5 default "/lib/modules/$(shell,uname --release)/.config"
6 default "/etc/kernel-config"
7 default "/boot/config-$(shell,uname --release)"
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))
33 config BUILDTIME_EXTABLE_SORT
36 config THREAD_INFO_IN_TASK
39 Select this to move thread_info off the stack into task_struct. To
40 make this work, an arch will need to remove all thread_info fields
41 except flags and fix any runtime bugs.
43 One subtle change that will be needed is to use try_get_task_stack()
44 and put_task_stack() in save_thread_stack_tsk() and get_wchan().
53 depends on BROKEN || !SMP
56 config INIT_ENV_ARG_LIMIT
61 Maximum of each of the number of arguments and environment
62 variables passed to init from the kernel command line.
65 bool "Compile also drivers which will not load"
69 Some drivers can be compiled on a different platform than they are
70 intended to be run on. Despite they cannot be loaded there (or even
71 when they load they cannot be used due to missing HW support),
72 developers still, opposing to distributors, might want to build such
73 drivers to compile-test them.
75 If you are a developer and want to build everything available, say Y
76 here. If you are a user/distributor, say N here to exclude useless
77 drivers to be distributed.
80 string "Local version - append to kernel release"
82 Append an extra string to the end of your kernel version.
83 This will show up when you type uname, for example.
84 The string you set here will be appended after the contents of
85 any files with a filename matching localversion* in your
86 object and source tree, in that order. Your total string can
87 be a maximum of 64 characters.
89 config LOCALVERSION_AUTO
90 bool "Automatically append version information to the version string"
92 depends on !COMPILE_TEST
94 This will try to automatically determine if the current tree is a
95 release tree by looking for git tags that belong to the current
98 A string of the format -gxxxxxxxx will be added to the localversion
99 if a git-based tree is found. The string generated by this will be
100 appended after any matching localversion* files, and after the value
101 set in CONFIG_LOCALVERSION.
103 (The actual string used here is the first eight characters produced
104 by running the command:
106 $ git rev-parse --verify HEAD
108 which is done within the script "scripts/setlocalversion".)
110 config HAVE_KERNEL_GZIP
113 config HAVE_KERNEL_BZIP2
116 config HAVE_KERNEL_LZMA
119 config HAVE_KERNEL_XZ
122 config HAVE_KERNEL_LZO
125 config HAVE_KERNEL_LZ4
129 prompt "Kernel compression mode"
131 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4
133 The linux kernel is a kind of self-extracting executable.
134 Several compression algorithms are available, which differ
135 in efficiency, compression and decompression speed.
136 Compression speed is only relevant when building a kernel.
137 Decompression speed is relevant at each boot.
139 If you have any problems with bzip2 or lzma compressed
140 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
141 version of this functionality (bzip2 only), for 2.4, was
142 supplied by Christian Ludwig)
144 High compression options are mostly useful for users, who
145 are low on disk space (embedded systems), but for whom ram
148 If in doubt, select 'gzip'
152 depends on HAVE_KERNEL_GZIP
154 The old and tried gzip compression. It provides a good balance
155 between compression ratio and decompression speed.
159 depends on HAVE_KERNEL_BZIP2
161 Its compression ratio and speed is intermediate.
162 Decompression speed is slowest among the choices. The kernel
163 size is about 10% smaller with bzip2, in comparison to gzip.
164 Bzip2 uses a large amount of memory. For modern kernels you
165 will need at least 8MB RAM or more for booting.
169 depends on HAVE_KERNEL_LZMA
171 This compression algorithm's ratio is best. Decompression speed
172 is between gzip and bzip2. Compression is slowest.
173 The kernel size is about 33% smaller with LZMA in comparison to gzip.
177 depends on HAVE_KERNEL_XZ
179 XZ uses the LZMA2 algorithm and instruction set specific
180 BCJ filters which can improve compression ratio of executable
181 code. The size of the kernel is about 30% smaller with XZ in
182 comparison to gzip. On architectures for which there is a BCJ
183 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ
184 will create a few percent smaller kernel than plain LZMA.
186 The speed is about the same as with LZMA: The decompression
187 speed of XZ is better than that of bzip2 but worse than gzip
188 and LZO. Compression is slow.
192 depends on HAVE_KERNEL_LZO
194 Its compression ratio is the poorest among the choices. The kernel
195 size is about 10% bigger than gzip; however its speed
196 (both compression and decompression) is the fastest.
200 depends on HAVE_KERNEL_LZ4
202 LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding.
203 A preliminary version of LZ4 de/compression tool is available at
204 <https://code.google.com/p/lz4/>.
206 Its compression ratio is worse than LZO. The size of the kernel
207 is about 8% bigger than LZO. But the decompression speed is
212 config DEFAULT_HOSTNAME
213 string "Default hostname"
216 This option determines the default system hostname before userspace
217 calls sethostname(2). The kernel traditionally uses "(none)" here,
218 but you may wish to use a different default here to make a minimal
219 system more usable with less configuration.
222 bool "Support for paging of anonymous memory (swap)"
223 depends on MMU && BLOCK
226 This option allows you to choose whether you want to have support
227 for so called swap devices or swap files in your kernel that are
228 used to provide more virtual memory than the actual RAM present
229 in your computer. If unsure say Y.
234 Inter Process Communication is a suite of library functions and
235 system calls which let processes (running programs) synchronize and
236 exchange information. It is generally considered to be a good thing,
237 and some programs won't run unless you say Y here. In particular, if
238 you want to run the DOS emulator dosemu under Linux (read the
239 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
240 you'll need to say Y here.
242 You can find documentation about IPC with "info ipc" and also in
243 section 6.4 of the Linux Programmer's Guide, available from
244 <http://www.tldp.org/guides.html>.
246 config SYSVIPC_SYSCTL
253 bool "POSIX Message Queues"
256 POSIX variant of message queues is a part of IPC. In POSIX message
257 queues every message has a priority which decides about succession
258 of receiving it by a process. If you want to compile and run
259 programs written e.g. for Solaris with use of its POSIX message
260 queues (functions mq_*) say Y here.
262 POSIX message queues are visible as a filesystem called 'mqueue'
263 and can be mounted somewhere if you want to do filesystem
264 operations on message queues.
268 config POSIX_MQUEUE_SYSCTL
270 depends on POSIX_MQUEUE
274 config CROSS_MEMORY_ATTACH
275 bool "Enable process_vm_readv/writev syscalls"
279 Enabling this option adds the system calls process_vm_readv and
280 process_vm_writev which allow a process with the correct privileges
281 to directly read from or write to another process' address space.
282 See the man page for more details.
285 bool "uselib syscall"
286 def_bool ALPHA || M68K || SPARC || X86_32 || IA32_EMULATION
288 This option enables the uselib syscall, a system call used in the
289 dynamic linker from libc5 and earlier. glibc does not use this
290 system call. If you intend to run programs built on libc5 or
291 earlier, you may need to enable this syscall. Current systems
292 running glibc can safely disable this.
295 bool "Auditing support"
298 Enable auditing infrastructure that can be used with another
299 kernel subsystem, such as SELinux (which requires this for
300 logging of avc messages output). System call auditing is included
301 on architectures which support it.
303 config HAVE_ARCH_AUDITSYSCALL
308 depends on AUDIT && HAVE_ARCH_AUDITSYSCALL
312 depends on AUDITSYSCALL
317 depends on AUDITSYSCALL
320 source "kernel/irq/Kconfig"
321 source "kernel/time/Kconfig"
323 menu "CPU/Task time and stats accounting"
325 config VIRT_CPU_ACCOUNTING
329 prompt "Cputime accounting"
330 default TICK_CPU_ACCOUNTING if !PPC64
331 default VIRT_CPU_ACCOUNTING_NATIVE if PPC64
333 # Kind of a stub config for the pure tick based cputime accounting
334 config TICK_CPU_ACCOUNTING
335 bool "Simple tick based cputime accounting"
336 depends on !S390 && !NO_HZ_FULL
338 This is the basic tick based cputime accounting that maintains
339 statistics about user, system and idle time spent on per jiffies
344 config VIRT_CPU_ACCOUNTING_NATIVE
345 bool "Deterministic task and CPU time accounting"
346 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL
347 select VIRT_CPU_ACCOUNTING
349 Select this option to enable more accurate task and CPU time
350 accounting. This is done by reading a CPU counter on each
351 kernel entry and exit and on transitions within the kernel
352 between system, softirq and hardirq state, so there is a
353 small performance impact. In the case of s390 or IBM POWER > 5,
354 this also enables accounting of stolen time on logically-partitioned
357 config VIRT_CPU_ACCOUNTING_GEN
358 bool "Full dynticks CPU time accounting"
359 depends on HAVE_CONTEXT_TRACKING
360 depends on HAVE_VIRT_CPU_ACCOUNTING_GEN
361 select VIRT_CPU_ACCOUNTING
362 select CONTEXT_TRACKING
364 Select this option to enable task and CPU time accounting on full
365 dynticks systems. This accounting is implemented by watching every
366 kernel-user boundaries using the context tracking subsystem.
367 The accounting is thus performed at the expense of some significant
370 For now this is only useful if you are working on the full
371 dynticks subsystem development.
377 config IRQ_TIME_ACCOUNTING
378 bool "Fine granularity task level IRQ time accounting"
379 depends on HAVE_IRQ_TIME_ACCOUNTING && !VIRT_CPU_ACCOUNTING_NATIVE
381 Select this option to enable fine granularity task irq time
382 accounting. This is done by reading a timestamp on each
383 transitions between softirq and hardirq state, so there can be a
384 small performance impact.
386 If in doubt, say N here.
388 config BSD_PROCESS_ACCT
389 bool "BSD Process Accounting"
392 If you say Y here, a user level program will be able to instruct the
393 kernel (via a special system call) to write process accounting
394 information to a file: whenever a process exits, information about
395 that process will be appended to the file by the kernel. The
396 information includes things such as creation time, owning user,
397 command name, memory usage, controlling terminal etc. (the complete
398 list is in the struct acct in <file:include/linux/acct.h>). It is
399 up to the user level program to do useful things with this
400 information. This is generally a good idea, so say Y.
402 config BSD_PROCESS_ACCT_V3
403 bool "BSD Process Accounting version 3 file format"
404 depends on BSD_PROCESS_ACCT
407 If you say Y here, the process accounting information is written
408 in a new file format that also logs the process IDs of each
409 process and it's parent. Note that this file format is incompatible
410 with previous v0/v1/v2 file formats, so you will need updated tools
411 for processing it. A preliminary version of these tools is available
412 at <http://www.gnu.org/software/acct/>.
415 bool "Export task/process statistics through netlink"
420 Export selected statistics for tasks/processes through the
421 generic netlink interface. Unlike BSD process accounting, the
422 statistics are available during the lifetime of tasks/processes as
423 responses to commands. Like BSD accounting, they are sent to user
428 config TASK_DELAY_ACCT
429 bool "Enable per-task delay accounting"
433 Collect information on time spent by a task waiting for system
434 resources like cpu, synchronous block I/O completion and swapping
435 in pages. Such statistics can help in setting a task's priorities
436 relative to other tasks for cpu, io, rss limits etc.
441 bool "Enable extended accounting over taskstats"
444 Collect extended task accounting data and send the data
445 to userland for processing over the taskstats interface.
449 config TASK_IO_ACCOUNTING
450 bool "Enable per-task storage I/O accounting"
451 depends on TASK_XACCT
453 Collect information on the number of bytes of storage I/O which this
458 endmenu # "CPU/Task time and stats accounting"
462 depends on SMP || COMPILE_TEST
465 Make sure that CPUs running critical tasks are not disturbed by
466 any source of "noise" such as unbound workqueues, timers, kthreads...
467 Unbound jobs get offloaded to housekeeping CPUs. This is driven by
468 the "isolcpus=" boot parameter.
472 source "kernel/rcu/Kconfig"
479 tristate "Kernel .config support"
482 This option enables the complete Linux kernel ".config" file
483 contents to be saved in the kernel. It provides documentation
484 of which kernel options are used in a running kernel or in an
485 on-disk kernel. This information can be extracted from the kernel
486 image file with the script scripts/extract-ikconfig and used as
487 input to rebuild the current kernel or to build another kernel.
488 It can also be extracted from a running kernel by reading
489 /proc/config.gz if enabled (below).
492 bool "Enable access to .config through /proc/config.gz"
493 depends on IKCONFIG && PROC_FS
495 This option enables access to the kernel configuration file
496 through /proc/config.gz.
499 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
504 Select the minimal kernel log buffer size as a power of 2.
505 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config
506 parameter, see below. Any higher size also might be forced
507 by "log_buf_len" boot parameter.
517 config LOG_CPU_MAX_BUF_SHIFT
518 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)"
521 default 12 if !BASE_SMALL
522 default 0 if BASE_SMALL
525 This option allows to increase the default ring buffer size
526 according to the number of CPUs. The value defines the contribution
527 of each CPU as a power of 2. The used space is typically only few
528 lines however it might be much more when problems are reported,
531 The increased size means that a new buffer has to be allocated and
532 the original static one is unused. It makes sense only on systems
533 with more CPUs. Therefore this value is used only when the sum of
534 contributions is greater than the half of the default kernel ring
535 buffer as defined by LOG_BUF_SHIFT. The default values are set
536 so that more than 64 CPUs are needed to trigger the allocation.
538 Also this option is ignored when "log_buf_len" kernel parameter is
539 used as it forces an exact (power of two) size of the ring buffer.
541 The number of possible CPUs is used for this computation ignoring
542 hotplugging making the computation optimal for the worst case
543 scenario while allowing a simple algorithm to be used from bootup.
545 Examples shift values and their meaning:
546 17 => 128 KB for each CPU
547 16 => 64 KB for each CPU
548 15 => 32 KB for each CPU
549 14 => 16 KB for each CPU
550 13 => 8 KB for each CPU
551 12 => 4 KB for each CPU
553 config PRINTK_SAFE_LOG_BUF_SHIFT
554 int "Temporary per-CPU printk log buffer size (12 => 4KB, 13 => 8KB)"
559 Select the size of an alternate printk per-CPU buffer where messages
560 printed from usafe contexts are temporary stored. One example would
561 be NMI messages, another one - printk recursion. The messages are
562 copied to the main log buffer in a safe context to avoid a deadlock.
563 The value defines the size as a power of 2.
565 Those messages are rare and limited. The largest one is when
566 a backtrace is printed. It usually fits into 4KB. Select
567 8KB if you want to be on the safe side.
570 17 => 128 KB for each CPU
571 16 => 64 KB for each CPU
572 15 => 32 KB for each CPU
573 14 => 16 KB for each CPU
574 13 => 8 KB for each CPU
575 12 => 4 KB for each CPU
578 # Architectures with an unreliable sched_clock() should select this:
580 config HAVE_UNSTABLE_SCHED_CLOCK
583 config GENERIC_SCHED_CLOCK
587 # For architectures that want to enable the support for NUMA-affine scheduler
590 config ARCH_SUPPORTS_NUMA_BALANCING
594 # For architectures that prefer to flush all TLBs after a number of pages
595 # are unmapped instead of sending one IPI per page to flush. The architecture
596 # must provide guarantees on what happens if a clean TLB cache entry is
597 # written after the unmap. Details are in mm/rmap.c near the check for
598 # should_defer_flush. The architecture should also consider if the full flush
599 # and the refill costs are offset by the savings of sending fewer IPIs.
600 config ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
604 # For architectures that know their GCC __int128 support is sound
606 config ARCH_SUPPORTS_INT128
609 # For architectures that (ab)use NUMA to represent different memory regions
610 # all cpu-local but of different latencies, such as SuperH.
612 config ARCH_WANT_NUMA_VARIABLE_LOCALITY
615 config NUMA_BALANCING
616 bool "Memory placement aware NUMA scheduler"
617 depends on ARCH_SUPPORTS_NUMA_BALANCING
618 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
619 depends on SMP && NUMA && MIGRATION
621 This option adds support for automatic NUMA aware memory/task placement.
622 The mechanism is quite primitive and is based on migrating memory when
623 it has references to the node the task is running on.
625 This system will be inactive on UMA systems.
627 config NUMA_BALANCING_DEFAULT_ENABLED
628 bool "Automatically enable NUMA aware memory/task placement"
630 depends on NUMA_BALANCING
632 If set, automatic NUMA balancing will be enabled if running on a NUMA
636 bool "Control Group support"
639 This option adds support for grouping sets of processes together, for
640 use with process control subsystems such as Cpusets, CFS, memory
641 controls or device isolation.
643 - Documentation/scheduler/sched-design-CFS.txt (CFS)
644 - Documentation/cgroup-v1/ (features for grouping, isolation
645 and resource control)
655 bool "Memory controller"
659 Provides control over the memory footprint of tasks in a cgroup.
662 bool "Swap controller"
663 depends on MEMCG && SWAP
665 Provides control over the swap space consumed by tasks in a cgroup.
667 config MEMCG_SWAP_ENABLED
668 bool "Swap controller enabled by default"
669 depends on MEMCG_SWAP
672 Memory Resource Controller Swap Extension comes with its price in
673 a bigger memory consumption. General purpose distribution kernels
674 which want to enable the feature but keep it disabled by default
675 and let the user enable it by swapaccount=1 boot command line
676 parameter should have this option unselected.
677 For those who want to have the feature enabled by default should
678 select this option (if, for some reason, they need to disable it
679 then swapaccount=0 does the trick).
686 Generic block IO controller cgroup interface. This is the common
687 cgroup interface which should be used by various IO controlling
690 Currently, CFQ IO scheduler uses it to recognize task groups and
691 control disk bandwidth allocation (proportional time slice allocation)
692 to such task groups. It is also used by bio throttling logic in
693 block layer to implement upper limit in IO rates on a device.
695 This option only enables generic Block IO controller infrastructure.
696 One needs to also enable actual IO controlling logic/policy. For
697 enabling proportional weight division of disk bandwidth in CFQ, set
698 CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set
699 CONFIG_BLK_DEV_THROTTLING=y.
701 See Documentation/cgroup-v1/blkio-controller.txt for more information.
703 config DEBUG_BLK_CGROUP
704 bool "IO controller debugging"
705 depends on BLK_CGROUP
708 Enable some debugging help. Currently it exports additional stat
709 files in a cgroup which can be useful for debugging.
711 config CGROUP_WRITEBACK
713 depends on MEMCG && BLK_CGROUP
716 menuconfig CGROUP_SCHED
717 bool "CPU controller"
720 This feature lets CPU scheduler recognize task groups and control CPU
721 bandwidth allocation to such task groups. It uses cgroups to group
725 config FAIR_GROUP_SCHED
726 bool "Group scheduling for SCHED_OTHER"
727 depends on CGROUP_SCHED
731 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED"
732 depends on FAIR_GROUP_SCHED
735 This option allows users to define CPU bandwidth rates (limits) for
736 tasks running within the fair group scheduler. Groups with no limit
737 set are considered to be unconstrained and will run with no
739 See Documentation/scheduler/sched-bwc.txt for more information.
741 config RT_GROUP_SCHED
742 bool "Group scheduling for SCHED_RR/FIFO"
743 depends on CGROUP_SCHED
746 This feature lets you explicitly allocate real CPU bandwidth
747 to task groups. If enabled, it will also make it impossible to
748 schedule realtime tasks for non-root users until you allocate
749 realtime bandwidth for them.
750 See Documentation/scheduler/sched-rt-group.txt for more information.
755 bool "PIDs controller"
757 Provides enforcement of process number limits in the scope of a
758 cgroup. Any attempt to fork more processes than is allowed in the
759 cgroup will fail. PIDs are fundamentally a global resource because it
760 is fairly trivial to reach PID exhaustion before you reach even a
761 conservative kmemcg limit. As a result, it is possible to grind a
762 system to halt without being limited by other cgroup policies. The
763 PIDs controller is designed to stop this from happening.
765 It should be noted that organisational operations (such as attaching
766 to a cgroup hierarchy will *not* be blocked by the PIDs controller),
767 since the PIDs limit only affects a process's ability to fork, not to
771 bool "RDMA controller"
773 Provides enforcement of RDMA resources defined by IB stack.
774 It is fairly easy for consumers to exhaust RDMA resources, which
775 can result into resource unavailability to other consumers.
776 RDMA controller is designed to stop this from happening.
777 Attaching processes with active RDMA resources to the cgroup
778 hierarchy is allowed even if can cross the hierarchy's limit.
780 config CGROUP_FREEZER
781 bool "Freezer controller"
783 Provides a way to freeze and unfreeze all tasks in a
786 This option affects the ORIGINAL cgroup interface. The cgroup2 memory
787 controller includes important in-kernel memory consumers per default.
789 If you're using cgroup2, say N.
791 config CGROUP_HUGETLB
792 bool "HugeTLB controller"
793 depends on HUGETLB_PAGE
797 Provides a cgroup controller for HugeTLB pages.
798 When you enable this, you can put a per cgroup limit on HugeTLB usage.
799 The limit is enforced during page fault. Since HugeTLB doesn't
800 support page reclaim, enforcing the limit at page fault time implies
801 that, the application will get SIGBUS signal if it tries to access
802 HugeTLB pages beyond its limit. This requires the application to know
803 beforehand how much HugeTLB pages it would require for its use. The
804 control group is tracked in the third page lru pointer. This means
805 that we cannot use the controller with huge page less than 3 pages.
808 bool "Cpuset controller"
811 This option will let you create and manage CPUSETs which
812 allow dynamically partitioning a system into sets of CPUs and
813 Memory Nodes and assigning tasks to run only within those sets.
814 This is primarily useful on large SMP or NUMA systems.
818 config PROC_PID_CPUSET
819 bool "Include legacy /proc/<pid>/cpuset file"
824 bool "Device controller"
826 Provides a cgroup controller implementing whitelists for
827 devices which a process in the cgroup can mknod or open.
829 config CGROUP_CPUACCT
830 bool "Simple CPU accounting controller"
832 Provides a simple controller for monitoring the
833 total CPU consumed by the tasks in a cgroup.
836 bool "Perf controller"
837 depends on PERF_EVENTS
839 This option extends the perf per-cpu mode to restrict monitoring
840 to threads which belong to the cgroup specified and run on the
846 bool "Support for eBPF programs attached to cgroups"
847 depends on BPF_SYSCALL
848 select SOCK_CGROUP_DATA
850 Allow attaching eBPF programs to a cgroup using the bpf(2)
851 syscall command BPF_PROG_ATTACH.
853 In which context these programs are accessed depends on the type
854 of attachment. For instance, programs that are attached using
855 BPF_CGROUP_INET_INGRESS will be executed on the ingress path of
859 bool "Debug controller"
861 depends on DEBUG_KERNEL
863 This option enables a simple controller that exports
864 debugging information about the cgroups framework. This
865 controller is for control cgroup debugging only. Its
866 interfaces are not stable.
870 config SOCK_CGROUP_DATA
876 menuconfig NAMESPACES
877 bool "Namespaces support" if EXPERT
881 Provides the way to make tasks work with different objects using
882 the same id. For example same IPC id may refer to different objects
883 or same user id or pid may refer to different tasks when used in
884 different namespaces.
892 In this namespace tasks see different info provided with the
897 depends on (SYSVIPC || POSIX_MQUEUE)
900 In this namespace tasks work with IPC ids which correspond to
901 different IPC objects in different namespaces.
904 bool "User namespace"
907 This allows containers, i.e. vservers, to use user namespaces
908 to provide different user info for different servers.
910 When user namespaces are enabled in the kernel it is
911 recommended that the MEMCG option also be enabled and that
912 user-space use the memory control groups to limit the amount
913 of memory a memory unprivileged users can use.
918 bool "PID Namespaces"
921 Support process id namespaces. This allows having multiple
922 processes with the same pid as long as they are in different
923 pid namespaces. This is a building block of containers.
926 bool "Network namespace"
930 Allow user space to create what appear to be multiple instances
931 of the network stack.
935 config SCHED_AUTOGROUP
936 bool "Automatic process group scheduling"
939 select FAIR_GROUP_SCHED
941 This option optimizes the scheduler for common desktop workloads by
942 automatically creating and populating task groups. This separation
943 of workloads isolates aggressive CPU burners (like build jobs) from
944 desktop applications. Task group autogeneration is currently based
947 config SYSFS_DEPRECATED
948 bool "Enable deprecated sysfs features to support old userspace tools"
952 This option adds code that switches the layout of the "block" class
953 devices, to not show up in /sys/class/block/, but only in
956 This switch is only active when the sysfs.deprecated=1 boot option is
957 passed or the SYSFS_DEPRECATED_V2 option is set.
959 This option allows new kernels to run on old distributions and tools,
960 which might get confused by /sys/class/block/. Since 2007/2008 all
961 major distributions and tools handle this just fine.
963 Recent distributions and userspace tools after 2009/2010 depend on
964 the existence of /sys/class/block/, and will not work with this
967 Only if you are using a new kernel on an old distribution, you might
970 config SYSFS_DEPRECATED_V2
971 bool "Enable deprecated sysfs features by default"
974 depends on SYSFS_DEPRECATED
976 Enable deprecated sysfs by default.
978 See the CONFIG_SYSFS_DEPRECATED option for more details about this
981 Only if you are using a new kernel on an old distribution, you might
982 need to say Y here. Even then, odds are you would not need it
983 enabled, you can always pass the boot option if absolutely necessary.
986 bool "Kernel->user space relay support (formerly relayfs)"
989 This option enables support for relay interface support in
990 certain file systems (such as debugfs).
991 It is designed to provide an efficient mechanism for tools and
992 facilities to relay large amounts of data from kernel space to
997 config BLK_DEV_INITRD
998 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
1000 The initial RAM filesystem is a ramfs which is loaded by the
1001 boot loader (loadlin or lilo) and that is mounted as root
1002 before the normal boot procedure. It is typically used to
1003 load modules needed to mount the "real" root file system,
1004 etc. See <file:Documentation/admin-guide/initrd.rst> for details.
1006 If RAM disk support (BLK_DEV_RAM) is also included, this
1007 also enables initial RAM disk (initrd) support and adds
1008 15 Kbytes (more on some other architectures) to the kernel size.
1014 source "usr/Kconfig"
1019 prompt "Compiler optimization level"
1020 default CC_OPTIMIZE_FOR_PERFORMANCE
1022 config CC_OPTIMIZE_FOR_PERFORMANCE
1023 bool "Optimize for performance"
1025 This is the default optimization level for the kernel, building
1026 with the "-O2" compiler flag for best performance and most
1027 helpful compile-time warnings.
1029 config CC_OPTIMIZE_FOR_SIZE
1030 bool "Optimize for size"
1032 Enabling this option will pass "-Os" instead of "-O2" to
1033 your compiler resulting in a smaller kernel.
1039 config HAVE_LD_DEAD_CODE_DATA_ELIMINATION
1042 This requires that the arch annotates or otherwise protects
1043 its external entry points from being discarded. Linker scripts
1044 must also merge .text.*, .data.*, and .bss.* correctly into
1045 output sections. Care must be taken not to pull in unrelated
1046 sections (e.g., '.text.init'). Typically '.' in section names
1047 is used to distinguish them from label names / C identifiers.
1049 config LD_DEAD_CODE_DATA_ELIMINATION
1050 bool "Dead code and data elimination (EXPERIMENTAL)"
1051 depends on HAVE_LD_DEAD_CODE_DATA_ELIMINATION
1054 Select this if the architecture wants to do dead code and
1055 data elimination with the linker by compiling with
1056 -ffunction-sections -fdata-sections, and linking with
1059 This can reduce on disk and in-memory size of the kernel
1060 code and static data, particularly for small configs and
1061 on small systems. This has the possibility of introducing
1062 silently broken kernel if the required annotations are not
1063 present. This option is not well tested yet, so use at your
1075 config SYSCTL_EXCEPTION_TRACE
1078 Enable support for /proc/sys/debug/exception-trace.
1080 config SYSCTL_ARCH_UNALIGN_NO_WARN
1083 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap
1084 Allows arch to define/use @no_unaligned_warning to possibly warn
1085 about unaligned access emulation going on under the hood.
1087 config SYSCTL_ARCH_UNALIGN_ALLOW
1090 Enable support for /proc/sys/kernel/unaligned-trap
1091 Allows arches to define/use @unaligned_enabled to runtime toggle
1092 the unaligned access emulation.
1093 see arch/parisc/kernel/unaligned.c for reference
1095 config HAVE_PCSPKR_PLATFORM
1098 # interpreter that classic socket filters depend on
1103 bool "Configure standard kernel features (expert users)"
1104 # Unhide debug options, to make the on-by-default options visible
1107 This option allows certain base kernel options and settings
1108 to be disabled or tweaked. This is for specialized
1109 environments which can tolerate a "non-standard" kernel.
1110 Only use this if you really know what you are doing.
1113 bool "Enable 16-bit UID system calls" if EXPERT
1114 depends on HAVE_UID16 && MULTIUSER
1117 This enables the legacy 16-bit UID syscall wrappers.
1120 bool "Multiple users, groups and capabilities support" if EXPERT
1123 This option enables support for non-root users, groups and
1126 If you say N here, all processes will run with UID 0, GID 0, and all
1127 possible capabilities. Saying N here also compiles out support for
1128 system calls related to UIDs, GIDs, and capabilities, such as setuid,
1131 If unsure, say Y here.
1133 config SGETMASK_SYSCALL
1134 bool "sgetmask/ssetmask syscalls support" if EXPERT
1135 def_bool PARISC || M68K || PPC || MIPS || X86 || SPARC || MICROBLAZE || SUPERH
1137 sys_sgetmask and sys_ssetmask are obsolete system calls
1138 no longer supported in libc but still enabled by default in some
1141 If unsure, leave the default option here.
1143 config SYSFS_SYSCALL
1144 bool "Sysfs syscall support" if EXPERT
1147 sys_sysfs is an obsolete system call no longer supported in libc.
1148 Note that disabling this option is more secure but might break
1149 compatibility with some systems.
1151 If unsure say Y here.
1153 config SYSCTL_SYSCALL
1154 bool "Sysctl syscall support" if EXPERT
1155 depends on PROC_SYSCTL
1159 sys_sysctl uses binary paths that have been found challenging
1160 to properly maintain and use. The interface in /proc/sys
1161 using paths with ascii names is now the primary path to this
1164 Almost nothing using the binary sysctl interface so if you are
1165 trying to save some space it is probably safe to disable this,
1166 making your kernel marginally smaller.
1168 If unsure say N here.
1171 bool "open by fhandle syscalls" if EXPERT
1175 If you say Y here, a user level program will be able to map
1176 file names to handle and then later use the handle for
1177 different file system operations. This is useful in implementing
1178 userspace file servers, which now track files using handles instead
1179 of names. The handle would remain the same even if file names
1180 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2)
1184 bool "Posix Clocks & timers" if EXPERT
1187 This includes native support for POSIX timers to the kernel.
1188 Some embedded systems have no use for them and therefore they
1189 can be configured out to reduce the size of the kernel image.
1191 When this option is disabled, the following syscalls won't be
1192 available: timer_create, timer_gettime: timer_getoverrun,
1193 timer_settime, timer_delete, clock_adjtime, getitimer,
1194 setitimer, alarm. Furthermore, the clock_settime, clock_gettime,
1195 clock_getres and clock_nanosleep syscalls will be limited to
1196 CLOCK_REALTIME, CLOCK_MONOTONIC and CLOCK_BOOTTIME only.
1202 bool "Enable support for printk" if EXPERT
1205 This option enables normal printk support. Removing it
1206 eliminates most of the message strings from the kernel image
1207 and makes the kernel more or less silent. As this makes it
1208 very difficult to diagnose system problems, saying N here is
1209 strongly discouraged.
1217 bool "BUG() support" if EXPERT
1220 Disabling this option eliminates support for BUG and WARN, reducing
1221 the size of your kernel image and potentially quietly ignoring
1222 numerous fatal conditions. You should only consider disabling this
1223 option for embedded systems with no facilities for reporting errors.
1229 bool "Enable ELF core dumps" if EXPERT
1231 Enable support for generating core dumps. Disabling saves about 4k.
1234 config PCSPKR_PLATFORM
1235 bool "Enable PC-Speaker support" if EXPERT
1236 depends on HAVE_PCSPKR_PLATFORM
1240 This option allows to disable the internal PC-Speaker
1241 support, saving some memory.
1245 bool "Enable full-sized data structures for core" if EXPERT
1247 Disabling this option reduces the size of miscellaneous core
1248 kernel data structures. This saves memory on small machines,
1249 but may reduce performance.
1252 bool "Enable futex support" if EXPERT
1256 Disabling this option will cause the kernel to be built without
1257 support for "fast userspace mutexes". The resulting kernel may not
1258 run glibc-based applications correctly.
1262 depends on FUTEX && RT_MUTEXES
1265 config HAVE_FUTEX_CMPXCHG
1269 Architectures should select this if futex_atomic_cmpxchg_inatomic()
1270 is implemented and always working. This removes a couple of runtime
1274 bool "Enable eventpoll support" if EXPERT
1278 Disabling this option will cause the kernel to be built without
1279 support for epoll family of system calls.
1282 bool "Enable signalfd() system call" if EXPERT
1286 Enable the signalfd() system call that allows to receive signals
1287 on a file descriptor.
1292 bool "Enable timerfd() system call" if EXPERT
1296 Enable the timerfd() system call that allows to receive timer
1297 events on a file descriptor.
1302 bool "Enable eventfd() system call" if EXPERT
1306 Enable the eventfd() system call that allows to receive both
1307 kernel notification (ie. KAIO) or userspace notifications.
1312 bool "Use full shmem filesystem" if EXPERT
1316 The shmem is an internal filesystem used to manage shared memory.
1317 It is backed by swap and manages resource limits. It is also exported
1318 to userspace as tmpfs if TMPFS is enabled. Disabling this
1319 option replaces shmem and tmpfs with the much simpler ramfs code,
1320 which may be appropriate on small systems without swap.
1323 bool "Enable AIO support" if EXPERT
1326 This option enables POSIX asynchronous I/O which may by used
1327 by some high performance threaded applications. Disabling
1328 this option saves about 7k.
1330 config ADVISE_SYSCALLS
1331 bool "Enable madvise/fadvise syscalls" if EXPERT
1334 This option enables the madvise and fadvise syscalls, used by
1335 applications to advise the kernel about their future memory or file
1336 usage, improving performance. If building an embedded system where no
1337 applications use these syscalls, you can disable this option to save
1341 bool "Enable membarrier() system call" if EXPERT
1344 Enable the membarrier() system call that allows issuing memory
1345 barriers across all running threads, which can be used to distribute
1346 the cost of user-space memory barriers asymmetrically by transforming
1347 pairs of memory barriers into pairs consisting of membarrier() and a
1352 config CHECKPOINT_RESTORE
1353 bool "Checkpoint/restore support" if EXPERT
1354 select PROC_CHILDREN
1357 Enables additional kernel features in a sake of checkpoint/restore.
1358 In particular it adds auxiliary prctl codes to setup process text,
1359 data and heap segment sizes, and a few additional /proc filesystem
1362 If unsure, say N here.
1365 bool "Load all symbols for debugging/ksymoops" if EXPERT
1368 Say Y here to let the kernel print out symbolic crash information and
1369 symbolic stack backtraces. This increases the size of the kernel
1370 somewhat, as all symbols have to be loaded into the kernel image.
1373 bool "Include all symbols in kallsyms"
1374 depends on DEBUG_KERNEL && KALLSYMS
1376 Normally kallsyms only contains the symbols of functions for nicer
1377 OOPS messages and backtraces (i.e., symbols from the text and inittext
1378 sections). This is sufficient for most cases. And only in very rare
1379 cases (e.g., when a debugger is used) all symbols are required (e.g.,
1380 names of variables from the data sections, etc).
1382 This option makes sure that all symbols are loaded into the kernel
1383 image (i.e., symbols from all sections) in cost of increased kernel
1384 size (depending on the kernel configuration, it may be 300KiB or
1385 something like this).
1387 Say N unless you really need all symbols.
1389 config KALLSYMS_ABSOLUTE_PERCPU
1392 default X86_64 && SMP
1394 config KALLSYMS_BASE_RELATIVE
1399 Instead of emitting them as absolute values in the native word size,
1400 emit the symbol references in the kallsyms table as 32-bit entries,
1401 each containing a relative value in the range [base, base + U32_MAX]
1402 or, when KALLSYMS_ABSOLUTE_PERCPU is in effect, each containing either
1403 an absolute value in the range [0, S32_MAX] or a relative value in the
1404 range [base, base + S32_MAX], where base is the lowest relative symbol
1405 address encountered in the image.
1407 On 64-bit builds, this reduces the size of the address table by 50%,
1408 but more importantly, it results in entries whose values are build
1409 time constants, and no relocation pass is required at runtime to fix
1410 up the entries based on the runtime load address of the kernel.
1412 # end of the "standard kernel features (expert users)" menu
1414 # syscall, maps, verifier
1416 bool "Enable bpf() system call"
1422 Enable the bpf() system call that allows to manipulate eBPF
1423 programs and maps via file descriptors.
1425 config BPF_JIT_ALWAYS_ON
1426 bool "Permanently enable BPF JIT and remove BPF interpreter"
1427 depends on BPF_SYSCALL && HAVE_EBPF_JIT && BPF_JIT
1429 Enables BPF JIT and removes BPF interpreter to avoid
1430 speculative execution of BPF instructions by the interpreter
1433 bool "Enable userfaultfd() system call"
1437 Enable the userfaultfd() system call that allows to intercept and
1438 handle page faults in userland.
1440 config ARCH_HAS_MEMBARRIER_CALLBACKS
1443 config ARCH_HAS_MEMBARRIER_SYNC_CORE
1447 bool "Enable rseq() system call" if EXPERT
1449 depends on HAVE_RSEQ
1452 Enable the restartable sequences system call. It provides a
1453 user-space cache for the current CPU number value, which
1454 speeds up getting the current CPU number from user-space,
1455 as well as an ABI to speed up user-space operations on
1462 bool "Enabled debugging of rseq() system call" if EXPERT
1463 depends on RSEQ && DEBUG_KERNEL
1465 Enable extra debugging checks for the rseq system call.
1470 bool "Embedded system"
1471 option allnoconfig_y
1474 This option should be enabled if compiling the kernel for
1475 an embedded system so certain expert options are available
1478 config HAVE_PERF_EVENTS
1481 See tools/perf/design.txt for details.
1483 config PERF_USE_VMALLOC
1486 See tools/perf/design.txt for details
1489 bool "PC/104 support" if EXPERT
1491 Expose PC/104 form factor device drivers and options available for
1492 selection and configuration. Enable this option if your target
1493 machine has a PC/104 bus.
1495 menu "Kernel Performance Events And Counters"
1498 bool "Kernel performance events and counters"
1499 default y if PROFILING
1500 depends on HAVE_PERF_EVENTS
1505 Enable kernel support for various performance events provided
1506 by software and hardware.
1508 Software events are supported either built-in or via the
1509 use of generic tracepoints.
1511 Most modern CPUs support performance events via performance
1512 counter registers. These registers count the number of certain
1513 types of hw events: such as instructions executed, cachemisses
1514 suffered, or branches mis-predicted - without slowing down the
1515 kernel or applications. These registers can also trigger interrupts
1516 when a threshold number of events have passed - and can thus be
1517 used to profile the code that runs on that CPU.
1519 The Linux Performance Event subsystem provides an abstraction of
1520 these software and hardware event capabilities, available via a
1521 system call and used by the "perf" utility in tools/perf/. It
1522 provides per task and per CPU counters, and it provides event
1523 capabilities on top of those.
1527 config DEBUG_PERF_USE_VMALLOC
1529 bool "Debug: use vmalloc to back perf mmap() buffers"
1530 depends on PERF_EVENTS && DEBUG_KERNEL && !PPC
1531 select PERF_USE_VMALLOC
1533 Use vmalloc memory to back perf mmap() buffers.
1535 Mostly useful for debugging the vmalloc code on platforms
1536 that don't require it.
1542 config VM_EVENT_COUNTERS
1544 bool "Enable VM event counters for /proc/vmstat" if EXPERT
1546 VM event counters are needed for event counts to be shown.
1547 This option allows the disabling of the VM event counters
1548 on EXPERT systems. /proc/vmstat will only show page counts
1549 if VM event counters are disabled.
1553 bool "Enable SLUB debugging support" if EXPERT
1554 depends on SLUB && SYSFS
1556 SLUB has extensive debug support features. Disabling these can
1557 result in significant savings in code size. This also disables
1558 SLUB sysfs support. /sys/slab will not exist and there will be
1559 no support for cache validation etc.
1561 config SLUB_MEMCG_SYSFS_ON
1563 bool "Enable memcg SLUB sysfs support by default" if EXPERT
1564 depends on SLUB && SYSFS && MEMCG
1566 SLUB creates a directory under /sys/kernel/slab for each
1567 allocation cache to host info and debug files. If memory
1568 cgroup is enabled, each cache can have per memory cgroup
1569 caches. SLUB can create the same sysfs directories for these
1570 caches under /sys/kernel/slab/CACHE/cgroup but it can lead
1571 to a very high number of debug files being created. This is
1572 controlled by slub_memcg_sysfs boot parameter and this
1573 config option determines the parameter's default value.
1576 bool "Disable heap randomization"
1579 Randomizing heap placement makes heap exploits harder, but it
1580 also breaks ancient binaries (including anything libc5 based).
1581 This option changes the bootup default to heap randomization
1582 disabled, and can be overridden at runtime by setting
1583 /proc/sys/kernel/randomize_va_space to 2.
1585 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1588 prompt "Choose SLAB allocator"
1591 This option allows to select a slab allocator.
1595 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1597 The regular slab allocator that is established and known to work
1598 well in all environments. It organizes cache hot objects in
1599 per cpu and per node queues.
1602 bool "SLUB (Unqueued Allocator)"
1603 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1605 SLUB is a slab allocator that minimizes cache line usage
1606 instead of managing queues of cached objects (SLAB approach).
1607 Per cpu caching is realized using slabs of objects instead
1608 of queues of objects. SLUB can use memory efficiently
1609 and has enhanced diagnostics. SLUB is the default choice for
1614 bool "SLOB (Simple Allocator)"
1616 SLOB replaces the stock allocator with a drastically simpler
1617 allocator. SLOB is generally more space efficient but
1618 does not perform as well on large systems.
1622 config SLAB_MERGE_DEFAULT
1623 bool "Allow slab caches to be merged"
1626 For reduced kernel memory fragmentation, slab caches can be
1627 merged when they share the same size and other characteristics.
1628 This carries a risk of kernel heap overflows being able to
1629 overwrite objects from merged caches (and more easily control
1630 cache layout), which makes such heap attacks easier to exploit
1631 by attackers. By keeping caches unmerged, these kinds of exploits
1632 can usually only damage objects in the same cache. To disable
1633 merging at runtime, "slab_nomerge" can be passed on the kernel
1636 config SLAB_FREELIST_RANDOM
1638 depends on SLAB || SLUB
1639 bool "SLAB freelist randomization"
1641 Randomizes the freelist order used on creating new pages. This
1642 security feature reduces the predictability of the kernel slab
1643 allocator against heap overflows.
1645 config SLAB_FREELIST_HARDENED
1646 bool "Harden slab freelist metadata"
1649 Many kernel heap attacks try to target slab cache metadata and
1650 other infrastructure. This options makes minor performance
1651 sacrifies to harden the kernel slab allocator against common
1652 freelist exploit methods.
1654 config SLUB_CPU_PARTIAL
1656 depends on SLUB && SMP
1657 bool "SLUB per cpu partial cache"
1659 Per cpu partial caches accellerate objects allocation and freeing
1660 that is local to a processor at the price of more indeterminism
1661 in the latency of the free. On overflow these caches will be cleared
1662 which requires the taking of locks that may cause latency spikes.
1663 Typically one would choose no for a realtime system.
1665 config MMAP_ALLOW_UNINITIALIZED
1666 bool "Allow mmapped anonymous memory to be uninitialized"
1667 depends on EXPERT && !MMU
1670 Normally, and according to the Linux spec, anonymous memory obtained
1671 from mmap() has it's contents cleared before it is passed to
1672 userspace. Enabling this config option allows you to request that
1673 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
1674 providing a huge performance boost. If this option is not enabled,
1675 then the flag will be ignored.
1677 This is taken advantage of by uClibc's malloc(), and also by
1678 ELF-FDPIC binfmt's brk and stack allocator.
1680 Because of the obvious security issues, this option should only be
1681 enabled on embedded devices where you control what is run in
1682 userspace. Since that isn't generally a problem on no-MMU systems,
1683 it is normally safe to say Y here.
1685 See Documentation/nommu-mmap.txt for more information.
1687 config SYSTEM_DATA_VERIFICATION
1689 select SYSTEM_TRUSTED_KEYRING
1693 select ASYMMETRIC_KEY_TYPE
1694 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE
1697 select X509_CERTIFICATE_PARSER
1698 select PKCS7_MESSAGE_PARSER
1700 Provide PKCS#7 message verification using the contents of the system
1701 trusted keyring to provide public keys. This then can be used for
1702 module verification, kexec image verification and firmware blob
1706 bool "Profiling support"
1708 Say Y here to enable the extended profiling support mechanisms used
1709 by profilers such as OProfile.
1712 # Place an empty function call at each tracepoint site. Can be
1713 # dynamically changed for a probe function.
1718 source "arch/Kconfig"
1720 endmenu # General setup
1722 config HAVE_GENERIC_DMA_COHERENT
1731 default 0 if BASE_FULL
1732 default 1 if !BASE_FULL
1735 bool "Enable loadable module support"
1738 Kernel modules are small pieces of compiled code which can
1739 be inserted in the running kernel, rather than being
1740 permanently built into the kernel. You use the "modprobe"
1741 tool to add (and sometimes remove) them. If you say Y here,
1742 many parts of the kernel can be built as modules (by
1743 answering M instead of Y where indicated): this is most
1744 useful for infrequently used options which are not required
1745 for booting. For more information, see the man pages for
1746 modprobe, lsmod, modinfo, insmod and rmmod.
1748 If you say Y here, you will need to run "make
1749 modules_install" to put the modules under /lib/modules/
1750 where modprobe can find them (you may need to be root to do
1757 config MODULE_FORCE_LOAD
1758 bool "Forced module loading"
1761 Allow loading of modules without version information (ie. modprobe
1762 --force). Forced module loading sets the 'F' (forced) taint flag and
1763 is usually a really bad idea.
1765 config MODULE_UNLOAD
1766 bool "Module unloading"
1768 Without this option you will not be able to unload any
1769 modules (note that some modules may not be unloadable
1770 anyway), which makes your kernel smaller, faster
1771 and simpler. If unsure, say Y.
1773 config MODULE_FORCE_UNLOAD
1774 bool "Forced module unloading"
1775 depends on MODULE_UNLOAD
1777 This option allows you to force a module to unload, even if the
1778 kernel believes it is unsafe: the kernel will remove the module
1779 without waiting for anyone to stop using it (using the -f option to
1780 rmmod). This is mainly for kernel developers and desperate users.
1784 bool "Module versioning support"
1786 Usually, you have to use modules compiled with your kernel.
1787 Saying Y here makes it sometimes possible to use modules
1788 compiled for different kernels, by adding enough information
1789 to the modules to (hopefully) spot any changes which would
1790 make them incompatible with the kernel you are running. If
1793 config MODULE_REL_CRCS
1795 depends on MODVERSIONS
1797 config MODULE_SRCVERSION_ALL
1798 bool "Source checksum for all modules"
1800 Modules which contain a MODULE_VERSION get an extra "srcversion"
1801 field inserted into their modinfo section, which contains a
1802 sum of the source files which made it. This helps maintainers
1803 see exactly which source was used to build a module (since
1804 others sometimes change the module source without updating
1805 the version). With this option, such a "srcversion" field
1806 will be created for all modules. If unsure, say N.
1809 bool "Module signature verification"
1811 select SYSTEM_DATA_VERIFICATION
1813 Check modules for valid signatures upon load: the signature
1814 is simply appended to the module. For more information see
1815 <file:Documentation/admin-guide/module-signing.rst>.
1817 Note that this option adds the OpenSSL development packages as a
1818 kernel build dependency so that the signing tool can use its crypto
1821 !!!WARNING!!! If you enable this option, you MUST make sure that the
1822 module DOES NOT get stripped after being signed. This includes the
1823 debuginfo strip done by some packagers (such as rpmbuild) and
1824 inclusion into an initramfs that wants the module size reduced.
1826 config MODULE_SIG_FORCE
1827 bool "Require modules to be validly signed"
1828 depends on MODULE_SIG
1830 Reject unsigned modules or signed modules for which we don't have a
1831 key. Without this, such modules will simply taint the kernel.
1833 config MODULE_SIG_ALL
1834 bool "Automatically sign all modules"
1836 depends on MODULE_SIG
1838 Sign all modules during make modules_install. Without this option,
1839 modules must be signed manually, using the scripts/sign-file tool.
1841 comment "Do not forget to sign required modules with scripts/sign-file"
1842 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL
1845 prompt "Which hash algorithm should modules be signed with?"
1846 depends on MODULE_SIG
1848 This determines which sort of hashing algorithm will be used during
1849 signature generation. This algorithm _must_ be built into the kernel
1850 directly so that signature verification can take place. It is not
1851 possible to load a signed module containing the algorithm to check
1852 the signature on that module.
1854 config MODULE_SIG_SHA1
1855 bool "Sign modules with SHA-1"
1858 config MODULE_SIG_SHA224
1859 bool "Sign modules with SHA-224"
1860 select CRYPTO_SHA256
1862 config MODULE_SIG_SHA256
1863 bool "Sign modules with SHA-256"
1864 select CRYPTO_SHA256
1866 config MODULE_SIG_SHA384
1867 bool "Sign modules with SHA-384"
1868 select CRYPTO_SHA512
1870 config MODULE_SIG_SHA512
1871 bool "Sign modules with SHA-512"
1872 select CRYPTO_SHA512
1876 config MODULE_SIG_HASH
1878 depends on MODULE_SIG
1879 default "sha1" if MODULE_SIG_SHA1
1880 default "sha224" if MODULE_SIG_SHA224
1881 default "sha256" if MODULE_SIG_SHA256
1882 default "sha384" if MODULE_SIG_SHA384
1883 default "sha512" if MODULE_SIG_SHA512
1885 config MODULE_COMPRESS
1886 bool "Compress modules on installation"
1890 Compresses kernel modules when 'make modules_install' is run; gzip or
1891 xz depending on "Compression algorithm" below.
1893 module-init-tools MAY support gzip, and kmod MAY support gzip and xz.
1895 Out-of-tree kernel modules installed using Kbuild will also be
1896 compressed upon installation.
1898 Note: for modules inside an initrd or initramfs, it's more efficient
1899 to compress the whole initrd or initramfs instead.
1901 Note: This is fully compatible with signed modules.
1906 prompt "Compression algorithm"
1907 depends on MODULE_COMPRESS
1908 default MODULE_COMPRESS_GZIP
1910 This determines which sort of compression will be used during
1911 'make modules_install'.
1913 GZIP (default) and XZ are supported.
1915 config MODULE_COMPRESS_GZIP
1918 config MODULE_COMPRESS_XZ
1923 config TRIM_UNUSED_KSYMS
1924 bool "Trim unused exported kernel symbols"
1925 depends on MODULES && !UNUSED_SYMBOLS
1927 The kernel and some modules make many symbols available for
1928 other modules to use via EXPORT_SYMBOL() and variants. Depending
1929 on the set of modules being selected in your kernel configuration,
1930 many of those exported symbols might never be used.
1932 This option allows for unused exported symbols to be dropped from
1933 the build. In turn, this provides the compiler more opportunities
1934 (especially when using LTO) for optimizing the code and reducing
1935 binary size. This might have some security advantages as well.
1937 If unsure, or if you need to build out-of-tree modules, say N.
1941 config MODULES_TREE_LOOKUP
1943 depends on PERF_EVENTS || TRACING
1945 config INIT_ALL_POSSIBLE
1948 Back when each arch used to define their own cpu_online_mask and
1949 cpu_possible_mask, some of them chose to initialize cpu_possible_mask
1950 with all 1s, and others with all 0s. When they were centralised,
1951 it was better to provide this option than to break all the archs
1952 and have several arch maintainers pursuing me down dark alleys.
1954 source "block/Kconfig"
1956 config PREEMPT_NOTIFIERS
1966 Build a simple ASN.1 grammar compiler that produces a bytecode output
1967 that can be interpreted by the ASN.1 stream decoder and used to
1968 inform it as to what tags are to be expected in a stream and what
1969 functions to call on what tags.
1971 source "kernel/Kconfig.locks"
1973 config ARCH_HAS_SYNC_CORE_BEFORE_USERMODE
1976 # It may be useful for an architecture to override the definitions of the
1977 # SYSCALL_DEFINE() and __SYSCALL_DEFINEx() macros in <linux/syscalls.h>
1978 # and the COMPAT_ variants in <linux/compat.h>, in particular to use a
1979 # different calling convention for syscalls. They can also override the
1980 # macros for not-implemented syscalls in kernel/sys_ni.c and
1981 # kernel/time/posix-stubs.c. All these overrides need to be available in
1982 # <asm/syscall_wrapper.h>.
1983 config ARCH_HAS_SYSCALL_WRAPPER