. For example, if one thread at priority 10 and another
688 thread at priority 5 are between themselves fully consuming
689 the CPU time on a given CPU, then RCU_KTHREAD_PRIO should be
690 set to priority 6 or higher.
692 Specify the real-time priority, or take the default if unsure.
694 config RCU_BOOST_DELAY
695 int "Milliseconds to delay boosting after RCU grace-period start"
700 This option specifies the time to wait after the beginning of
701 a given grace period before priority-boosting preempted RCU
702 readers blocking that grace period. Note that any RCU reader
703 blocking an expedited RCU grace period is boosted immediately.
705 Accept the default if unsure.
708 bool "Offload RCU callback processing from boot-selected CPUs"
709 depends on TREE_RCU || PREEMPT_RCU
710 depends on RCU_EXPERT || NO_HZ_FULL
713 Use this option to reduce OS jitter for aggressive HPC or
714 real-time workloads. It can also be used to offload RCU
715 callback invocation to energy-efficient CPUs in battery-powered
716 asymmetric multiprocessors.
718 This option offloads callback invocation from the set of
719 CPUs specified at boot time by the rcu_nocbs parameter.
720 For each such CPU, a kthread ("rcuox/N") will be created to
721 invoke callbacks, where the "N" is the CPU being offloaded,
722 and where the "x" is "b" for RCU-bh, "p" for RCU-preempt, and
723 "s" for RCU-sched. Nothing prevents this kthread from running
724 on the specified CPUs, but (1) the kthreads may be preempted
725 between each callback, and (2) affinity or cgroups can be used
726 to force the kthreads to run on whatever set of CPUs is desired.
728 Say Y here if you want to help to debug reduced OS jitter.
729 Say N here if you are unsure.
732 prompt "Build-forced no-CBs CPUs"
733 default RCU_NOCB_CPU_NONE
734 depends on RCU_NOCB_CPU
736 This option allows no-CBs CPUs (whose RCU callbacks are invoked
737 from kthreads rather than from softirq context) to be specified
738 at build time. Additional no-CBs CPUs may be specified by
739 the rcu_nocbs= boot parameter.
741 config RCU_NOCB_CPU_NONE
742 bool "No build_forced no-CBs CPUs"
744 This option does not force any of the CPUs to be no-CBs CPUs.
745 Only CPUs designated by the rcu_nocbs= boot parameter will be
746 no-CBs CPUs, whose RCU callbacks will be invoked by per-CPU
747 kthreads whose names begin with "rcuo". All other CPUs will
748 invoke their own RCU callbacks in softirq context.
750 Select this option if you want to choose no-CBs CPUs at
751 boot time, for example, to allow testing of different no-CBs
752 configurations without having to rebuild the kernel each time.
754 config RCU_NOCB_CPU_ZERO
755 bool "CPU 0 is a build_forced no-CBs CPU"
757 This option forces CPU 0 to be a no-CBs CPU, so that its RCU
758 callbacks are invoked by a per-CPU kthread whose name begins
759 with "rcuo". Additional CPUs may be designated as no-CBs
760 CPUs using the rcu_nocbs= boot parameter will be no-CBs CPUs.
761 All other CPUs will invoke their own RCU callbacks in softirq
764 Select this if CPU 0 needs to be a no-CBs CPU for real-time
765 or energy-efficiency reasons, but the real reason it exists
766 is to ensure that randconfig testing covers mixed systems.
768 config RCU_NOCB_CPU_ALL
769 bool "All CPUs are build_forced no-CBs CPUs"
771 This option forces all CPUs to be no-CBs CPUs. The rcu_nocbs=
772 boot parameter will be ignored. All CPUs' RCU callbacks will
773 be executed in the context of per-CPU rcuo kthreads created for
774 this purpose. Assuming that the kthreads whose names start with
775 "rcuo" are bound to "housekeeping" CPUs, this reduces OS jitter
776 on the remaining CPUs, but might decrease memory locality during
777 RCU-callback invocation, thus potentially degrading throughput.
779 Select this if all CPUs need to be no-CBs CPUs for real-time
780 or energy-efficiency reasons.
784 config RCU_EXPEDITE_BOOT
788 This option enables expedited grace periods at boot time,
789 as if rcu_expedite_gp() had been invoked early in boot.
790 The corresponding rcu_unexpedite_gp() is invoked from
791 rcu_end_inkernel_boot(), which is intended to be invoked
792 at the end of the kernel-only boot sequence, just before
795 Accept the default if unsure.
797 endmenu # "RCU Subsystem"
804 tristate "Kernel .config support"
807 This option enables the complete Linux kernel ".config" file
808 contents to be saved in the kernel. It provides documentation
809 of which kernel options are used in a running kernel or in an
810 on-disk kernel. This information can be extracted from the kernel
811 image file with the script scripts/extract-ikconfig and used as
812 input to rebuild the current kernel or to build another kernel.
813 It can also be extracted from a running kernel by reading
814 /proc/config.gz if enabled (below).
817 bool "Enable access to .config through /proc/config.gz"
818 depends on IKCONFIG && PROC_FS
820 This option enables access to the kernel configuration file
821 through /proc/config.gz.
824 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
829 Select the minimal kernel log buffer size as a power of 2.
830 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config
831 parameter, see below. Any higher size also might be forced
832 by "log_buf_len" boot parameter.
842 config LOG_CPU_MAX_BUF_SHIFT
843 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)"
846 default 12 if !BASE_SMALL
847 default 0 if BASE_SMALL
850 This option allows to increase the default ring buffer size
851 according to the number of CPUs. The value defines the contribution
852 of each CPU as a power of 2. The used space is typically only few
853 lines however it might be much more when problems are reported,
856 The increased size means that a new buffer has to be allocated and
857 the original static one is unused. It makes sense only on systems
858 with more CPUs. Therefore this value is used only when the sum of
859 contributions is greater than the half of the default kernel ring
860 buffer as defined by LOG_BUF_SHIFT. The default values are set
861 so that more than 64 CPUs are needed to trigger the allocation.
863 Also this option is ignored when "log_buf_len" kernel parameter is
864 used as it forces an exact (power of two) size of the ring buffer.
866 The number of possible CPUs is used for this computation ignoring
867 hotplugging making the computation optimal for the worst case
868 scenario while allowing a simple algorithm to be used from bootup.
870 Examples shift values and their meaning:
871 17 => 128 KB for each CPU
872 16 => 64 KB for each CPU
873 15 => 32 KB for each CPU
874 14 => 16 KB for each CPU
875 13 => 8 KB for each CPU
876 12 => 4 KB for each CPU
878 config NMI_LOG_BUF_SHIFT
879 int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)"
882 depends on PRINTK_NMI
884 Select the size of a per-CPU buffer where NMI messages are temporary
885 stored. They are copied to the main log buffer in a safe context
886 to avoid a deadlock. The value defines the size as a power of 2.
888 NMI messages are rare and limited. The largest one is when
889 a backtrace is printed. It usually fits into 4KB. Select
890 8KB if you want to be on the safe side.
893 17 => 128 KB for each CPU
894 16 => 64 KB for each CPU
895 15 => 32 KB for each CPU
896 14 => 16 KB for each CPU
897 13 => 8 KB for each CPU
898 12 => 4 KB for each CPU
901 # Architectures with an unreliable sched_clock() should select this:
903 config HAVE_UNSTABLE_SCHED_CLOCK
906 config GENERIC_SCHED_CLOCK
910 # For architectures that want to enable the support for NUMA-affine scheduler
913 config ARCH_SUPPORTS_NUMA_BALANCING
917 # For architectures that prefer to flush all TLBs after a number of pages
918 # are unmapped instead of sending one IPI per page to flush. The architecture
919 # must provide guarantees on what happens if a clean TLB cache entry is
920 # written after the unmap. Details are in mm/rmap.c near the check for
921 # should_defer_flush. The architecture should also consider if the full flush
922 # and the refill costs are offset by the savings of sending fewer IPIs.
923 config ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
927 # For architectures that know their GCC __int128 support is sound
929 config ARCH_SUPPORTS_INT128
932 # For architectures that (ab)use NUMA to represent different memory regions
933 # all cpu-local but of different latencies, such as SuperH.
935 config ARCH_WANT_NUMA_VARIABLE_LOCALITY
938 config NUMA_BALANCING
939 bool "Memory placement aware NUMA scheduler"
940 depends on ARCH_SUPPORTS_NUMA_BALANCING
941 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
942 depends on SMP && NUMA && MIGRATION
944 This option adds support for automatic NUMA aware memory/task placement.
945 The mechanism is quite primitive and is based on migrating memory when
946 it has references to the node the task is running on.
948 This system will be inactive on UMA systems.
950 config NUMA_BALANCING_DEFAULT_ENABLED
951 bool "Automatically enable NUMA aware memory/task placement"
953 depends on NUMA_BALANCING
955 If set, automatic NUMA balancing will be enabled if running on a NUMA
959 bool "Control Group support"
962 This option adds support for grouping sets of processes together, for
963 use with process control subsystems such as Cpusets, CFS, memory
964 controls or device isolation.
966 - Documentation/scheduler/sched-design-CFS.txt (CFS)
967 - Documentation/cgroup-v1/ (features for grouping, isolation
968 and resource control)
978 bool "Memory controller"
982 Provides control over the memory footprint of tasks in a cgroup.
985 bool "Swap controller"
986 depends on MEMCG && SWAP
988 Provides control over the swap space consumed by tasks in a cgroup.
990 config MEMCG_SWAP_ENABLED
991 bool "Swap controller enabled by default"
992 depends on MEMCG_SWAP
995 Memory Resource Controller Swap Extension comes with its price in
996 a bigger memory consumption. General purpose distribution kernels
997 which want to enable the feature but keep it disabled by default
998 and let the user enable it by swapaccount=1 boot command line
999 parameter should have this option unselected.
1000 For those who want to have the feature enabled by default should
1001 select this option (if, for some reason, they need to disable it
1002 then swapaccount=0 does the trick).
1005 bool "IO controller"
1009 Generic block IO controller cgroup interface. This is the common
1010 cgroup interface which should be used by various IO controlling
1013 Currently, CFQ IO scheduler uses it to recognize task groups and
1014 control disk bandwidth allocation (proportional time slice allocation)
1015 to such task groups. It is also used by bio throttling logic in
1016 block layer to implement upper limit in IO rates on a device.
1018 This option only enables generic Block IO controller infrastructure.
1019 One needs to also enable actual IO controlling logic/policy. For
1020 enabling proportional weight division of disk bandwidth in CFQ, set
1021 CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set
1022 CONFIG_BLK_DEV_THROTTLING=y.
1024 See Documentation/cgroup-v1/blkio-controller.txt for more information.
1026 config DEBUG_BLK_CGROUP
1027 bool "IO controller debugging"
1028 depends on BLK_CGROUP
1031 Enable some debugging help. Currently it exports additional stat
1032 files in a cgroup which can be useful for debugging.
1034 config CGROUP_WRITEBACK
1036 depends on MEMCG && BLK_CGROUP
1039 menuconfig CGROUP_SCHED
1040 bool "CPU controller"
1043 This feature lets CPU scheduler recognize task groups and control CPU
1044 bandwidth allocation to such task groups. It uses cgroups to group
1048 config FAIR_GROUP_SCHED
1049 bool "Group scheduling for SCHED_OTHER"
1050 depends on CGROUP_SCHED
1051 default CGROUP_SCHED
1053 config CFS_BANDWIDTH
1054 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED"
1055 depends on FAIR_GROUP_SCHED
1058 This option allows users to define CPU bandwidth rates (limits) for
1059 tasks running within the fair group scheduler. Groups with no limit
1060 set are considered to be unconstrained and will run with no
1062 See tip/Documentation/scheduler/sched-bwc.txt for more information.
1064 config RT_GROUP_SCHED
1065 bool "Group scheduling for SCHED_RR/FIFO"
1066 depends on CGROUP_SCHED
1069 This feature lets you explicitly allocate real CPU bandwidth
1070 to task groups. If enabled, it will also make it impossible to
1071 schedule realtime tasks for non-root users until you allocate
1072 realtime bandwidth for them.
1073 See Documentation/scheduler/sched-rt-group.txt for more information.
1078 bool "PIDs controller"
1080 Provides enforcement of process number limits in the scope of a
1081 cgroup. Any attempt to fork more processes than is allowed in the
1082 cgroup will fail. PIDs are fundamentally a global resource because it
1083 is fairly trivial to reach PID exhaustion before you reach even a
1084 conservative kmemcg limit. As a result, it is possible to grind a
1085 system to halt without being limited by other cgroup policies. The
1086 PIDs controller is designed to stop this from happening.
1088 It should be noted that organisational operations (such as attaching
1089 to a cgroup hierarchy will *not* be blocked by the PIDs controller),
1090 since the PIDs limit only affects a process's ability to fork, not to
1093 config CGROUP_FREEZER
1094 bool "Freezer controller"
1096 Provides a way to freeze and unfreeze all tasks in a
1099 This option affects the ORIGINAL cgroup interface. The cgroup2 memory
1100 controller includes important in-kernel memory consumers per default.
1102 If you're using cgroup2, say N.
1104 config CGROUP_HUGETLB
1105 bool "HugeTLB controller"
1106 depends on HUGETLB_PAGE
1110 Provides a cgroup controller for HugeTLB pages.
1111 When you enable this, you can put a per cgroup limit on HugeTLB usage.
1112 The limit is enforced during page fault. Since HugeTLB doesn't
1113 support page reclaim, enforcing the limit at page fault time implies
1114 that, the application will get SIGBUS signal if it tries to access
1115 HugeTLB pages beyond its limit. This requires the application to know
1116 beforehand how much HugeTLB pages it would require for its use. The
1117 control group is tracked in the third page lru pointer. This means
1118 that we cannot use the controller with huge page less than 3 pages.
1121 bool "Cpuset controller"
1123 This option will let you create and manage CPUSETs which
1124 allow dynamically partitioning a system into sets of CPUs and
1125 Memory Nodes and assigning tasks to run only within those sets.
1126 This is primarily useful on large SMP or NUMA systems.
1130 config PROC_PID_CPUSET
1131 bool "Include legacy /proc/<pid>/cpuset file"
1135 config CGROUP_DEVICE
1136 bool "Device controller"
1138 Provides a cgroup controller implementing whitelists for
1139 devices which a process in the cgroup can mknod or open.
1141 config CGROUP_CPUACCT
1142 bool "Simple CPU accounting controller"
1144 Provides a simple controller for monitoring the
1145 total CPU consumed by the tasks in a cgroup.
1148 bool "Perf controller"
1149 depends on PERF_EVENTS
1151 This option extends the perf per-cpu mode to restrict monitoring
1152 to threads which belong to the cgroup specified and run on the
1158 bool "Support for eBPF programs attached to cgroups"
1159 depends on BPF_SYSCALL
1160 select SOCK_CGROUP_DATA
1162 Allow attaching eBPF programs to a cgroup using the bpf(2)
1163 syscall command BPF_PROG_ATTACH.
1165 In which context these programs are accessed depends on the type
1166 of attachment. For instance, programs that are attached using
1167 BPF_CGROUP_INET_INGRESS will be executed on the ingress path of
1171 bool "Example controller"
1174 This option enables a simple controller that exports
1175 debugging information about the cgroups framework.
1181 config CHECKPOINT_RESTORE
1182 bool "Checkpoint/restore support" if EXPERT
1183 select PROC_CHILDREN
1186 Enables additional kernel features in a sake of checkpoint/restore.
1187 In particular it adds auxiliary prctl codes to setup process text,
1188 data and heap segment sizes, and a few additional /proc filesystem
1191 If unsure, say N here.
1193 menuconfig NAMESPACES
1194 bool "Namespaces support" if EXPERT
1195 depends on MULTIUSER
1198 Provides the way to make tasks work with different objects using
1199 the same id. For example same IPC id may refer to different objects
1200 or same user id or pid may refer to different tasks when used in
1201 different namespaces.
1206 bool "UTS namespace"
1209 In this namespace tasks see different info provided with the
1213 bool "IPC namespace"
1214 depends on (SYSVIPC || POSIX_MQUEUE)
1217 In this namespace tasks work with IPC ids which correspond to
1218 different IPC objects in different namespaces.
1221 bool "User namespace"
1224 This allows containers, i.e. vservers, to use user namespaces
1225 to provide different user info for different servers.
1227 When user namespaces are enabled in the kernel it is
1228 recommended that the MEMCG option also be enabled and that
1229 user-space use the memory control groups to limit the amount
1230 of memory a memory unprivileged users can use.
1235 bool "PID Namespaces"
1238 Support process id namespaces. This allows having multiple
1239 processes with the same pid as long as they are in different
1240 pid namespaces. This is a building block of containers.
1243 bool "Network namespace"
1247 Allow user space to create what appear to be multiple instances
1248 of the network stack.
1252 config SCHED_AUTOGROUP
1253 bool "Automatic process group scheduling"
1256 select FAIR_GROUP_SCHED
1258 This option optimizes the scheduler for common desktop workloads by
1259 automatically creating and populating task groups. This separation
1260 of workloads isolates aggressive CPU burners (like build jobs) from
1261 desktop applications. Task group autogeneration is currently based
1264 config SYSFS_DEPRECATED
1265 bool "Enable deprecated sysfs features to support old userspace tools"
1269 This option adds code that switches the layout of the "block" class
1270 devices, to not show up in /sys/class/block/, but only in
1273 This switch is only active when the sysfs.deprecated=1 boot option is
1274 passed or the SYSFS_DEPRECATED_V2 option is set.
1276 This option allows new kernels to run on old distributions and tools,
1277 which might get confused by /sys/class/block/. Since 2007/2008 all
1278 major distributions and tools handle this just fine.
1280 Recent distributions and userspace tools after 2009/2010 depend on
1281 the existence of /sys/class/block/, and will not work with this
1284 Only if you are using a new kernel on an old distribution, you might
1287 config SYSFS_DEPRECATED_V2
1288 bool "Enable deprecated sysfs features by default"
1291 depends on SYSFS_DEPRECATED
1293 Enable deprecated sysfs by default.
1295 See the CONFIG_SYSFS_DEPRECATED option for more details about this
1298 Only if you are using a new kernel on an old distribution, you might
1299 need to say Y here. Even then, odds are you would not need it
1300 enabled, you can always pass the boot option if absolutely necessary.
1303 bool "Kernel->user space relay support (formerly relayfs)"
1306 This option enables support for relay interface support in
1307 certain file systems (such as debugfs).
1308 It is designed to provide an efficient mechanism for tools and
1309 facilities to relay large amounts of data from kernel space to
1314 config BLK_DEV_INITRD
1315 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
1316 depends on BROKEN || !FRV
1318 The initial RAM filesystem is a ramfs which is loaded by the
1319 boot loader (loadlin or lilo) and that is mounted as root
1320 before the normal boot procedure. It is typically used to
1321 load modules needed to mount the "real" root file system,
1322 etc. See <file:Documentation/admin-guide/initrd.rst> for details.
1324 If RAM disk support (BLK_DEV_RAM) is also included, this
1325 also enables initial RAM disk (initrd) support and adds
1326 15 Kbytes (more on some other architectures) to the kernel size.
1332 source "usr/Kconfig"
1337 prompt "Compiler optimization level"
1338 default CONFIG_CC_OPTIMIZE_FOR_PERFORMANCE
1340 config CC_OPTIMIZE_FOR_PERFORMANCE
1341 bool "Optimize for performance"
1343 This is the default optimization level for the kernel, building
1344 with the "-O2" compiler flag for best performance and most
1345 helpful compile-time warnings.
1347 config CC_OPTIMIZE_FOR_SIZE
1348 bool "Optimize for size"
1350 Enabling this option will pass "-Os" instead of "-O2" to
1351 your compiler resulting in a smaller kernel.
1366 config SYSCTL_EXCEPTION_TRACE
1369 Enable support for /proc/sys/debug/exception-trace.
1371 config SYSCTL_ARCH_UNALIGN_NO_WARN
1374 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap
1375 Allows arch to define/use @no_unaligned_warning to possibly warn
1376 about unaligned access emulation going on under the hood.
1378 config SYSCTL_ARCH_UNALIGN_ALLOW
1381 Enable support for /proc/sys/kernel/unaligned-trap
1382 Allows arches to define/use @unaligned_enabled to runtime toggle
1383 the unaligned access emulation.
1384 see arch/parisc/kernel/unaligned.c for reference
1386 config HAVE_PCSPKR_PLATFORM
1389 # interpreter that classic socket filters depend on
1394 bool "Configure standard kernel features (expert users)"
1395 # Unhide debug options, to make the on-by-default options visible
1398 This option allows certain base kernel options and settings
1399 to be disabled or tweaked. This is for specialized
1400 environments which can tolerate a "non-standard" kernel.
1401 Only use this if you really know what you are doing.
1404 bool "Enable 16-bit UID system calls" if EXPERT
1405 depends on HAVE_UID16 && MULTIUSER
1408 This enables the legacy 16-bit UID syscall wrappers.
1411 bool "Multiple users, groups and capabilities support" if EXPERT
1414 This option enables support for non-root users, groups and
1417 If you say N here, all processes will run with UID 0, GID 0, and all
1418 possible capabilities. Saying N here also compiles out support for
1419 system calls related to UIDs, GIDs, and capabilities, such as setuid,
1422 If unsure, say Y here.
1424 config SGETMASK_SYSCALL
1425 bool "sgetmask/ssetmask syscalls support" if EXPERT
1426 def_bool PARISC || MN10300 || BLACKFIN || M68K || PPC || MIPS || X86 || SPARC || CRIS || MICROBLAZE || SUPERH
1428 sys_sgetmask and sys_ssetmask are obsolete system calls
1429 no longer supported in libc but still enabled by default in some
1432 If unsure, leave the default option here.
1434 config SYSFS_SYSCALL
1435 bool "Sysfs syscall support" if EXPERT
1438 sys_sysfs is an obsolete system call no longer supported in libc.
1439 Note that disabling this option is more secure but might break
1440 compatibility with some systems.
1442 If unsure say Y here.
1444 config SYSCTL_SYSCALL
1445 bool "Sysctl syscall support" if EXPERT
1446 depends on PROC_SYSCTL
1450 sys_sysctl uses binary paths that have been found challenging
1451 to properly maintain and use. The interface in /proc/sys
1452 using paths with ascii names is now the primary path to this
1455 Almost nothing using the binary sysctl interface so if you are
1456 trying to save some space it is probably safe to disable this,
1457 making your kernel marginally smaller.
1459 If unsure say N here.
1462 bool "Posix Clocks & timers" if EXPERT
1465 This includes native support for POSIX timers to the kernel.
1466 Some embedded systems have no use for them and therefore they
1467 can be configured out to reduce the size of the kernel image.
1469 When this option is disabled, the following syscalls won't be
1470 available: timer_create, timer_gettime: timer_getoverrun,
1471 timer_settime, timer_delete, clock_adjtime, getitimer,
1472 setitimer, alarm. Furthermore, the clock_settime, clock_gettime,
1473 clock_getres and clock_nanosleep syscalls will be limited to
1474 CLOCK_REALTIME, CLOCK_MONOTONIC and CLOCK_BOOTTIME only.
1479 bool "Load all symbols for debugging/ksymoops" if EXPERT
1482 Say Y here to let the kernel print out symbolic crash information and
1483 symbolic stack backtraces. This increases the size of the kernel
1484 somewhat, as all symbols have to be loaded into the kernel image.
1487 bool "Include all symbols in kallsyms"
1488 depends on DEBUG_KERNEL && KALLSYMS
1490 Normally kallsyms only contains the symbols of functions for nicer
1491 OOPS messages and backtraces (i.e., symbols from the text and inittext
1492 sections). This is sufficient for most cases. And only in very rare
1493 cases (e.g., when a debugger is used) all symbols are required (e.g.,
1494 names of variables from the data sections, etc).
1496 This option makes sure that all symbols are loaded into the kernel
1497 image (i.e., symbols from all sections) in cost of increased kernel
1498 size (depending on the kernel configuration, it may be 300KiB or
1499 something like this).
1501 Say N unless you really need all symbols.
1503 config KALLSYMS_ABSOLUTE_PERCPU
1506 default X86_64 && SMP
1508 config KALLSYMS_BASE_RELATIVE
1511 default !IA64 && !(TILE && 64BIT)
1513 Instead of emitting them as absolute values in the native word size,
1514 emit the symbol references in the kallsyms table as 32-bit entries,
1515 each containing a relative value in the range [base, base + U32_MAX]
1516 or, when KALLSYMS_ABSOLUTE_PERCPU is in effect, each containing either
1517 an absolute value in the range [0, S32_MAX] or a relative value in the
1518 range [base, base + S32_MAX], where base is the lowest relative symbol
1519 address encountered in the image.
1521 On 64-bit builds, this reduces the size of the address table by 50%,
1522 but more importantly, it results in entries whose values are build
1523 time constants, and no relocation pass is required at runtime to fix
1524 up the entries based on the runtime load address of the kernel.
1528 bool "Enable support for printk" if EXPERT
1531 This option enables normal printk support. Removing it
1532 eliminates most of the message strings from the kernel image
1533 and makes the kernel more or less silent. As this makes it
1534 very difficult to diagnose system problems, saying N here is
1535 strongly discouraged.
1543 bool "BUG() support" if EXPERT
1546 Disabling this option eliminates support for BUG and WARN, reducing
1547 the size of your kernel image and potentially quietly ignoring
1548 numerous fatal conditions. You should only consider disabling this
1549 option for embedded systems with no facilities for reporting errors.
1555 bool "Enable ELF core dumps" if EXPERT
1557 Enable support for generating core dumps. Disabling saves about 4k.
1560 config PCSPKR_PLATFORM
1561 bool "Enable PC-Speaker support" if EXPERT
1562 depends on HAVE_PCSPKR_PLATFORM
1566 This option allows to disable the internal PC-Speaker
1567 support, saving some memory.
1571 bool "Enable full-sized data structures for core" if EXPERT
1573 Disabling this option reduces the size of miscellaneous core
1574 kernel data structures. This saves memory on small machines,
1575 but may reduce performance.
1578 bool "Enable futex support" if EXPERT
1582 Disabling this option will cause the kernel to be built without
1583 support for "fast userspace mutexes". The resulting kernel may not
1584 run glibc-based applications correctly.
1586 config HAVE_FUTEX_CMPXCHG
1590 Architectures should select this if futex_atomic_cmpxchg_inatomic()
1591 is implemented and always working. This removes a couple of runtime
1595 bool "Enable eventpoll support" if EXPERT
1599 Disabling this option will cause the kernel to be built without
1600 support for epoll family of system calls.
1603 bool "Enable signalfd() system call" if EXPERT
1607 Enable the signalfd() system call that allows to receive signals
1608 on a file descriptor.
1613 bool "Enable timerfd() system call" if EXPERT
1617 Enable the timerfd() system call that allows to receive timer
1618 events on a file descriptor.
1623 bool "Enable eventfd() system call" if EXPERT
1627 Enable the eventfd() system call that allows to receive both
1628 kernel notification (ie. KAIO) or userspace notifications.
1632 # syscall, maps, verifier
1634 bool "Enable bpf() system call"
1639 Enable the bpf() system call that allows to manipulate eBPF
1640 programs and maps via file descriptors.
1643 bool "Use full shmem filesystem" if EXPERT
1647 The shmem is an internal filesystem used to manage shared memory.
1648 It is backed by swap and manages resource limits. It is also exported
1649 to userspace as tmpfs if TMPFS is enabled. Disabling this
1650 option replaces shmem and tmpfs with the much simpler ramfs code,
1651 which may be appropriate on small systems without swap.
1654 bool "Enable AIO support" if EXPERT
1657 This option enables POSIX asynchronous I/O which may by used
1658 by some high performance threaded applications. Disabling
1659 this option saves about 7k.
1661 config ADVISE_SYSCALLS
1662 bool "Enable madvise/fadvise syscalls" if EXPERT
1665 This option enables the madvise and fadvise syscalls, used by
1666 applications to advise the kernel about their future memory or file
1667 usage, improving performance. If building an embedded system where no
1668 applications use these syscalls, you can disable this option to save
1672 bool "Enable userfaultfd() system call"
1676 Enable the userfaultfd() system call that allows to intercept and
1677 handle page faults in userland.
1681 bool "Enable PCI quirk workarounds" if EXPERT
1684 This enables workarounds for various PCI chipset
1685 bugs/quirks. Disable this only if your target machine is
1686 unaffected by PCI quirks.
1689 bool "Enable membarrier() system call" if EXPERT
1692 Enable the membarrier() system call that allows issuing memory
1693 barriers across all running threads, which can be used to distribute
1694 the cost of user-space memory barriers asymmetrically by transforming
1695 pairs of memory barriers into pairs consisting of membarrier() and a
1701 bool "Embedded system"
1702 option allnoconfig_y
1705 This option should be enabled if compiling the kernel for
1706 an embedded system so certain expert options are available
1709 config HAVE_PERF_EVENTS
1712 See tools/perf/design.txt for details.
1714 config PERF_USE_VMALLOC
1717 See tools/perf/design.txt for details
1719 menu "Kernel Performance Events And Counters"
1722 bool "Kernel performance events and counters"
1723 default y if PROFILING
1724 depends on HAVE_PERF_EVENTS
1729 Enable kernel support for various performance events provided
1730 by software and hardware.
1732 Software events are supported either built-in or via the
1733 use of generic tracepoints.
1735 Most modern CPUs support performance events via performance
1736 counter registers. These registers count the number of certain
1737 types of hw events: such as instructions executed, cachemisses
1738 suffered, or branches mis-predicted - without slowing down the
1739 kernel or applications. These registers can also trigger interrupts
1740 when a threshold number of events have passed - and can thus be
1741 used to profile the code that runs on that CPU.
1743 The Linux Performance Event subsystem provides an abstraction of
1744 these software and hardware event capabilities, available via a
1745 system call and used by the "perf" utility in tools/perf/. It
1746 provides per task and per CPU counters, and it provides event
1747 capabilities on top of those.
1751 config DEBUG_PERF_USE_VMALLOC
1753 bool "Debug: use vmalloc to back perf mmap() buffers"
1754 depends on PERF_EVENTS && DEBUG_KERNEL && !PPC
1755 select PERF_USE_VMALLOC
1757 Use vmalloc memory to back perf mmap() buffers.
1759 Mostly useful for debugging the vmalloc code on platforms
1760 that don't require it.
1766 config VM_EVENT_COUNTERS
1768 bool "Enable VM event counters for /proc/vmstat" if EXPERT
1770 VM event counters are needed for event counts to be shown.
1771 This option allows the disabling of the VM event counters
1772 on EXPERT systems. /proc/vmstat will only show page counts
1773 if VM event counters are disabled.
1777 bool "Enable SLUB debugging support" if EXPERT
1778 depends on SLUB && SYSFS
1780 SLUB has extensive debug support features. Disabling these can
1781 result in significant savings in code size. This also disables
1782 SLUB sysfs support. /sys/slab will not exist and there will be
1783 no support for cache validation etc.
1786 bool "Disable heap randomization"
1789 Randomizing heap placement makes heap exploits harder, but it
1790 also breaks ancient binaries (including anything libc5 based).
1791 This option changes the bootup default to heap randomization
1792 disabled, and can be overridden at runtime by setting
1793 /proc/sys/kernel/randomize_va_space to 2.
1795 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1798 prompt "Choose SLAB allocator"
1801 This option allows to select a slab allocator.
1805 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1807 The regular slab allocator that is established and known to work
1808 well in all environments. It organizes cache hot objects in
1809 per cpu and per node queues.
1812 bool "SLUB (Unqueued Allocator)"
1813 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1815 SLUB is a slab allocator that minimizes cache line usage
1816 instead of managing queues of cached objects (SLAB approach).
1817 Per cpu caching is realized using slabs of objects instead
1818 of queues of objects. SLUB can use memory efficiently
1819 and has enhanced diagnostics. SLUB is the default choice for
1824 bool "SLOB (Simple Allocator)"
1826 SLOB replaces the stock allocator with a drastically simpler
1827 allocator. SLOB is generally more space efficient but
1828 does not perform as well on large systems.
1832 config SLAB_FREELIST_RANDOM
1834 depends on SLAB || SLUB
1835 bool "SLAB freelist randomization"
1837 Randomizes the freelist order used on creating new pages. This
1838 security feature reduces the predictability of the kernel slab
1839 allocator against heap overflows.
1841 config SLUB_CPU_PARTIAL
1843 depends on SLUB && SMP
1844 bool "SLUB per cpu partial cache"
1846 Per cpu partial caches accellerate objects allocation and freeing
1847 that is local to a processor at the price of more indeterminism
1848 in the latency of the free. On overflow these caches will be cleared
1849 which requires the taking of locks that may cause latency spikes.
1850 Typically one would choose no for a realtime system.
1852 config MMAP_ALLOW_UNINITIALIZED
1853 bool "Allow mmapped anonymous memory to be uninitialized"
1854 depends on EXPERT && !MMU
1857 Normally, and according to the Linux spec, anonymous memory obtained
1858 from mmap() has it's contents cleared before it is passed to
1859 userspace. Enabling this config option allows you to request that
1860 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
1861 providing a huge performance boost. If this option is not enabled,
1862 then the flag will be ignored.
1864 This is taken advantage of by uClibc's malloc(), and also by
1865 ELF-FDPIC binfmt's brk and stack allocator.
1867 Because of the obvious security issues, this option should only be
1868 enabled on embedded devices where you control what is run in
1869 userspace. Since that isn't generally a problem on no-MMU systems,
1870 it is normally safe to say Y here.
1872 See Documentation/nommu-mmap.txt for more information.
1874 config SYSTEM_DATA_VERIFICATION
1876 select SYSTEM_TRUSTED_KEYRING
1880 select ASYMMETRIC_KEY_TYPE
1881 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE
1884 select X509_CERTIFICATE_PARSER
1885 select PKCS7_MESSAGE_PARSER
1887 Provide PKCS#7 message verification using the contents of the system
1888 trusted keyring to provide public keys. This then can be used for
1889 module verification, kexec image verification and firmware blob
1893 bool "Profiling support"
1895 Say Y here to enable the extended profiling support mechanisms used
1896 by profilers such as OProfile.
1899 # Place an empty function call at each tracepoint site. Can be
1900 # dynamically changed for a probe function.
1905 source "arch/Kconfig"
1907 endmenu # General setup
1909 config HAVE_GENERIC_DMA_COHERENT
1916 depends on SLAB || SLUB_DEBUG
1924 default 0 if BASE_FULL
1925 default 1 if !BASE_FULL
1928 bool "Enable loadable module support"
1931 Kernel modules are small pieces of compiled code which can
1932 be inserted in the running kernel, rather than being
1933 permanently built into the kernel. You use the "modprobe"
1934 tool to add (and sometimes remove) them. If you say Y here,
1935 many parts of the kernel can be built as modules (by
1936 answering M instead of Y where indicated): this is most
1937 useful for infrequently used options which are not required
1938 for booting. For more information, see the man pages for
1939 modprobe, lsmod, modinfo, insmod and rmmod.
1941 If you say Y here, you will need to run "make
1942 modules_install" to put the modules under /lib/modules/
1943 where modprobe can find them (you may need to be root to do
1950 config MODULE_FORCE_LOAD
1951 bool "Forced module loading"
1954 Allow loading of modules without version information (ie. modprobe
1955 --force). Forced module loading sets the 'F' (forced) taint flag and
1956 is usually a really bad idea.
1958 config MODULE_UNLOAD
1959 bool "Module unloading"
1961 Without this option you will not be able to unload any
1962 modules (note that some modules may not be unloadable
1963 anyway), which makes your kernel smaller, faster
1964 and simpler. If unsure, say Y.
1966 config MODULE_FORCE_UNLOAD
1967 bool "Forced module unloading"
1968 depends on MODULE_UNLOAD
1970 This option allows you to force a module to unload, even if the
1971 kernel believes it is unsafe: the kernel will remove the module
1972 without waiting for anyone to stop using it (using the -f option to
1973 rmmod). This is mainly for kernel developers and desperate users.
1977 bool "Module versioning support"
1979 Usually, you have to use modules compiled with your kernel.
1980 Saying Y here makes it sometimes possible to use modules
1981 compiled for different kernels, by adding enough information
1982 to the modules to (hopefully) spot any changes which would
1983 make them incompatible with the kernel you are running. If
1986 config MODULE_SRCVERSION_ALL
1987 bool "Source checksum for all modules"
1989 Modules which contain a MODULE_VERSION get an extra "srcversion"
1990 field inserted into their modinfo section, which contains a
1991 sum of the source files which made it. This helps maintainers
1992 see exactly which source was used to build a module (since
1993 others sometimes change the module source without updating
1994 the version). With this option, such a "srcversion" field
1995 will be created for all modules. If unsure, say N.
1998 bool "Module signature verification"
2000 select SYSTEM_DATA_VERIFICATION
2002 Check modules for valid signatures upon load: the signature
2003 is simply appended to the module. For more information see
2004 Documentation/module-signing.txt.
2006 Note that this option adds the OpenSSL development packages as a
2007 kernel build dependency so that the signing tool can use its crypto
2010 !!!WARNING!!! If you enable this option, you MUST make sure that the
2011 module DOES NOT get stripped after being signed. This includes the
2012 debuginfo strip done by some packagers (such as rpmbuild) and
2013 inclusion into an initramfs that wants the module size reduced.
2015 config MODULE_SIG_FORCE
2016 bool "Require modules to be validly signed"
2017 depends on MODULE_SIG
2019 Reject unsigned modules or signed modules for which we don't have a
2020 key. Without this, such modules will simply taint the kernel.
2022 config MODULE_SIG_ALL
2023 bool "Automatically sign all modules"
2025 depends on MODULE_SIG
2027 Sign all modules during make modules_install. Without this option,
2028 modules must be signed manually, using the scripts/sign-file tool.
2030 comment "Do not forget to sign required modules with scripts/sign-file"
2031 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL
2034 prompt "Which hash algorithm should modules be signed with?"
2035 depends on MODULE_SIG
2037 This determines which sort of hashing algorithm will be used during
2038 signature generation. This algorithm _must_ be built into the kernel
2039 directly so that signature verification can take place. It is not
2040 possible to load a signed module containing the algorithm to check
2041 the signature on that module.
2043 config MODULE_SIG_SHA1
2044 bool "Sign modules with SHA-1"
2047 config MODULE_SIG_SHA224
2048 bool "Sign modules with SHA-224"
2049 select CRYPTO_SHA256
2051 config MODULE_SIG_SHA256
2052 bool "Sign modules with SHA-256"
2053 select CRYPTO_SHA256
2055 config MODULE_SIG_SHA384
2056 bool "Sign modules with SHA-384"
2057 select CRYPTO_SHA512
2059 config MODULE_SIG_SHA512
2060 bool "Sign modules with SHA-512"
2061 select CRYPTO_SHA512
2065 config MODULE_SIG_HASH
2067 depends on MODULE_SIG
2068 default "sha1" if MODULE_SIG_SHA1
2069 default "sha224" if MODULE_SIG_SHA224
2070 default "sha256" if MODULE_SIG_SHA256
2071 default "sha384" if MODULE_SIG_SHA384
2072 default "sha512" if MODULE_SIG_SHA512
2074 config MODULE_COMPRESS
2075 bool "Compress modules on installation"
2079 Compresses kernel modules when 'make modules_install' is run; gzip or
2080 xz depending on "Compression algorithm" below.
2082 module-init-tools MAY support gzip, and kmod MAY support gzip and xz.
2084 Out-of-tree kernel modules installed using Kbuild will also be
2085 compressed upon installation.
2087 Note: for modules inside an initrd or initramfs, it's more efficient
2088 to compress the whole initrd or initramfs instead.
2090 Note: This is fully compatible with signed modules.
2095 prompt "Compression algorithm"
2096 depends on MODULE_COMPRESS
2097 default MODULE_COMPRESS_GZIP
2099 This determines which sort of compression will be used during
2100 'make modules_install'.
2102 GZIP (default) and XZ are supported.
2104 config MODULE_COMPRESS_GZIP
2107 config MODULE_COMPRESS_XZ
2112 config TRIM_UNUSED_KSYMS
2113 bool "Trim unused exported kernel symbols"
2114 depends on MODULES && !UNUSED_SYMBOLS
2116 The kernel and some modules make many symbols available for
2117 other modules to use via EXPORT_SYMBOL() and variants. Depending
2118 on the set of modules being selected in your kernel configuration,
2119 many of those exported symbols might never be used.
2121 This option allows for unused exported symbols to be dropped from
2122 the build. In turn, this provides the compiler more opportunities
2123 (especially when using LTO) for optimizing the code and reducing
2124 binary size. This might have some security advantages as well.
2126 If unsure, or if you need to build out-of-tree modules, say N.
2130 config MODULES_TREE_LOOKUP
2132 depends on PERF_EVENTS || TRACING
2134 config INIT_ALL_POSSIBLE
2137 Back when each arch used to define their own cpu_online_mask and
2138 cpu_possible_mask, some of them chose to initialize cpu_possible_mask
2139 with all 1s, and others with all 0s. When they were centralised,
2140 it was better to provide this option than to break all the archs
2141 and have several arch maintainers pursuing me down dark alleys.
2143 source "block/Kconfig"
2145 config PREEMPT_NOTIFIERS
2155 Build a simple ASN.1 grammar compiler that produces a bytecode output
2156 that can be interpreted by the ASN.1 stream decoder and used to
2157 inform it as to what tags are to be expected in a stream and what
2158 functions to call on what tags.
2160 source "kernel/Kconfig.locks"