5 default "/lib/modules/$(shell,uname -r)/.config"
6 default "/etc/kernel-config"
7 default "/boot/config-$(shell,uname -r)"
9 default "arch/$(ARCH)/defconfig"
12 def_bool $(success,$(CC) --version | head -n 1 | grep -q gcc)
16 default $(shell,$(srctree)/scripts/gcc-version.sh -p $(CC) | sed 's/^0*//') if CC_IS_GCC
20 def_bool $(success,$(CC) --version | head -n 1 | grep -q clang)
24 default $(shell,$(srctree)/scripts/clang-version.sh $(CC))
26 config CC_HAS_ASM_GOTO
27 def_bool $(success,$(srctree)/scripts/gcc-goto.sh $(CC))
36 config BUILDTIME_EXTABLE_SORT
39 config THREAD_INFO_IN_TASK
42 Select this to move thread_info off the stack into task_struct. To
43 make this work, an arch will need to remove all thread_info fields
44 except flags and fix any runtime bugs.
46 One subtle change that will be needed is to use try_get_task_stack()
47 and put_task_stack() in save_thread_stack_tsk() and get_wchan().
56 depends on BROKEN || !SMP
59 config INIT_ENV_ARG_LIMIT
64 Maximum of each of the number of arguments and environment
65 variables passed to init from the kernel command line.
68 bool "Compile also drivers which will not load"
72 Some drivers can be compiled on a different platform than they are
73 intended to be run on. Despite they cannot be loaded there (or even
74 when they load they cannot be used due to missing HW support),
75 developers still, opposing to distributors, might want to build such
76 drivers to compile-test them.
78 If you are a developer and want to build everything available, say Y
79 here. If you are a user/distributor, say N here to exclude useless
80 drivers to be distributed.
83 string "Local version - append to kernel release"
85 Append an extra string to the end of your kernel version.
86 This will show up when you type uname, for example.
87 The string you set here will be appended after the contents of
88 any files with a filename matching localversion* in your
89 object and source tree, in that order. Your total string can
90 be a maximum of 64 characters.
92 config LOCALVERSION_AUTO
93 bool "Automatically append version information to the version string"
95 depends on !COMPILE_TEST
97 This will try to automatically determine if the current tree is a
98 release tree by looking for git tags that belong to the current
101 A string of the format -gxxxxxxxx will be added to the localversion
102 if a git-based tree is found. The string generated by this will be
103 appended after any matching localversion* files, and after the value
104 set in CONFIG_LOCALVERSION.
106 (The actual string used here is the first eight characters produced
107 by running the command:
109 $ git rev-parse --verify HEAD
111 which is done within the script "scripts/setlocalversion".)
114 string "Build ID Salt"
117 The build ID is used to link binaries and their debug info. Setting
118 this option will use the value in the calculation of the build id.
119 This is mostly useful for distributions which want to ensure the
120 build is unique between builds. It's safe to leave the default.
122 config HAVE_KERNEL_GZIP
125 config HAVE_KERNEL_BZIP2
128 config HAVE_KERNEL_LZMA
131 config HAVE_KERNEL_XZ
134 config HAVE_KERNEL_LZO
137 config HAVE_KERNEL_LZ4
140 config HAVE_KERNEL_UNCOMPRESSED
144 prompt "Kernel compression mode"
146 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4 || HAVE_KERNEL_UNCOMPRESSED
148 The linux kernel is a kind of self-extracting executable.
149 Several compression algorithms are available, which differ
150 in efficiency, compression and decompression speed.
151 Compression speed is only relevant when building a kernel.
152 Decompression speed is relevant at each boot.
154 If you have any problems with bzip2 or lzma compressed
155 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
156 version of this functionality (bzip2 only), for 2.4, was
157 supplied by Christian Ludwig)
159 High compression options are mostly useful for users, who
160 are low on disk space (embedded systems), but for whom ram
163 If in doubt, select 'gzip'
167 depends on HAVE_KERNEL_GZIP
169 The old and tried gzip compression. It provides a good balance
170 between compression ratio and decompression speed.
174 depends on HAVE_KERNEL_BZIP2
176 Its compression ratio and speed is intermediate.
177 Decompression speed is slowest among the choices. The kernel
178 size is about 10% smaller with bzip2, in comparison to gzip.
179 Bzip2 uses a large amount of memory. For modern kernels you
180 will need at least 8MB RAM or more for booting.
184 depends on HAVE_KERNEL_LZMA
186 This compression algorithm's ratio is best. Decompression speed
187 is between gzip and bzip2. Compression is slowest.
188 The kernel size is about 33% smaller with LZMA in comparison to gzip.
192 depends on HAVE_KERNEL_XZ
194 XZ uses the LZMA2 algorithm and instruction set specific
195 BCJ filters which can improve compression ratio of executable
196 code. The size of the kernel is about 30% smaller with XZ in
197 comparison to gzip. On architectures for which there is a BCJ
198 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ
199 will create a few percent smaller kernel than plain LZMA.
201 The speed is about the same as with LZMA: The decompression
202 speed of XZ is better than that of bzip2 but worse than gzip
203 and LZO. Compression is slow.
207 depends on HAVE_KERNEL_LZO
209 Its compression ratio is the poorest among the choices. The kernel
210 size is about 10% bigger than gzip; however its speed
211 (both compression and decompression) is the fastest.
215 depends on HAVE_KERNEL_LZ4
217 LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding.
218 A preliminary version of LZ4 de/compression tool is available at
219 <https://code.google.com/p/lz4/>.
221 Its compression ratio is worse than LZO. The size of the kernel
222 is about 8% bigger than LZO. But the decompression speed is
225 config KERNEL_UNCOMPRESSED
227 depends on HAVE_KERNEL_UNCOMPRESSED
229 Produce uncompressed kernel image. This option is usually not what
230 you want. It is useful for debugging the kernel in slow simulation
231 environments, where decompressing and moving the kernel is awfully
232 slow. This option allows early boot code to skip the decompressor
233 and jump right at uncompressed kernel image.
237 config DEFAULT_HOSTNAME
238 string "Default hostname"
241 This option determines the default system hostname before userspace
242 calls sethostname(2). The kernel traditionally uses "(none)" here,
243 but you may wish to use a different default here to make a minimal
244 system more usable with less configuration.
247 # For some reason microblaze and nios2 hard code SWAP=n. Hopefully we can
248 # add proper SWAP support to them, in which case this can be remove.
254 bool "Support for paging of anonymous memory (swap)"
255 depends on MMU && BLOCK && !ARCH_NO_SWAP
258 This option allows you to choose whether you want to have support
259 for so called swap devices or swap files in your kernel that are
260 used to provide more virtual memory than the actual RAM present
261 in your computer. If unsure say Y.
266 Inter Process Communication is a suite of library functions and
267 system calls which let processes (running programs) synchronize and
268 exchange information. It is generally considered to be a good thing,
269 and some programs won't run unless you say Y here. In particular, if
270 you want to run the DOS emulator dosemu under Linux (read the
271 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
272 you'll need to say Y here.
274 You can find documentation about IPC with "info ipc" and also in
275 section 6.4 of the Linux Programmer's Guide, available from
276 <http://www.tldp.org/guides.html>.
278 config SYSVIPC_SYSCTL
285 bool "POSIX Message Queues"
288 POSIX variant of message queues is a part of IPC. In POSIX message
289 queues every message has a priority which decides about succession
290 of receiving it by a process. If you want to compile and run
291 programs written e.g. for Solaris with use of its POSIX message
292 queues (functions mq_*) say Y here.
294 POSIX message queues are visible as a filesystem called 'mqueue'
295 and can be mounted somewhere if you want to do filesystem
296 operations on message queues.
300 config POSIX_MQUEUE_SYSCTL
302 depends on POSIX_MQUEUE
306 config CROSS_MEMORY_ATTACH
307 bool "Enable process_vm_readv/writev syscalls"
311 Enabling this option adds the system calls process_vm_readv and
312 process_vm_writev which allow a process with the correct privileges
313 to directly read from or write to another process' address space.
314 See the man page for more details.
317 bool "uselib syscall"
318 def_bool ALPHA || M68K || SPARC || X86_32 || IA32_EMULATION
320 This option enables the uselib syscall, a system call used in the
321 dynamic linker from libc5 and earlier. glibc does not use this
322 system call. If you intend to run programs built on libc5 or
323 earlier, you may need to enable this syscall. Current systems
324 running glibc can safely disable this.
327 bool "Auditing support"
330 Enable auditing infrastructure that can be used with another
331 kernel subsystem, such as SELinux (which requires this for
332 logging of avc messages output). System call auditing is included
333 on architectures which support it.
335 config HAVE_ARCH_AUDITSYSCALL
340 depends on AUDIT && HAVE_ARCH_AUDITSYSCALL
344 depends on AUDITSYSCALL
349 depends on AUDITSYSCALL
352 source "kernel/irq/Kconfig"
353 source "kernel/time/Kconfig"
354 source "kernel/Kconfig.preempt"
356 menu "CPU/Task time and stats accounting"
358 config VIRT_CPU_ACCOUNTING
362 prompt "Cputime accounting"
363 default TICK_CPU_ACCOUNTING if !PPC64
364 default VIRT_CPU_ACCOUNTING_NATIVE if PPC64
366 # Kind of a stub config for the pure tick based cputime accounting
367 config TICK_CPU_ACCOUNTING
368 bool "Simple tick based cputime accounting"
369 depends on !S390 && !NO_HZ_FULL
371 This is the basic tick based cputime accounting that maintains
372 statistics about user, system and idle time spent on per jiffies
377 config VIRT_CPU_ACCOUNTING_NATIVE
378 bool "Deterministic task and CPU time accounting"
379 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL
380 select VIRT_CPU_ACCOUNTING
382 Select this option to enable more accurate task and CPU time
383 accounting. This is done by reading a CPU counter on each
384 kernel entry and exit and on transitions within the kernel
385 between system, softirq and hardirq state, so there is a
386 small performance impact. In the case of s390 or IBM POWER > 5,
387 this also enables accounting of stolen time on logically-partitioned
390 config VIRT_CPU_ACCOUNTING_GEN
391 bool "Full dynticks CPU time accounting"
392 depends on HAVE_CONTEXT_TRACKING
393 depends on HAVE_VIRT_CPU_ACCOUNTING_GEN
394 select VIRT_CPU_ACCOUNTING
395 select CONTEXT_TRACKING
397 Select this option to enable task and CPU time accounting on full
398 dynticks systems. This accounting is implemented by watching every
399 kernel-user boundaries using the context tracking subsystem.
400 The accounting is thus performed at the expense of some significant
403 For now this is only useful if you are working on the full
404 dynticks subsystem development.
410 config IRQ_TIME_ACCOUNTING
411 bool "Fine granularity task level IRQ time accounting"
412 depends on HAVE_IRQ_TIME_ACCOUNTING && !VIRT_CPU_ACCOUNTING_NATIVE
414 Select this option to enable fine granularity task irq time
415 accounting. This is done by reading a timestamp on each
416 transitions between softirq and hardirq state, so there can be a
417 small performance impact.
419 If in doubt, say N here.
421 config HAVE_SCHED_AVG_IRQ
423 depends on IRQ_TIME_ACCOUNTING || PARAVIRT_TIME_ACCOUNTING
426 config BSD_PROCESS_ACCT
427 bool "BSD Process Accounting"
430 If you say Y here, a user level program will be able to instruct the
431 kernel (via a special system call) to write process accounting
432 information to a file: whenever a process exits, information about
433 that process will be appended to the file by the kernel. The
434 information includes things such as creation time, owning user,
435 command name, memory usage, controlling terminal etc. (the complete
436 list is in the struct acct in <file:include/linux/acct.h>). It is
437 up to the user level program to do useful things with this
438 information. This is generally a good idea, so say Y.
440 config BSD_PROCESS_ACCT_V3
441 bool "BSD Process Accounting version 3 file format"
442 depends on BSD_PROCESS_ACCT
445 If you say Y here, the process accounting information is written
446 in a new file format that also logs the process IDs of each
447 process and its parent. Note that this file format is incompatible
448 with previous v0/v1/v2 file formats, so you will need updated tools
449 for processing it. A preliminary version of these tools is available
450 at <http://www.gnu.org/software/acct/>.
453 bool "Export task/process statistics through netlink"
458 Export selected statistics for tasks/processes through the
459 generic netlink interface. Unlike BSD process accounting, the
460 statistics are available during the lifetime of tasks/processes as
461 responses to commands. Like BSD accounting, they are sent to user
466 config TASK_DELAY_ACCT
467 bool "Enable per-task delay accounting"
471 Collect information on time spent by a task waiting for system
472 resources like cpu, synchronous block I/O completion and swapping
473 in pages. Such statistics can help in setting a task's priorities
474 relative to other tasks for cpu, io, rss limits etc.
479 bool "Enable extended accounting over taskstats"
482 Collect extended task accounting data and send the data
483 to userland for processing over the taskstats interface.
487 config TASK_IO_ACCOUNTING
488 bool "Enable per-task storage I/O accounting"
489 depends on TASK_XACCT
491 Collect information on the number of bytes of storage I/O which this
496 endmenu # "CPU/Task time and stats accounting"
500 depends on SMP || COMPILE_TEST
503 Make sure that CPUs running critical tasks are not disturbed by
504 any source of "noise" such as unbound workqueues, timers, kthreads...
505 Unbound jobs get offloaded to housekeeping CPUs. This is driven by
506 the "isolcpus=" boot parameter.
510 source "kernel/rcu/Kconfig"
517 tristate "Kernel .config support"
520 This option enables the complete Linux kernel ".config" file
521 contents to be saved in the kernel. It provides documentation
522 of which kernel options are used in a running kernel or in an
523 on-disk kernel. This information can be extracted from the kernel
524 image file with the script scripts/extract-ikconfig and used as
525 input to rebuild the current kernel or to build another kernel.
526 It can also be extracted from a running kernel by reading
527 /proc/config.gz if enabled (below).
530 bool "Enable access to .config through /proc/config.gz"
531 depends on IKCONFIG && PROC_FS
533 This option enables access to the kernel configuration file
534 through /proc/config.gz.
537 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
542 Select the minimal kernel log buffer size as a power of 2.
543 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config
544 parameter, see below. Any higher size also might be forced
545 by "log_buf_len" boot parameter.
555 config LOG_CPU_MAX_BUF_SHIFT
556 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)"
559 default 12 if !BASE_SMALL
560 default 0 if BASE_SMALL
563 This option allows to increase the default ring buffer size
564 according to the number of CPUs. The value defines the contribution
565 of each CPU as a power of 2. The used space is typically only few
566 lines however it might be much more when problems are reported,
569 The increased size means that a new buffer has to be allocated and
570 the original static one is unused. It makes sense only on systems
571 with more CPUs. Therefore this value is used only when the sum of
572 contributions is greater than the half of the default kernel ring
573 buffer as defined by LOG_BUF_SHIFT. The default values are set
574 so that more than 64 CPUs are needed to trigger the allocation.
576 Also this option is ignored when "log_buf_len" kernel parameter is
577 used as it forces an exact (power of two) size of the ring buffer.
579 The number of possible CPUs is used for this computation ignoring
580 hotplugging making the computation optimal for the worst case
581 scenario while allowing a simple algorithm to be used from bootup.
583 Examples shift values and their meaning:
584 17 => 128 KB for each CPU
585 16 => 64 KB for each CPU
586 15 => 32 KB for each CPU
587 14 => 16 KB for each CPU
588 13 => 8 KB for each CPU
589 12 => 4 KB for each CPU
591 config PRINTK_SAFE_LOG_BUF_SHIFT
592 int "Temporary per-CPU printk log buffer size (12 => 4KB, 13 => 8KB)"
597 Select the size of an alternate printk per-CPU buffer where messages
598 printed from usafe contexts are temporary stored. One example would
599 be NMI messages, another one - printk recursion. The messages are
600 copied to the main log buffer in a safe context to avoid a deadlock.
601 The value defines the size as a power of 2.
603 Those messages are rare and limited. The largest one is when
604 a backtrace is printed. It usually fits into 4KB. Select
605 8KB if you want to be on the safe side.
608 17 => 128 KB for each CPU
609 16 => 64 KB for each CPU
610 15 => 32 KB for each CPU
611 14 => 16 KB for each CPU
612 13 => 8 KB for each CPU
613 12 => 4 KB for each CPU
616 # Architectures with an unreliable sched_clock() should select this:
618 config HAVE_UNSTABLE_SCHED_CLOCK
621 config GENERIC_SCHED_CLOCK
625 # For architectures that want to enable the support for NUMA-affine scheduler
628 config ARCH_SUPPORTS_NUMA_BALANCING
632 # For architectures that prefer to flush all TLBs after a number of pages
633 # are unmapped instead of sending one IPI per page to flush. The architecture
634 # must provide guarantees on what happens if a clean TLB cache entry is
635 # written after the unmap. Details are in mm/rmap.c near the check for
636 # should_defer_flush. The architecture should also consider if the full flush
637 # and the refill costs are offset by the savings of sending fewer IPIs.
638 config ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
642 # For architectures that know their GCC __int128 support is sound
644 config ARCH_SUPPORTS_INT128
647 # For architectures that (ab)use NUMA to represent different memory regions
648 # all cpu-local but of different latencies, such as SuperH.
650 config ARCH_WANT_NUMA_VARIABLE_LOCALITY
653 config NUMA_BALANCING
654 bool "Memory placement aware NUMA scheduler"
655 depends on ARCH_SUPPORTS_NUMA_BALANCING
656 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
657 depends on SMP && NUMA && MIGRATION
659 This option adds support for automatic NUMA aware memory/task placement.
660 The mechanism is quite primitive and is based on migrating memory when
661 it has references to the node the task is running on.
663 This system will be inactive on UMA systems.
665 config NUMA_BALANCING_DEFAULT_ENABLED
666 bool "Automatically enable NUMA aware memory/task placement"
668 depends on NUMA_BALANCING
670 If set, automatic NUMA balancing will be enabled if running on a NUMA
674 bool "Control Group support"
677 This option adds support for grouping sets of processes together, for
678 use with process control subsystems such as Cpusets, CFS, memory
679 controls or device isolation.
681 - Documentation/scheduler/sched-design-CFS.txt (CFS)
682 - Documentation/cgroup-v1/ (features for grouping, isolation
683 and resource control)
693 bool "Memory controller"
697 Provides control over the memory footprint of tasks in a cgroup.
700 bool "Swap controller"
701 depends on MEMCG && SWAP
703 Provides control over the swap space consumed by tasks in a cgroup.
705 config MEMCG_SWAP_ENABLED
706 bool "Swap controller enabled by default"
707 depends on MEMCG_SWAP
710 Memory Resource Controller Swap Extension comes with its price in
711 a bigger memory consumption. General purpose distribution kernels
712 which want to enable the feature but keep it disabled by default
713 and let the user enable it by swapaccount=1 boot command line
714 parameter should have this option unselected.
715 For those who want to have the feature enabled by default should
716 select this option (if, for some reason, they need to disable it
717 then swapaccount=0 does the trick).
721 depends on MEMCG && !SLOB
729 Generic block IO controller cgroup interface. This is the common
730 cgroup interface which should be used by various IO controlling
733 Currently, CFQ IO scheduler uses it to recognize task groups and
734 control disk bandwidth allocation (proportional time slice allocation)
735 to such task groups. It is also used by bio throttling logic in
736 block layer to implement upper limit in IO rates on a device.
738 This option only enables generic Block IO controller infrastructure.
739 One needs to also enable actual IO controlling logic/policy. For
740 enabling proportional weight division of disk bandwidth in CFQ, set
741 CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set
742 CONFIG_BLK_DEV_THROTTLING=y.
744 See Documentation/cgroup-v1/blkio-controller.txt for more information.
746 config DEBUG_BLK_CGROUP
747 bool "IO controller debugging"
748 depends on BLK_CGROUP
751 Enable some debugging help. Currently it exports additional stat
752 files in a cgroup which can be useful for debugging.
754 config CGROUP_WRITEBACK
756 depends on MEMCG && BLK_CGROUP
759 menuconfig CGROUP_SCHED
760 bool "CPU controller"
763 This feature lets CPU scheduler recognize task groups and control CPU
764 bandwidth allocation to such task groups. It uses cgroups to group
768 config FAIR_GROUP_SCHED
769 bool "Group scheduling for SCHED_OTHER"
770 depends on CGROUP_SCHED
774 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED"
775 depends on FAIR_GROUP_SCHED
778 This option allows users to define CPU bandwidth rates (limits) for
779 tasks running within the fair group scheduler. Groups with no limit
780 set are considered to be unconstrained and will run with no
782 See Documentation/scheduler/sched-bwc.txt for more information.
784 config RT_GROUP_SCHED
785 bool "Group scheduling for SCHED_RR/FIFO"
786 depends on CGROUP_SCHED
789 This feature lets you explicitly allocate real CPU bandwidth
790 to task groups. If enabled, it will also make it impossible to
791 schedule realtime tasks for non-root users until you allocate
792 realtime bandwidth for them.
793 See Documentation/scheduler/sched-rt-group.txt for more information.
798 bool "PIDs controller"
800 Provides enforcement of process number limits in the scope of a
801 cgroup. Any attempt to fork more processes than is allowed in the
802 cgroup will fail. PIDs are fundamentally a global resource because it
803 is fairly trivial to reach PID exhaustion before you reach even a
804 conservative kmemcg limit. As a result, it is possible to grind a
805 system to halt without being limited by other cgroup policies. The
806 PIDs controller is designed to stop this from happening.
808 It should be noted that organisational operations (such as attaching
809 to a cgroup hierarchy will *not* be blocked by the PIDs controller),
810 since the PIDs limit only affects a process's ability to fork, not to
814 bool "RDMA controller"
816 Provides enforcement of RDMA resources defined by IB stack.
817 It is fairly easy for consumers to exhaust RDMA resources, which
818 can result into resource unavailability to other consumers.
819 RDMA controller is designed to stop this from happening.
820 Attaching processes with active RDMA resources to the cgroup
821 hierarchy is allowed even if can cross the hierarchy's limit.
823 config CGROUP_FREEZER
824 bool "Freezer controller"
826 Provides a way to freeze and unfreeze all tasks in a
829 This option affects the ORIGINAL cgroup interface. The cgroup2 memory
830 controller includes important in-kernel memory consumers per default.
832 If you're using cgroup2, say N.
834 config CGROUP_HUGETLB
835 bool "HugeTLB controller"
836 depends on HUGETLB_PAGE
840 Provides a cgroup controller for HugeTLB pages.
841 When you enable this, you can put a per cgroup limit on HugeTLB usage.
842 The limit is enforced during page fault. Since HugeTLB doesn't
843 support page reclaim, enforcing the limit at page fault time implies
844 that, the application will get SIGBUS signal if it tries to access
845 HugeTLB pages beyond its limit. This requires the application to know
846 beforehand how much HugeTLB pages it would require for its use. The
847 control group is tracked in the third page lru pointer. This means
848 that we cannot use the controller with huge page less than 3 pages.
851 bool "Cpuset controller"
854 This option will let you create and manage CPUSETs which
855 allow dynamically partitioning a system into sets of CPUs and
856 Memory Nodes and assigning tasks to run only within those sets.
857 This is primarily useful on large SMP or NUMA systems.
861 config PROC_PID_CPUSET
862 bool "Include legacy /proc/<pid>/cpuset file"
867 bool "Device controller"
869 Provides a cgroup controller implementing whitelists for
870 devices which a process in the cgroup can mknod or open.
872 config CGROUP_CPUACCT
873 bool "Simple CPU accounting controller"
875 Provides a simple controller for monitoring the
876 total CPU consumed by the tasks in a cgroup.
879 bool "Perf controller"
880 depends on PERF_EVENTS
882 This option extends the perf per-cpu mode to restrict monitoring
883 to threads which belong to the cgroup specified and run on the
889 bool "Support for eBPF programs attached to cgroups"
890 depends on BPF_SYSCALL
891 select SOCK_CGROUP_DATA
893 Allow attaching eBPF programs to a cgroup using the bpf(2)
894 syscall command BPF_PROG_ATTACH.
896 In which context these programs are accessed depends on the type
897 of attachment. For instance, programs that are attached using
898 BPF_CGROUP_INET_INGRESS will be executed on the ingress path of
902 bool "Debug controller"
904 depends on DEBUG_KERNEL
906 This option enables a simple controller that exports
907 debugging information about the cgroups framework. This
908 controller is for control cgroup debugging only. Its
909 interfaces are not stable.
913 config SOCK_CGROUP_DATA
919 menuconfig NAMESPACES
920 bool "Namespaces support" if EXPERT
924 Provides the way to make tasks work with different objects using
925 the same id. For example same IPC id may refer to different objects
926 or same user id or pid may refer to different tasks when used in
927 different namespaces.
935 In this namespace tasks see different info provided with the
940 depends on (SYSVIPC || POSIX_MQUEUE)
943 In this namespace tasks work with IPC ids which correspond to
944 different IPC objects in different namespaces.
947 bool "User namespace"
950 This allows containers, i.e. vservers, to use user namespaces
951 to provide different user info for different servers.
953 When user namespaces are enabled in the kernel it is
954 recommended that the MEMCG option also be enabled and that
955 user-space use the memory control groups to limit the amount
956 of memory a memory unprivileged users can use.
961 bool "PID Namespaces"
964 Support process id namespaces. This allows having multiple
965 processes with the same pid as long as they are in different
966 pid namespaces. This is a building block of containers.
969 bool "Network namespace"
973 Allow user space to create what appear to be multiple instances
974 of the network stack.
978 config CHECKPOINT_RESTORE
979 bool "Checkpoint/restore support"
983 Enables additional kernel features in a sake of checkpoint/restore.
984 In particular it adds auxiliary prctl codes to setup process text,
985 data and heap segment sizes, and a few additional /proc filesystem
988 If unsure, say N here.
990 config SCHED_AUTOGROUP
991 bool "Automatic process group scheduling"
994 select FAIR_GROUP_SCHED
996 This option optimizes the scheduler for common desktop workloads by
997 automatically creating and populating task groups. This separation
998 of workloads isolates aggressive CPU burners (like build jobs) from
999 desktop applications. Task group autogeneration is currently based
1002 config SYSFS_DEPRECATED
1003 bool "Enable deprecated sysfs features to support old userspace tools"
1007 This option adds code that switches the layout of the "block" class
1008 devices, to not show up in /sys/class/block/, but only in
1011 This switch is only active when the sysfs.deprecated=1 boot option is
1012 passed or the SYSFS_DEPRECATED_V2 option is set.
1014 This option allows new kernels to run on old distributions and tools,
1015 which might get confused by /sys/class/block/. Since 2007/2008 all
1016 major distributions and tools handle this just fine.
1018 Recent distributions and userspace tools after 2009/2010 depend on
1019 the existence of /sys/class/block/, and will not work with this
1022 Only if you are using a new kernel on an old distribution, you might
1025 config SYSFS_DEPRECATED_V2
1026 bool "Enable deprecated sysfs features by default"
1029 depends on SYSFS_DEPRECATED
1031 Enable deprecated sysfs by default.
1033 See the CONFIG_SYSFS_DEPRECATED option for more details about this
1036 Only if you are using a new kernel on an old distribution, you might
1037 need to say Y here. Even then, odds are you would not need it
1038 enabled, you can always pass the boot option if absolutely necessary.
1041 bool "Kernel->user space relay support (formerly relayfs)"
1044 This option enables support for relay interface support in
1045 certain file systems (such as debugfs).
1046 It is designed to provide an efficient mechanism for tools and
1047 facilities to relay large amounts of data from kernel space to
1052 config BLK_DEV_INITRD
1053 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
1055 The initial RAM filesystem is a ramfs which is loaded by the
1056 boot loader (loadlin or lilo) and that is mounted as root
1057 before the normal boot procedure. It is typically used to
1058 load modules needed to mount the "real" root file system,
1059 etc. See <file:Documentation/admin-guide/initrd.rst> for details.
1061 If RAM disk support (BLK_DEV_RAM) is also included, this
1062 also enables initial RAM disk (initrd) support and adds
1063 15 Kbytes (more on some other architectures) to the kernel size.
1069 source "usr/Kconfig"
1074 prompt "Compiler optimization level"
1075 default CC_OPTIMIZE_FOR_PERFORMANCE
1077 config CC_OPTIMIZE_FOR_PERFORMANCE
1078 bool "Optimize for performance"
1080 This is the default optimization level for the kernel, building
1081 with the "-O2" compiler flag for best performance and most
1082 helpful compile-time warnings.
1084 config CC_OPTIMIZE_FOR_SIZE
1085 bool "Optimize for size"
1087 Enabling this option will pass "-Os" instead of "-O2" to
1088 your compiler resulting in a smaller kernel.
1094 config HAVE_LD_DEAD_CODE_DATA_ELIMINATION
1097 This requires that the arch annotates or otherwise protects
1098 its external entry points from being discarded. Linker scripts
1099 must also merge .text.*, .data.*, and .bss.* correctly into
1100 output sections. Care must be taken not to pull in unrelated
1101 sections (e.g., '.text.init'). Typically '.' in section names
1102 is used to distinguish them from label names / C identifiers.
1104 config LD_DEAD_CODE_DATA_ELIMINATION
1105 bool "Dead code and data elimination (EXPERIMENTAL)"
1106 depends on HAVE_LD_DEAD_CODE_DATA_ELIMINATION
1108 depends on !(FUNCTION_TRACER && CC_IS_GCC && GCC_VERSION < 40800)
1109 depends on $(cc-option,-ffunction-sections -fdata-sections)
1110 depends on $(ld-option,--gc-sections)
1112 Enable this if you want to do dead code and data elimination with
1113 the linker by compiling with -ffunction-sections -fdata-sections,
1114 and linking with --gc-sections.
1116 This can reduce on disk and in-memory size of the kernel
1117 code and static data, particularly for small configs and
1118 on small systems. This has the possibility of introducing
1119 silently broken kernel if the required annotations are not
1120 present. This option is not well tested yet, so use at your
1132 config SYSCTL_EXCEPTION_TRACE
1135 Enable support for /proc/sys/debug/exception-trace.
1137 config SYSCTL_ARCH_UNALIGN_NO_WARN
1140 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap
1141 Allows arch to define/use @no_unaligned_warning to possibly warn
1142 about unaligned access emulation going on under the hood.
1144 config SYSCTL_ARCH_UNALIGN_ALLOW
1147 Enable support for /proc/sys/kernel/unaligned-trap
1148 Allows arches to define/use @unaligned_enabled to runtime toggle
1149 the unaligned access emulation.
1150 see arch/parisc/kernel/unaligned.c for reference
1152 config HAVE_PCSPKR_PLATFORM
1155 # interpreter that classic socket filters depend on
1160 bool "Configure standard kernel features (expert users)"
1161 # Unhide debug options, to make the on-by-default options visible
1164 This option allows certain base kernel options and settings
1165 to be disabled or tweaked. This is for specialized
1166 environments which can tolerate a "non-standard" kernel.
1167 Only use this if you really know what you are doing.
1170 bool "Enable 16-bit UID system calls" if EXPERT
1171 depends on HAVE_UID16 && MULTIUSER
1174 This enables the legacy 16-bit UID syscall wrappers.
1177 bool "Multiple users, groups and capabilities support" if EXPERT
1180 This option enables support for non-root users, groups and
1183 If you say N here, all processes will run with UID 0, GID 0, and all
1184 possible capabilities. Saying N here also compiles out support for
1185 system calls related to UIDs, GIDs, and capabilities, such as setuid,
1188 If unsure, say Y here.
1190 config SGETMASK_SYSCALL
1191 bool "sgetmask/ssetmask syscalls support" if EXPERT
1192 def_bool PARISC || M68K || PPC || MIPS || X86 || SPARC || MICROBLAZE || SUPERH
1194 sys_sgetmask and sys_ssetmask are obsolete system calls
1195 no longer supported in libc but still enabled by default in some
1198 If unsure, leave the default option here.
1200 config SYSFS_SYSCALL
1201 bool "Sysfs syscall support" if EXPERT
1204 sys_sysfs is an obsolete system call no longer supported in libc.
1205 Note that disabling this option is more secure but might break
1206 compatibility with some systems.
1208 If unsure say Y here.
1210 config SYSCTL_SYSCALL
1211 bool "Sysctl syscall support" if EXPERT
1212 depends on PROC_SYSCTL
1216 sys_sysctl uses binary paths that have been found challenging
1217 to properly maintain and use. The interface in /proc/sys
1218 using paths with ascii names is now the primary path to this
1221 Almost nothing using the binary sysctl interface so if you are
1222 trying to save some space it is probably safe to disable this,
1223 making your kernel marginally smaller.
1225 If unsure say N here.
1228 bool "open by fhandle syscalls" if EXPERT
1232 If you say Y here, a user level program will be able to map
1233 file names to handle and then later use the handle for
1234 different file system operations. This is useful in implementing
1235 userspace file servers, which now track files using handles instead
1236 of names. The handle would remain the same even if file names
1237 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2)
1241 bool "Posix Clocks & timers" if EXPERT
1244 This includes native support for POSIX timers to the kernel.
1245 Some embedded systems have no use for them and therefore they
1246 can be configured out to reduce the size of the kernel image.
1248 When this option is disabled, the following syscalls won't be
1249 available: timer_create, timer_gettime: timer_getoverrun,
1250 timer_settime, timer_delete, clock_adjtime, getitimer,
1251 setitimer, alarm. Furthermore, the clock_settime, clock_gettime,
1252 clock_getres and clock_nanosleep syscalls will be limited to
1253 CLOCK_REALTIME, CLOCK_MONOTONIC and CLOCK_BOOTTIME only.
1259 bool "Enable support for printk" if EXPERT
1262 This option enables normal printk support. Removing it
1263 eliminates most of the message strings from the kernel image
1264 and makes the kernel more or less silent. As this makes it
1265 very difficult to diagnose system problems, saying N here is
1266 strongly discouraged.
1274 bool "BUG() support" if EXPERT
1277 Disabling this option eliminates support for BUG and WARN, reducing
1278 the size of your kernel image and potentially quietly ignoring
1279 numerous fatal conditions. You should only consider disabling this
1280 option for embedded systems with no facilities for reporting errors.
1286 bool "Enable ELF core dumps" if EXPERT
1288 Enable support for generating core dumps. Disabling saves about 4k.
1291 config PCSPKR_PLATFORM
1292 bool "Enable PC-Speaker support" if EXPERT
1293 depends on HAVE_PCSPKR_PLATFORM
1297 This option allows to disable the internal PC-Speaker
1298 support, saving some memory.
1302 bool "Enable full-sized data structures for core" if EXPERT
1304 Disabling this option reduces the size of miscellaneous core
1305 kernel data structures. This saves memory on small machines,
1306 but may reduce performance.
1309 bool "Enable futex support" if EXPERT
1313 Disabling this option will cause the kernel to be built without
1314 support for "fast userspace mutexes". The resulting kernel may not
1315 run glibc-based applications correctly.
1319 depends on FUTEX && RT_MUTEXES
1322 config HAVE_FUTEX_CMPXCHG
1326 Architectures should select this if futex_atomic_cmpxchg_inatomic()
1327 is implemented and always working. This removes a couple of runtime
1331 bool "Enable eventpoll support" if EXPERT
1335 Disabling this option will cause the kernel to be built without
1336 support for epoll family of system calls.
1339 bool "Enable signalfd() system call" if EXPERT
1343 Enable the signalfd() system call that allows to receive signals
1344 on a file descriptor.
1349 bool "Enable timerfd() system call" if EXPERT
1353 Enable the timerfd() system call that allows to receive timer
1354 events on a file descriptor.
1359 bool "Enable eventfd() system call" if EXPERT
1363 Enable the eventfd() system call that allows to receive both
1364 kernel notification (ie. KAIO) or userspace notifications.
1369 bool "Use full shmem filesystem" if EXPERT
1373 The shmem is an internal filesystem used to manage shared memory.
1374 It is backed by swap and manages resource limits. It is also exported
1375 to userspace as tmpfs if TMPFS is enabled. Disabling this
1376 option replaces shmem and tmpfs with the much simpler ramfs code,
1377 which may be appropriate on small systems without swap.
1380 bool "Enable AIO support" if EXPERT
1383 This option enables POSIX asynchronous I/O which may by used
1384 by some high performance threaded applications. Disabling
1385 this option saves about 7k.
1387 config ADVISE_SYSCALLS
1388 bool "Enable madvise/fadvise syscalls" if EXPERT
1391 This option enables the madvise and fadvise syscalls, used by
1392 applications to advise the kernel about their future memory or file
1393 usage, improving performance. If building an embedded system where no
1394 applications use these syscalls, you can disable this option to save
1398 bool "Enable membarrier() system call" if EXPERT
1401 Enable the membarrier() system call that allows issuing memory
1402 barriers across all running threads, which can be used to distribute
1403 the cost of user-space memory barriers asymmetrically by transforming
1404 pairs of memory barriers into pairs consisting of membarrier() and a
1410 bool "Load all symbols for debugging/ksymoops" if EXPERT
1413 Say Y here to let the kernel print out symbolic crash information and
1414 symbolic stack backtraces. This increases the size of the kernel
1415 somewhat, as all symbols have to be loaded into the kernel image.
1418 bool "Include all symbols in kallsyms"
1419 depends on DEBUG_KERNEL && KALLSYMS
1421 Normally kallsyms only contains the symbols of functions for nicer
1422 OOPS messages and backtraces (i.e., symbols from the text and inittext
1423 sections). This is sufficient for most cases. And only in very rare
1424 cases (e.g., when a debugger is used) all symbols are required (e.g.,
1425 names of variables from the data sections, etc).
1427 This option makes sure that all symbols are loaded into the kernel
1428 image (i.e., symbols from all sections) in cost of increased kernel
1429 size (depending on the kernel configuration, it may be 300KiB or
1430 something like this).
1432 Say N unless you really need all symbols.
1434 config KALLSYMS_ABSOLUTE_PERCPU
1437 default X86_64 && SMP
1439 config KALLSYMS_BASE_RELATIVE
1444 Instead of emitting them as absolute values in the native word size,
1445 emit the symbol references in the kallsyms table as 32-bit entries,
1446 each containing a relative value in the range [base, base + U32_MAX]
1447 or, when KALLSYMS_ABSOLUTE_PERCPU is in effect, each containing either
1448 an absolute value in the range [0, S32_MAX] or a relative value in the
1449 range [base, base + S32_MAX], where base is the lowest relative symbol
1450 address encountered in the image.
1452 On 64-bit builds, this reduces the size of the address table by 50%,
1453 but more importantly, it results in entries whose values are build
1454 time constants, and no relocation pass is required at runtime to fix
1455 up the entries based on the runtime load address of the kernel.
1457 # end of the "standard kernel features (expert users)" menu
1459 # syscall, maps, verifier
1461 bool "Enable bpf() system call"
1467 Enable the bpf() system call that allows to manipulate eBPF
1468 programs and maps via file descriptors.
1470 config BPF_JIT_ALWAYS_ON
1471 bool "Permanently enable BPF JIT and remove BPF interpreter"
1472 depends on BPF_SYSCALL && HAVE_EBPF_JIT && BPF_JIT
1474 Enables BPF JIT and removes BPF interpreter to avoid
1475 speculative execution of BPF instructions by the interpreter
1478 bool "Enable userfaultfd() system call"
1482 Enable the userfaultfd() system call that allows to intercept and
1483 handle page faults in userland.
1485 config ARCH_HAS_MEMBARRIER_CALLBACKS
1488 config ARCH_HAS_MEMBARRIER_SYNC_CORE
1492 bool "Enable rseq() system call" if EXPERT
1494 depends on HAVE_RSEQ
1497 Enable the restartable sequences system call. It provides a
1498 user-space cache for the current CPU number value, which
1499 speeds up getting the current CPU number from user-space,
1500 as well as an ABI to speed up user-space operations on
1507 bool "Enabled debugging of rseq() system call" if EXPERT
1508 depends on RSEQ && DEBUG_KERNEL
1510 Enable extra debugging checks for the rseq system call.
1515 bool "Embedded system"
1516 option allnoconfig_y
1519 This option should be enabled if compiling the kernel for
1520 an embedded system so certain expert options are available
1523 config HAVE_PERF_EVENTS
1526 See tools/perf/design.txt for details.
1528 config PERF_USE_VMALLOC
1531 See tools/perf/design.txt for details
1534 bool "PC/104 support" if EXPERT
1536 Expose PC/104 form factor device drivers and options available for
1537 selection and configuration. Enable this option if your target
1538 machine has a PC/104 bus.
1540 menu "Kernel Performance Events And Counters"
1543 bool "Kernel performance events and counters"
1544 default y if PROFILING
1545 depends on HAVE_PERF_EVENTS
1550 Enable kernel support for various performance events provided
1551 by software and hardware.
1553 Software events are supported either built-in or via the
1554 use of generic tracepoints.
1556 Most modern CPUs support performance events via performance
1557 counter registers. These registers count the number of certain
1558 types of hw events: such as instructions executed, cachemisses
1559 suffered, or branches mis-predicted - without slowing down the
1560 kernel or applications. These registers can also trigger interrupts
1561 when a threshold number of events have passed - and can thus be
1562 used to profile the code that runs on that CPU.
1564 The Linux Performance Event subsystem provides an abstraction of
1565 these software and hardware event capabilities, available via a
1566 system call and used by the "perf" utility in tools/perf/. It
1567 provides per task and per CPU counters, and it provides event
1568 capabilities on top of those.
1572 config DEBUG_PERF_USE_VMALLOC
1574 bool "Debug: use vmalloc to back perf mmap() buffers"
1575 depends on PERF_EVENTS && DEBUG_KERNEL && !PPC
1576 select PERF_USE_VMALLOC
1578 Use vmalloc memory to back perf mmap() buffers.
1580 Mostly useful for debugging the vmalloc code on platforms
1581 that don't require it.
1587 config VM_EVENT_COUNTERS
1589 bool "Enable VM event counters for /proc/vmstat" if EXPERT
1591 VM event counters are needed for event counts to be shown.
1592 This option allows the disabling of the VM event counters
1593 on EXPERT systems. /proc/vmstat will only show page counts
1594 if VM event counters are disabled.
1598 bool "Enable SLUB debugging support" if EXPERT
1599 depends on SLUB && SYSFS
1601 SLUB has extensive debug support features. Disabling these can
1602 result in significant savings in code size. This also disables
1603 SLUB sysfs support. /sys/slab will not exist and there will be
1604 no support for cache validation etc.
1606 config SLUB_MEMCG_SYSFS_ON
1608 bool "Enable memcg SLUB sysfs support by default" if EXPERT
1609 depends on SLUB && SYSFS && MEMCG
1611 SLUB creates a directory under /sys/kernel/slab for each
1612 allocation cache to host info and debug files. If memory
1613 cgroup is enabled, each cache can have per memory cgroup
1614 caches. SLUB can create the same sysfs directories for these
1615 caches under /sys/kernel/slab/CACHE/cgroup but it can lead
1616 to a very high number of debug files being created. This is
1617 controlled by slub_memcg_sysfs boot parameter and this
1618 config option determines the parameter's default value.
1621 bool "Disable heap randomization"
1624 Randomizing heap placement makes heap exploits harder, but it
1625 also breaks ancient binaries (including anything libc5 based).
1626 This option changes the bootup default to heap randomization
1627 disabled, and can be overridden at runtime by setting
1628 /proc/sys/kernel/randomize_va_space to 2.
1630 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1633 prompt "Choose SLAB allocator"
1636 This option allows to select a slab allocator.
1640 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1642 The regular slab allocator that is established and known to work
1643 well in all environments. It organizes cache hot objects in
1644 per cpu and per node queues.
1647 bool "SLUB (Unqueued Allocator)"
1648 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1650 SLUB is a slab allocator that minimizes cache line usage
1651 instead of managing queues of cached objects (SLAB approach).
1652 Per cpu caching is realized using slabs of objects instead
1653 of queues of objects. SLUB can use memory efficiently
1654 and has enhanced diagnostics. SLUB is the default choice for
1659 bool "SLOB (Simple Allocator)"
1661 SLOB replaces the stock allocator with a drastically simpler
1662 allocator. SLOB is generally more space efficient but
1663 does not perform as well on large systems.
1667 config SLAB_MERGE_DEFAULT
1668 bool "Allow slab caches to be merged"
1671 For reduced kernel memory fragmentation, slab caches can be
1672 merged when they share the same size and other characteristics.
1673 This carries a risk of kernel heap overflows being able to
1674 overwrite objects from merged caches (and more easily control
1675 cache layout), which makes such heap attacks easier to exploit
1676 by attackers. By keeping caches unmerged, these kinds of exploits
1677 can usually only damage objects in the same cache. To disable
1678 merging at runtime, "slab_nomerge" can be passed on the kernel
1681 config SLAB_FREELIST_RANDOM
1683 depends on SLAB || SLUB
1684 bool "SLAB freelist randomization"
1686 Randomizes the freelist order used on creating new pages. This
1687 security feature reduces the predictability of the kernel slab
1688 allocator against heap overflows.
1690 config SLAB_FREELIST_HARDENED
1691 bool "Harden slab freelist metadata"
1694 Many kernel heap attacks try to target slab cache metadata and
1695 other infrastructure. This options makes minor performance
1696 sacrifies to harden the kernel slab allocator against common
1697 freelist exploit methods.
1699 config SLUB_CPU_PARTIAL
1701 depends on SLUB && SMP
1702 bool "SLUB per cpu partial cache"
1704 Per cpu partial caches accellerate objects allocation and freeing
1705 that is local to a processor at the price of more indeterminism
1706 in the latency of the free. On overflow these caches will be cleared
1707 which requires the taking of locks that may cause latency spikes.
1708 Typically one would choose no for a realtime system.
1710 config MMAP_ALLOW_UNINITIALIZED
1711 bool "Allow mmapped anonymous memory to be uninitialized"
1712 depends on EXPERT && !MMU
1715 Normally, and according to the Linux spec, anonymous memory obtained
1716 from mmap() has its contents cleared before it is passed to
1717 userspace. Enabling this config option allows you to request that
1718 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
1719 providing a huge performance boost. If this option is not enabled,
1720 then the flag will be ignored.
1722 This is taken advantage of by uClibc's malloc(), and also by
1723 ELF-FDPIC binfmt's brk and stack allocator.
1725 Because of the obvious security issues, this option should only be
1726 enabled on embedded devices where you control what is run in
1727 userspace. Since that isn't generally a problem on no-MMU systems,
1728 it is normally safe to say Y here.
1730 See Documentation/nommu-mmap.txt for more information.
1732 config SYSTEM_DATA_VERIFICATION
1734 select SYSTEM_TRUSTED_KEYRING
1738 select ASYMMETRIC_KEY_TYPE
1739 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE
1742 select X509_CERTIFICATE_PARSER
1743 select PKCS7_MESSAGE_PARSER
1745 Provide PKCS#7 message verification using the contents of the system
1746 trusted keyring to provide public keys. This then can be used for
1747 module verification, kexec image verification and firmware blob
1751 bool "Profiling support"
1753 Say Y here to enable the extended profiling support mechanisms used
1754 by profilers such as OProfile.
1757 # Place an empty function call at each tracepoint site. Can be
1758 # dynamically changed for a probe function.
1763 endmenu # General setup
1765 source "arch/Kconfig"
1772 default 0 if BASE_FULL
1773 default 1 if !BASE_FULL
1776 bool "Enable loadable module support"
1779 Kernel modules are small pieces of compiled code which can
1780 be inserted in the running kernel, rather than being
1781 permanently built into the kernel. You use the "modprobe"
1782 tool to add (and sometimes remove) them. If you say Y here,
1783 many parts of the kernel can be built as modules (by
1784 answering M instead of Y where indicated): this is most
1785 useful for infrequently used options which are not required
1786 for booting. For more information, see the man pages for
1787 modprobe, lsmod, modinfo, insmod and rmmod.
1789 If you say Y here, you will need to run "make
1790 modules_install" to put the modules under /lib/modules/
1791 where modprobe can find them (you may need to be root to do
1798 config MODULE_FORCE_LOAD
1799 bool "Forced module loading"
1802 Allow loading of modules without version information (ie. modprobe
1803 --force). Forced module loading sets the 'F' (forced) taint flag and
1804 is usually a really bad idea.
1806 config MODULE_UNLOAD
1807 bool "Module unloading"
1809 Without this option you will not be able to unload any
1810 modules (note that some modules may not be unloadable
1811 anyway), which makes your kernel smaller, faster
1812 and simpler. If unsure, say Y.
1814 config MODULE_FORCE_UNLOAD
1815 bool "Forced module unloading"
1816 depends on MODULE_UNLOAD
1818 This option allows you to force a module to unload, even if the
1819 kernel believes it is unsafe: the kernel will remove the module
1820 without waiting for anyone to stop using it (using the -f option to
1821 rmmod). This is mainly for kernel developers and desperate users.
1825 bool "Module versioning support"
1827 Usually, you have to use modules compiled with your kernel.
1828 Saying Y here makes it sometimes possible to use modules
1829 compiled for different kernels, by adding enough information
1830 to the modules to (hopefully) spot any changes which would
1831 make them incompatible with the kernel you are running. If
1834 config MODULE_REL_CRCS
1836 depends on MODVERSIONS
1838 config MODULE_SRCVERSION_ALL
1839 bool "Source checksum for all modules"
1841 Modules which contain a MODULE_VERSION get an extra "srcversion"
1842 field inserted into their modinfo section, which contains a
1843 sum of the source files which made it. This helps maintainers
1844 see exactly which source was used to build a module (since
1845 others sometimes change the module source without updating
1846 the version). With this option, such a "srcversion" field
1847 will be created for all modules. If unsure, say N.
1850 bool "Module signature verification"
1852 select SYSTEM_DATA_VERIFICATION
1854 Check modules for valid signatures upon load: the signature
1855 is simply appended to the module. For more information see
1856 <file:Documentation/admin-guide/module-signing.rst>.
1858 Note that this option adds the OpenSSL development packages as a
1859 kernel build dependency so that the signing tool can use its crypto
1862 !!!WARNING!!! If you enable this option, you MUST make sure that the
1863 module DOES NOT get stripped after being signed. This includes the
1864 debuginfo strip done by some packagers (such as rpmbuild) and
1865 inclusion into an initramfs that wants the module size reduced.
1867 config MODULE_SIG_FORCE
1868 bool "Require modules to be validly signed"
1869 depends on MODULE_SIG
1871 Reject unsigned modules or signed modules for which we don't have a
1872 key. Without this, such modules will simply taint the kernel.
1874 config MODULE_SIG_ALL
1875 bool "Automatically sign all modules"
1877 depends on MODULE_SIG
1879 Sign all modules during make modules_install. Without this option,
1880 modules must be signed manually, using the scripts/sign-file tool.
1882 comment "Do not forget to sign required modules with scripts/sign-file"
1883 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL
1886 prompt "Which hash algorithm should modules be signed with?"
1887 depends on MODULE_SIG
1889 This determines which sort of hashing algorithm will be used during
1890 signature generation. This algorithm _must_ be built into the kernel
1891 directly so that signature verification can take place. It is not
1892 possible to load a signed module containing the algorithm to check
1893 the signature on that module.
1895 config MODULE_SIG_SHA1
1896 bool "Sign modules with SHA-1"
1899 config MODULE_SIG_SHA224
1900 bool "Sign modules with SHA-224"
1901 select CRYPTO_SHA256
1903 config MODULE_SIG_SHA256
1904 bool "Sign modules with SHA-256"
1905 select CRYPTO_SHA256
1907 config MODULE_SIG_SHA384
1908 bool "Sign modules with SHA-384"
1909 select CRYPTO_SHA512
1911 config MODULE_SIG_SHA512
1912 bool "Sign modules with SHA-512"
1913 select CRYPTO_SHA512
1917 config MODULE_SIG_HASH
1919 depends on MODULE_SIG
1920 default "sha1" if MODULE_SIG_SHA1
1921 default "sha224" if MODULE_SIG_SHA224
1922 default "sha256" if MODULE_SIG_SHA256
1923 default "sha384" if MODULE_SIG_SHA384
1924 default "sha512" if MODULE_SIG_SHA512
1926 config MODULE_COMPRESS
1927 bool "Compress modules on installation"
1931 Compresses kernel modules when 'make modules_install' is run; gzip or
1932 xz depending on "Compression algorithm" below.
1934 module-init-tools MAY support gzip, and kmod MAY support gzip and xz.
1936 Out-of-tree kernel modules installed using Kbuild will also be
1937 compressed upon installation.
1939 Note: for modules inside an initrd or initramfs, it's more efficient
1940 to compress the whole initrd or initramfs instead.
1942 Note: This is fully compatible with signed modules.
1947 prompt "Compression algorithm"
1948 depends on MODULE_COMPRESS
1949 default MODULE_COMPRESS_GZIP
1951 This determines which sort of compression will be used during
1952 'make modules_install'.
1954 GZIP (default) and XZ are supported.
1956 config MODULE_COMPRESS_GZIP
1959 config MODULE_COMPRESS_XZ
1964 config TRIM_UNUSED_KSYMS
1965 bool "Trim unused exported kernel symbols"
1966 depends on MODULES && !UNUSED_SYMBOLS
1968 The kernel and some modules make many symbols available for
1969 other modules to use via EXPORT_SYMBOL() and variants. Depending
1970 on the set of modules being selected in your kernel configuration,
1971 many of those exported symbols might never be used.
1973 This option allows for unused exported symbols to be dropped from
1974 the build. In turn, this provides the compiler more opportunities
1975 (especially when using LTO) for optimizing the code and reducing
1976 binary size. This might have some security advantages as well.
1978 If unsure, or if you need to build out-of-tree modules, say N.
1982 config MODULES_TREE_LOOKUP
1984 depends on PERF_EVENTS || TRACING
1986 config INIT_ALL_POSSIBLE
1989 Back when each arch used to define their own cpu_online_mask and
1990 cpu_possible_mask, some of them chose to initialize cpu_possible_mask
1991 with all 1s, and others with all 0s. When they were centralised,
1992 it was better to provide this option than to break all the archs
1993 and have several arch maintainers pursuing me down dark alleys.
1995 source "block/Kconfig"
1997 config PREEMPT_NOTIFIERS
2007 Build a simple ASN.1 grammar compiler that produces a bytecode output
2008 that can be interpreted by the ASN.1 stream decoder and used to
2009 inform it as to what tags are to be expected in a stream and what
2010 functions to call on what tags.
2012 source "kernel/Kconfig.locks"
2014 config ARCH_HAS_SYNC_CORE_BEFORE_USERMODE
2017 # It may be useful for an architecture to override the definitions of the
2018 # SYSCALL_DEFINE() and __SYSCALL_DEFINEx() macros in <linux/syscalls.h>
2019 # and the COMPAT_ variants in <linux/compat.h>, in particular to use a
2020 # different calling convention for syscalls. They can also override the
2021 # macros for not-implemented syscalls in kernel/sys_ni.c and
2022 # kernel/time/posix-stubs.c. All these overrides need to be available in
2023 # <asm/syscall_wrapper.h>.
2024 config ARCH_HAS_SYSCALL_WRAPPER