7 option env="KERNELVERSION"
13 default "/lib/modules/$UNAME_RELEASE/.config"
14 default "/etc/kernel-config"
15 default "/boot/config-$UNAME_RELEASE"
16 default "$ARCH_DEFCONFIG"
17 default "arch/$ARCH/defconfig"
26 config BUILDTIME_EXTABLE_SORT
29 config THREAD_INFO_IN_TASK
32 Select this to move thread_info off the stack into task_struct. To
33 make this work, an arch will need to remove all thread_info fields
34 except flags and fix any runtime bugs.
36 One subtle change that will be needed is to use try_get_task_stack()
37 and put_task_stack() in save_thread_stack_tsk() and get_wchan().
46 depends on BROKEN || !SMP
49 config INIT_ENV_ARG_LIMIT
54 Maximum of each of the number of arguments and environment
55 variables passed to init from the kernel command line.
59 string "Cross-compiler tool prefix"
61 Same as running 'make CROSS_COMPILE=prefix-' but stored for
62 default make runs in this kernel build directory. You don't
63 need to set this unless you want the configured kernel build
64 directory to select the cross-compiler automatically.
67 bool "Compile also drivers which will not load"
71 Some drivers can be compiled on a different platform than they are
72 intended to be run on. Despite they cannot be loaded there (or even
73 when they load they cannot be used due to missing HW support),
74 developers still, opposing to distributors, might want to build such
75 drivers to compile-test them.
77 If you are a developer and want to build everything available, say Y
78 here. If you are a user/distributor, say N here to exclude useless
79 drivers to be distributed.
82 string "Local version - append to kernel release"
84 Append an extra string to the end of your kernel version.
85 This will show up when you type uname, for example.
86 The string you set here will be appended after the contents of
87 any files with a filename matching localversion* in your
88 object and source tree, in that order. Your total string can
89 be a maximum of 64 characters.
91 config LOCALVERSION_AUTO
92 bool "Automatically append version information to the version string"
94 depends on !COMPILE_TEST
96 This will try to automatically determine if the current tree is a
97 release tree by looking for git tags that belong to the current
100 A string of the format -gxxxxxxxx will be added to the localversion
101 if a git-based tree is found. The string generated by this will be
102 appended after any matching localversion* files, and after the value
103 set in CONFIG_LOCALVERSION.
105 (The actual string used here is the first eight characters produced
106 by running the command:
108 $ git rev-parse --verify HEAD
110 which is done within the script "scripts/setlocalversion".)
112 config HAVE_KERNEL_GZIP
115 config HAVE_KERNEL_BZIP2
118 config HAVE_KERNEL_LZMA
121 config HAVE_KERNEL_XZ
124 config HAVE_KERNEL_LZO
127 config HAVE_KERNEL_LZ4
131 prompt "Kernel compression mode"
133 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4
135 The linux kernel is a kind of self-extracting executable.
136 Several compression algorithms are available, which differ
137 in efficiency, compression and decompression speed.
138 Compression speed is only relevant when building a kernel.
139 Decompression speed is relevant at each boot.
141 If you have any problems with bzip2 or lzma compressed
142 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
143 version of this functionality (bzip2 only), for 2.4, was
144 supplied by Christian Ludwig)
146 High compression options are mostly useful for users, who
147 are low on disk space (embedded systems), but for whom ram
150 If in doubt, select 'gzip'
154 depends on HAVE_KERNEL_GZIP
156 The old and tried gzip compression. It provides a good balance
157 between compression ratio and decompression speed.
161 depends on HAVE_KERNEL_BZIP2
163 Its compression ratio and speed is intermediate.
164 Decompression speed is slowest among the choices. The kernel
165 size is about 10% smaller with bzip2, in comparison to gzip.
166 Bzip2 uses a large amount of memory. For modern kernels you
167 will need at least 8MB RAM or more for booting.
171 depends on HAVE_KERNEL_LZMA
173 This compression algorithm's ratio is best. Decompression speed
174 is between gzip and bzip2. Compression is slowest.
175 The kernel size is about 33% smaller with LZMA in comparison to gzip.
179 depends on HAVE_KERNEL_XZ
181 XZ uses the LZMA2 algorithm and instruction set specific
182 BCJ filters which can improve compression ratio of executable
183 code. The size of the kernel is about 30% smaller with XZ in
184 comparison to gzip. On architectures for which there is a BCJ
185 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ
186 will create a few percent smaller kernel than plain LZMA.
188 The speed is about the same as with LZMA: The decompression
189 speed of XZ is better than that of bzip2 but worse than gzip
190 and LZO. Compression is slow.
194 depends on HAVE_KERNEL_LZO
196 Its compression ratio is the poorest among the choices. The kernel
197 size is about 10% bigger than gzip; however its speed
198 (both compression and decompression) is the fastest.
202 depends on HAVE_KERNEL_LZ4
204 LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding.
205 A preliminary version of LZ4 de/compression tool is available at
206 <https://code.google.com/p/lz4/>.
208 Its compression ratio is worse than LZO. The size of the kernel
209 is about 8% bigger than LZO. But the decompression speed is
214 config DEFAULT_HOSTNAME
215 string "Default hostname"
218 This option determines the default system hostname before userspace
219 calls sethostname(2). The kernel traditionally uses "(none)" here,
220 but you may wish to use a different default here to make a minimal
221 system more usable with less configuration.
224 bool "Support for paging of anonymous memory (swap)"
225 depends on MMU && BLOCK
228 This option allows you to choose whether you want to have support
229 for so called swap devices or swap files in your kernel that are
230 used to provide more virtual memory than the actual RAM present
231 in your computer. If unsure say Y.
236 Inter Process Communication is a suite of library functions and
237 system calls which let processes (running programs) synchronize and
238 exchange information. It is generally considered to be a good thing,
239 and some programs won't run unless you say Y here. In particular, if
240 you want to run the DOS emulator dosemu under Linux (read the
241 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
242 you'll need to say Y here.
244 You can find documentation about IPC with "info ipc" and also in
245 section 6.4 of the Linux Programmer's Guide, available from
246 <http://www.tldp.org/guides.html>.
248 config SYSVIPC_SYSCTL
255 bool "POSIX Message Queues"
258 POSIX variant of message queues is a part of IPC. In POSIX message
259 queues every message has a priority which decides about succession
260 of receiving it by a process. If you want to compile and run
261 programs written e.g. for Solaris with use of its POSIX message
262 queues (functions mq_*) say Y here.
264 POSIX message queues are visible as a filesystem called 'mqueue'
265 and can be mounted somewhere if you want to do filesystem
266 operations on message queues.
270 config POSIX_MQUEUE_SYSCTL
272 depends on POSIX_MQUEUE
277 tristate "kdbus interprocess communication"
280 D-Bus is a system for low-latency, low-overhead, easy to use
281 interprocess communication (IPC).
283 See the man-pages and HTML files in Documentation/kdbus/
284 that are generated by 'make mandocs' and 'make htmldocs'.
286 If you have an ordinary machine, select M here. The module
287 will be called kdbus.
289 config CROSS_MEMORY_ATTACH
290 bool "Enable process_vm_readv/writev syscalls"
294 Enabling this option adds the system calls process_vm_readv and
295 process_vm_writev which allow a process with the correct privileges
296 to directly read from or write to another process' address space.
297 See the man page for more details.
300 bool "open by fhandle syscalls" if EXPERT
304 If you say Y here, a user level program will be able to map
305 file names to handle and then later use the handle for
306 different file system operations. This is useful in implementing
307 userspace file servers, which now track files using handles instead
308 of names. The handle would remain the same even if file names
309 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2)
313 bool "uselib syscall"
314 def_bool ALPHA || M68K || SPARC || X86_32 || IA32_EMULATION
316 This option enables the uselib syscall, a system call used in the
317 dynamic linker from libc5 and earlier. glibc does not use this
318 system call. If you intend to run programs built on libc5 or
319 earlier, you may need to enable this syscall. Current systems
320 running glibc can safely disable this.
323 bool "Auditing support"
326 Enable auditing infrastructure that can be used with another
327 kernel subsystem, such as SELinux (which requires this for
328 logging of avc messages output). System call auditing is included
329 on architectures which support it.
331 config HAVE_ARCH_AUDITSYSCALL
336 depends on AUDIT && HAVE_ARCH_AUDITSYSCALL
340 depends on AUDITSYSCALL
345 depends on AUDITSYSCALL
348 source "kernel/irq/Kconfig"
349 source "kernel/time/Kconfig"
351 menu "CPU/Task time and stats accounting"
353 config VIRT_CPU_ACCOUNTING
357 prompt "Cputime accounting"
358 default TICK_CPU_ACCOUNTING if !PPC64
359 default VIRT_CPU_ACCOUNTING_NATIVE if PPC64
361 # Kind of a stub config for the pure tick based cputime accounting
362 config TICK_CPU_ACCOUNTING
363 bool "Simple tick based cputime accounting"
364 depends on !S390 && !NO_HZ_FULL
366 This is the basic tick based cputime accounting that maintains
367 statistics about user, system and idle time spent on per jiffies
372 config VIRT_CPU_ACCOUNTING_NATIVE
373 bool "Deterministic task and CPU time accounting"
374 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL
375 select VIRT_CPU_ACCOUNTING
377 Select this option to enable more accurate task and CPU time
378 accounting. This is done by reading a CPU counter on each
379 kernel entry and exit and on transitions within the kernel
380 between system, softirq and hardirq state, so there is a
381 small performance impact. In the case of s390 or IBM POWER > 5,
382 this also enables accounting of stolen time on logically-partitioned
385 config VIRT_CPU_ACCOUNTING_GEN
386 bool "Full dynticks CPU time accounting"
387 depends on HAVE_CONTEXT_TRACKING
388 depends on HAVE_VIRT_CPU_ACCOUNTING_GEN
389 select VIRT_CPU_ACCOUNTING
390 select CONTEXT_TRACKING
392 Select this option to enable task and CPU time accounting on full
393 dynticks systems. This accounting is implemented by watching every
394 kernel-user boundaries using the context tracking subsystem.
395 The accounting is thus performed at the expense of some significant
398 For now this is only useful if you are working on the full
399 dynticks subsystem development.
405 config IRQ_TIME_ACCOUNTING
406 bool "Fine granularity task level IRQ time accounting"
407 depends on HAVE_IRQ_TIME_ACCOUNTING && !VIRT_CPU_ACCOUNTING_NATIVE
409 Select this option to enable fine granularity task irq time
410 accounting. This is done by reading a timestamp on each
411 transitions between softirq and hardirq state, so there can be a
412 small performance impact.
414 If in doubt, say N here.
416 config BSD_PROCESS_ACCT
417 bool "BSD Process Accounting"
420 If you say Y here, a user level program will be able to instruct the
421 kernel (via a special system call) to write process accounting
422 information to a file: whenever a process exits, information about
423 that process will be appended to the file by the kernel. The
424 information includes things such as creation time, owning user,
425 command name, memory usage, controlling terminal etc. (the complete
426 list is in the struct acct in <file:include/linux/acct.h>). It is
427 up to the user level program to do useful things with this
428 information. This is generally a good idea, so say Y.
430 config BSD_PROCESS_ACCT_V3
431 bool "BSD Process Accounting version 3 file format"
432 depends on BSD_PROCESS_ACCT
435 If you say Y here, the process accounting information is written
436 in a new file format that also logs the process IDs of each
437 process and it's parent. Note that this file format is incompatible
438 with previous v0/v1/v2 file formats, so you will need updated tools
439 for processing it. A preliminary version of these tools is available
440 at <http://www.gnu.org/software/acct/>.
443 bool "Export task/process statistics through netlink"
448 Export selected statistics for tasks/processes through the
449 generic netlink interface. Unlike BSD process accounting, the
450 statistics are available during the lifetime of tasks/processes as
451 responses to commands. Like BSD accounting, they are sent to user
456 config TASK_DELAY_ACCT
457 bool "Enable per-task delay accounting"
461 Collect information on time spent by a task waiting for system
462 resources like cpu, synchronous block I/O completion and swapping
463 in pages. Such statistics can help in setting a task's priorities
464 relative to other tasks for cpu, io, rss limits etc.
469 bool "Enable extended accounting over taskstats"
472 Collect extended task accounting data and send the data
473 to userland for processing over the taskstats interface.
477 config TASK_IO_ACCOUNTING
478 bool "Enable per-task storage I/O accounting"
479 depends on TASK_XACCT
481 Collect information on the number of bytes of storage I/O which this
486 endmenu # "CPU/Task time and stats accounting"
488 source "kernel/rcu/Kconfig"
495 tristate "Kernel .config support"
498 This option enables the complete Linux kernel ".config" file
499 contents to be saved in the kernel. It provides documentation
500 of which kernel options are used in a running kernel or in an
501 on-disk kernel. This information can be extracted from the kernel
502 image file with the script scripts/extract-ikconfig and used as
503 input to rebuild the current kernel or to build another kernel.
504 It can also be extracted from a running kernel by reading
505 /proc/config.gz if enabled (below).
508 bool "Enable access to .config through /proc/config.gz"
509 depends on IKCONFIG && PROC_FS
511 This option enables access to the kernel configuration file
512 through /proc/config.gz.
515 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
520 Select the minimal kernel log buffer size as a power of 2.
521 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config
522 parameter, see below. Any higher size also might be forced
523 by "log_buf_len" boot parameter.
533 config LOG_CPU_MAX_BUF_SHIFT
534 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)"
537 default 12 if !BASE_SMALL
538 default 0 if BASE_SMALL
541 This option allows to increase the default ring buffer size
542 according to the number of CPUs. The value defines the contribution
543 of each CPU as a power of 2. The used space is typically only few
544 lines however it might be much more when problems are reported,
547 The increased size means that a new buffer has to be allocated and
548 the original static one is unused. It makes sense only on systems
549 with more CPUs. Therefore this value is used only when the sum of
550 contributions is greater than the half of the default kernel ring
551 buffer as defined by LOG_BUF_SHIFT. The default values are set
552 so that more than 64 CPUs are needed to trigger the allocation.
554 Also this option is ignored when "log_buf_len" kernel parameter is
555 used as it forces an exact (power of two) size of the ring buffer.
557 The number of possible CPUs is used for this computation ignoring
558 hotplugging making the computation optimal for the worst case
559 scenario while allowing a simple algorithm to be used from bootup.
561 Examples shift values and their meaning:
562 17 => 128 KB for each CPU
563 16 => 64 KB for each CPU
564 15 => 32 KB for each CPU
565 14 => 16 KB for each CPU
566 13 => 8 KB for each CPU
567 12 => 4 KB for each CPU
569 config PRINTK_SAFE_LOG_BUF_SHIFT
570 int "Temporary per-CPU printk log buffer size (12 => 4KB, 13 => 8KB)"
575 Select the size of an alternate printk per-CPU buffer where messages
576 printed from usafe contexts are temporary stored. One example would
577 be NMI messages, another one - printk recursion. The messages are
578 copied to the main log buffer in a safe context to avoid a deadlock.
579 The value defines the size as a power of 2.
581 Those messages are rare and limited. The largest one is when
582 a backtrace is printed. It usually fits into 4KB. Select
583 8KB if you want to be on the safe side.
586 17 => 128 KB for each CPU
587 16 => 64 KB for each CPU
588 15 => 32 KB for each CPU
589 14 => 16 KB for each CPU
590 13 => 8 KB for each CPU
591 12 => 4 KB for each CPU
594 # Architectures with an unreliable sched_clock() should select this:
596 config HAVE_UNSTABLE_SCHED_CLOCK
599 config GENERIC_SCHED_CLOCK
603 # For architectures that want to enable the support for NUMA-affine scheduler
606 config ARCH_SUPPORTS_NUMA_BALANCING
610 # For architectures that prefer to flush all TLBs after a number of pages
611 # are unmapped instead of sending one IPI per page to flush. The architecture
612 # must provide guarantees on what happens if a clean TLB cache entry is
613 # written after the unmap. Details are in mm/rmap.c near the check for
614 # should_defer_flush. The architecture should also consider if the full flush
615 # and the refill costs are offset by the savings of sending fewer IPIs.
616 config ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
620 # For architectures that know their GCC __int128 support is sound
622 config ARCH_SUPPORTS_INT128
625 # For architectures that (ab)use NUMA to represent different memory regions
626 # all cpu-local but of different latencies, such as SuperH.
628 config ARCH_WANT_NUMA_VARIABLE_LOCALITY
631 config NUMA_BALANCING
632 bool "Memory placement aware NUMA scheduler"
633 depends on ARCH_SUPPORTS_NUMA_BALANCING
634 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
635 depends on SMP && NUMA && MIGRATION
637 This option adds support for automatic NUMA aware memory/task placement.
638 The mechanism is quite primitive and is based on migrating memory when
639 it has references to the node the task is running on.
641 This system will be inactive on UMA systems.
643 config NUMA_BALANCING_DEFAULT_ENABLED
644 bool "Automatically enable NUMA aware memory/task placement"
646 depends on NUMA_BALANCING
648 If set, automatic NUMA balancing will be enabled if running on a NUMA
652 bool "Control Group support"
655 This option adds support for grouping sets of processes together, for
656 use with process control subsystems such as Cpusets, CFS, memory
657 controls or device isolation.
659 - Documentation/scheduler/sched-design-CFS.txt (CFS)
660 - Documentation/cgroup-v1/ (features for grouping, isolation
661 and resource control)
671 bool "Memory controller"
675 Provides control over the memory footprint of tasks in a cgroup.
678 bool "Swap controller"
679 depends on MEMCG && SWAP
681 Provides control over the swap space consumed by tasks in a cgroup.
683 config MEMCG_SWAP_ENABLED
684 bool "Swap controller enabled by default"
685 depends on MEMCG_SWAP
688 Memory Resource Controller Swap Extension comes with its price in
689 a bigger memory consumption. General purpose distribution kernels
690 which want to enable the feature but keep it disabled by default
691 and let the user enable it by swapaccount=1 boot command line
692 parameter should have this option unselected.
693 For those who want to have the feature enabled by default should
694 select this option (if, for some reason, they need to disable it
695 then swapaccount=0 does the trick).
702 Generic block IO controller cgroup interface. This is the common
703 cgroup interface which should be used by various IO controlling
706 Currently, CFQ IO scheduler uses it to recognize task groups and
707 control disk bandwidth allocation (proportional time slice allocation)
708 to such task groups. It is also used by bio throttling logic in
709 block layer to implement upper limit in IO rates on a device.
711 This option only enables generic Block IO controller infrastructure.
712 One needs to also enable actual IO controlling logic/policy. For
713 enabling proportional weight division of disk bandwidth in CFQ, set
714 CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set
715 CONFIG_BLK_DEV_THROTTLING=y.
717 See Documentation/cgroup-v1/blkio-controller.txt for more information.
719 config DEBUG_BLK_CGROUP
720 bool "IO controller debugging"
721 depends on BLK_CGROUP
724 Enable some debugging help. Currently it exports additional stat
725 files in a cgroup which can be useful for debugging.
727 config CGROUP_WRITEBACK
729 depends on MEMCG && BLK_CGROUP
732 menuconfig CGROUP_SCHED
733 bool "CPU controller"
736 This feature lets CPU scheduler recognize task groups and control CPU
737 bandwidth allocation to such task groups. It uses cgroups to group
741 config FAIR_GROUP_SCHED
742 bool "Group scheduling for SCHED_OTHER"
743 depends on CGROUP_SCHED
747 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED"
748 depends on FAIR_GROUP_SCHED
751 This option allows users to define CPU bandwidth rates (limits) for
752 tasks running within the fair group scheduler. Groups with no limit
753 set are considered to be unconstrained and will run with no
755 See tip/Documentation/scheduler/sched-bwc.txt for more information.
757 config RT_GROUP_SCHED
758 bool "Group scheduling for SCHED_RR/FIFO"
759 depends on CGROUP_SCHED
762 This feature lets you explicitly allocate real CPU bandwidth
763 to task groups. If enabled, it will also make it impossible to
764 schedule realtime tasks for non-root users until you allocate
765 realtime bandwidth for them.
766 See Documentation/scheduler/sched-rt-group.txt for more information.
771 bool "PIDs controller"
773 Provides enforcement of process number limits in the scope of a
774 cgroup. Any attempt to fork more processes than is allowed in the
775 cgroup will fail. PIDs are fundamentally a global resource because it
776 is fairly trivial to reach PID exhaustion before you reach even a
777 conservative kmemcg limit. As a result, it is possible to grind a
778 system to halt without being limited by other cgroup policies. The
779 PIDs controller is designed to stop this from happening.
781 It should be noted that organisational operations (such as attaching
782 to a cgroup hierarchy will *not* be blocked by the PIDs controller),
783 since the PIDs limit only affects a process's ability to fork, not to
787 bool "RDMA controller"
789 Provides enforcement of RDMA resources defined by IB stack.
790 It is fairly easy for consumers to exhaust RDMA resources, which
791 can result into resource unavailability to other consumers.
792 RDMA controller is designed to stop this from happening.
793 Attaching processes with active RDMA resources to the cgroup
794 hierarchy is allowed even if can cross the hierarchy's limit.
796 config CGROUP_FREEZER
797 bool "Freezer controller"
799 Provides a way to freeze and unfreeze all tasks in a
802 This option affects the ORIGINAL cgroup interface. The cgroup2 memory
803 controller includes important in-kernel memory consumers per default.
805 If you're using cgroup2, say N.
807 config CGROUP_HUGETLB
808 bool "HugeTLB controller"
809 depends on HUGETLB_PAGE
813 Provides a cgroup controller for HugeTLB pages.
814 When you enable this, you can put a per cgroup limit on HugeTLB usage.
815 The limit is enforced during page fault. Since HugeTLB doesn't
816 support page reclaim, enforcing the limit at page fault time implies
817 that, the application will get SIGBUS signal if it tries to access
818 HugeTLB pages beyond its limit. This requires the application to know
819 beforehand how much HugeTLB pages it would require for its use. The
820 control group is tracked in the third page lru pointer. This means
821 that we cannot use the controller with huge page less than 3 pages.
824 bool "Cpuset controller"
827 This option will let you create and manage CPUSETs which
828 allow dynamically partitioning a system into sets of CPUs and
829 Memory Nodes and assigning tasks to run only within those sets.
830 This is primarily useful on large SMP or NUMA systems.
834 config PROC_PID_CPUSET
835 bool "Include legacy /proc/<pid>/cpuset file"
840 bool "Device controller"
842 Provides a cgroup controller implementing whitelists for
843 devices which a process in the cgroup can mknod or open.
845 config CGROUP_CPUACCT
846 bool "Simple CPU accounting controller"
848 Provides a simple controller for monitoring the
849 total CPU consumed by the tasks in a cgroup.
852 bool "Perf controller"
853 depends on PERF_EVENTS
855 This option extends the perf per-cpu mode to restrict monitoring
856 to threads which belong to the cgroup specified and run on the
862 bool "Support for eBPF programs attached to cgroups"
863 depends on BPF_SYSCALL
864 select SOCK_CGROUP_DATA
866 Allow attaching eBPF programs to a cgroup using the bpf(2)
867 syscall command BPF_PROG_ATTACH.
869 In which context these programs are accessed depends on the type
870 of attachment. For instance, programs that are attached using
871 BPF_CGROUP_INET_INGRESS will be executed on the ingress path of
875 bool "Debug controller"
877 depends on DEBUG_KERNEL
879 This option enables a simple controller that exports
880 debugging information about the cgroups framework. This
881 controller is for control cgroup debugging only. Its
882 interfaces are not stable.
886 config SOCK_CGROUP_DATA
892 config CHECKPOINT_RESTORE
893 bool "Checkpoint/restore support" if EXPERT
897 Enables additional kernel features in a sake of checkpoint/restore.
898 In particular it adds auxiliary prctl codes to setup process text,
899 data and heap segment sizes, and a few additional /proc filesystem
902 If unsure, say N here.
904 menuconfig NAMESPACES
905 bool "Namespaces support" if EXPERT
909 Provides the way to make tasks work with different objects using
910 the same id. For example same IPC id may refer to different objects
911 or same user id or pid may refer to different tasks when used in
912 different namespaces.
920 In this namespace tasks see different info provided with the
925 depends on (SYSVIPC || POSIX_MQUEUE)
928 In this namespace tasks work with IPC ids which correspond to
929 different IPC objects in different namespaces.
932 bool "User namespace"
935 This allows containers, i.e. vservers, to use user namespaces
936 to provide different user info for different servers.
938 When user namespaces are enabled in the kernel it is
939 recommended that the MEMCG option also be enabled and that
940 user-space use the memory control groups to limit the amount
941 of memory a memory unprivileged users can use.
946 bool "PID Namespaces"
949 Support process id namespaces. This allows having multiple
950 processes with the same pid as long as they are in different
951 pid namespaces. This is a building block of containers.
954 bool "Network namespace"
958 Allow user space to create what appear to be multiple instances
959 of the network stack.
963 config SCHED_AUTOGROUP
964 bool "Automatic process group scheduling"
967 select FAIR_GROUP_SCHED
969 This option optimizes the scheduler for common desktop workloads by
970 automatically creating and populating task groups. This separation
971 of workloads isolates aggressive CPU burners (like build jobs) from
972 desktop applications. Task group autogeneration is currently based
975 config SYSFS_DEPRECATED
976 bool "Enable deprecated sysfs features to support old userspace tools"
980 This option adds code that switches the layout of the "block" class
981 devices, to not show up in /sys/class/block/, but only in
984 This switch is only active when the sysfs.deprecated=1 boot option is
985 passed or the SYSFS_DEPRECATED_V2 option is set.
987 This option allows new kernels to run on old distributions and tools,
988 which might get confused by /sys/class/block/. Since 2007/2008 all
989 major distributions and tools handle this just fine.
991 Recent distributions and userspace tools after 2009/2010 depend on
992 the existence of /sys/class/block/, and will not work with this
995 Only if you are using a new kernel on an old distribution, you might
998 config SYSFS_DEPRECATED_V2
999 bool "Enable deprecated sysfs features by default"
1002 depends on SYSFS_DEPRECATED
1004 Enable deprecated sysfs by default.
1006 See the CONFIG_SYSFS_DEPRECATED option for more details about this
1009 Only if you are using a new kernel on an old distribution, you might
1010 need to say Y here. Even then, odds are you would not need it
1011 enabled, you can always pass the boot option if absolutely necessary.
1014 bool "Kernel->user space relay support (formerly relayfs)"
1017 This option enables support for relay interface support in
1018 certain file systems (such as debugfs).
1019 It is designed to provide an efficient mechanism for tools and
1020 facilities to relay large amounts of data from kernel space to
1025 config BLK_DEV_INITRD
1026 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
1027 depends on BROKEN || !FRV
1029 The initial RAM filesystem is a ramfs which is loaded by the
1030 boot loader (loadlin or lilo) and that is mounted as root
1031 before the normal boot procedure. It is typically used to
1032 load modules needed to mount the "real" root file system,
1033 etc. See <file:Documentation/admin-guide/initrd.rst> for details.
1035 If RAM disk support (BLK_DEV_RAM) is also included, this
1036 also enables initial RAM disk (initrd) support and adds
1037 15 Kbytes (more on some other architectures) to the kernel size.
1043 source "usr/Kconfig"
1048 prompt "Compiler optimization level"
1049 default CC_OPTIMIZE_FOR_PERFORMANCE
1051 config CC_OPTIMIZE_FOR_PERFORMANCE
1052 bool "Optimize for performance"
1054 This is the default optimization level for the kernel, building
1055 with the "-O2" compiler flag for best performance and most
1056 helpful compile-time warnings.
1058 config CC_OPTIMIZE_FOR_SIZE
1059 bool "Optimize for size"
1061 Enabling this option will pass "-Os" instead of "-O2" to
1062 your compiler resulting in a smaller kernel.
1077 config SYSCTL_EXCEPTION_TRACE
1080 Enable support for /proc/sys/debug/exception-trace.
1082 config SYSCTL_ARCH_UNALIGN_NO_WARN
1085 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap
1086 Allows arch to define/use @no_unaligned_warning to possibly warn
1087 about unaligned access emulation going on under the hood.
1089 config SYSCTL_ARCH_UNALIGN_ALLOW
1092 Enable support for /proc/sys/kernel/unaligned-trap
1093 Allows arches to define/use @unaligned_enabled to runtime toggle
1094 the unaligned access emulation.
1095 see arch/parisc/kernel/unaligned.c for reference
1097 config HAVE_PCSPKR_PLATFORM
1100 # interpreter that classic socket filters depend on
1105 bool "Configure standard kernel features (expert users)"
1106 # Unhide debug options, to make the on-by-default options visible
1109 This option allows certain base kernel options and settings
1110 to be disabled or tweaked. This is for specialized
1111 environments which can tolerate a "non-standard" kernel.
1112 Only use this if you really know what you are doing.
1115 bool "Enable 16-bit UID system calls" if EXPERT
1116 depends on HAVE_UID16 && MULTIUSER
1119 This enables the legacy 16-bit UID syscall wrappers.
1122 bool "Multiple users, groups and capabilities support" if EXPERT
1125 This option enables support for non-root users, groups and
1128 If you say N here, all processes will run with UID 0, GID 0, and all
1129 possible capabilities. Saying N here also compiles out support for
1130 system calls related to UIDs, GIDs, and capabilities, such as setuid,
1133 If unsure, say Y here.
1135 config SGETMASK_SYSCALL
1136 bool "sgetmask/ssetmask syscalls support" if EXPERT
1137 def_bool PARISC || MN10300 || BLACKFIN || M68K || PPC || MIPS || X86 || SPARC || CRIS || MICROBLAZE || SUPERH
1139 sys_sgetmask and sys_ssetmask are obsolete system calls
1140 no longer supported in libc but still enabled by default in some
1143 If unsure, leave the default option here.
1145 config SYSFS_SYSCALL
1146 bool "Sysfs syscall support" if EXPERT
1149 sys_sysfs is an obsolete system call no longer supported in libc.
1150 Note that disabling this option is more secure but might break
1151 compatibility with some systems.
1153 If unsure say Y here.
1155 config SYSCTL_SYSCALL
1156 bool "Sysctl syscall support" if EXPERT
1157 depends on PROC_SYSCTL
1161 sys_sysctl uses binary paths that have been found challenging
1162 to properly maintain and use. The interface in /proc/sys
1163 using paths with ascii names is now the primary path to this
1166 Almost nothing using the binary sysctl interface so if you are
1167 trying to save some space it is probably safe to disable this,
1168 making your kernel marginally smaller.
1170 If unsure say N here.
1173 bool "Posix Clocks & timers" if EXPERT
1176 This includes native support for POSIX timers to the kernel.
1177 Some embedded systems have no use for them and therefore they
1178 can be configured out to reduce the size of the kernel image.
1180 When this option is disabled, the following syscalls won't be
1181 available: timer_create, timer_gettime: timer_getoverrun,
1182 timer_settime, timer_delete, clock_adjtime, getitimer,
1183 setitimer, alarm. Furthermore, the clock_settime, clock_gettime,
1184 clock_getres and clock_nanosleep syscalls will be limited to
1185 CLOCK_REALTIME, CLOCK_MONOTONIC and CLOCK_BOOTTIME only.
1190 bool "Load all symbols for debugging/ksymoops" if EXPERT
1193 Say Y here to let the kernel print out symbolic crash information and
1194 symbolic stack backtraces. This increases the size of the kernel
1195 somewhat, as all symbols have to be loaded into the kernel image.
1198 bool "Include all symbols in kallsyms"
1199 depends on DEBUG_KERNEL && KALLSYMS
1201 Normally kallsyms only contains the symbols of functions for nicer
1202 OOPS messages and backtraces (i.e., symbols from the text and inittext
1203 sections). This is sufficient for most cases. And only in very rare
1204 cases (e.g., when a debugger is used) all symbols are required (e.g.,
1205 names of variables from the data sections, etc).
1207 This option makes sure that all symbols are loaded into the kernel
1208 image (i.e., symbols from all sections) in cost of increased kernel
1209 size (depending on the kernel configuration, it may be 300KiB or
1210 something like this).
1212 Say N unless you really need all symbols.
1214 config KALLSYMS_ABSOLUTE_PERCPU
1217 default X86_64 && SMP
1219 config KALLSYMS_BASE_RELATIVE
1222 default !IA64 && !(TILE && 64BIT)
1224 Instead of emitting them as absolute values in the native word size,
1225 emit the symbol references in the kallsyms table as 32-bit entries,
1226 each containing a relative value in the range [base, base + U32_MAX]
1227 or, when KALLSYMS_ABSOLUTE_PERCPU is in effect, each containing either
1228 an absolute value in the range [0, S32_MAX] or a relative value in the
1229 range [base, base + S32_MAX], where base is the lowest relative symbol
1230 address encountered in the image.
1232 On 64-bit builds, this reduces the size of the address table by 50%,
1233 but more importantly, it results in entries whose values are build
1234 time constants, and no relocation pass is required at runtime to fix
1235 up the entries based on the runtime load address of the kernel.
1239 bool "Enable support for printk" if EXPERT
1242 This option enables normal printk support. Removing it
1243 eliminates most of the message strings from the kernel image
1244 and makes the kernel more or less silent. As this makes it
1245 very difficult to diagnose system problems, saying N here is
1246 strongly discouraged.
1254 bool "BUG() support" if EXPERT
1257 Disabling this option eliminates support for BUG and WARN, reducing
1258 the size of your kernel image and potentially quietly ignoring
1259 numerous fatal conditions. You should only consider disabling this
1260 option for embedded systems with no facilities for reporting errors.
1266 bool "Enable ELF core dumps" if EXPERT
1268 Enable support for generating core dumps. Disabling saves about 4k.
1271 config PCSPKR_PLATFORM
1272 bool "Enable PC-Speaker support" if EXPERT
1273 depends on HAVE_PCSPKR_PLATFORM
1277 This option allows to disable the internal PC-Speaker
1278 support, saving some memory.
1282 bool "Enable full-sized data structures for core" if EXPERT
1284 Disabling this option reduces the size of miscellaneous core
1285 kernel data structures. This saves memory on small machines,
1286 but may reduce performance.
1289 bool "Enable futex support" if EXPERT
1293 Disabling this option will cause the kernel to be built without
1294 support for "fast userspace mutexes". The resulting kernel may not
1295 run glibc-based applications correctly.
1299 depends on FUTEX && RT_MUTEXES
1302 config HAVE_FUTEX_CMPXCHG
1306 Architectures should select this if futex_atomic_cmpxchg_inatomic()
1307 is implemented and always working. This removes a couple of runtime
1311 bool "Enable eventpoll support" if EXPERT
1315 Disabling this option will cause the kernel to be built without
1316 support for epoll family of system calls.
1319 bool "Enable signalfd() system call" if EXPERT
1323 Enable the signalfd() system call that allows to receive signals
1324 on a file descriptor.
1329 bool "Enable timerfd() system call" if EXPERT
1333 Enable the timerfd() system call that allows to receive timer
1334 events on a file descriptor.
1339 bool "Enable eventfd() system call" if EXPERT
1343 Enable the eventfd() system call that allows to receive both
1344 kernel notification (ie. KAIO) or userspace notifications.
1348 # syscall, maps, verifier
1350 bool "Enable bpf() system call"
1355 Enable the bpf() system call that allows to manipulate eBPF
1356 programs and maps via file descriptors.
1358 config BPF_JIT_ALWAYS_ON
1359 bool "Permanently enable BPF JIT and remove BPF interpreter"
1360 depends on BPF_SYSCALL && HAVE_EBPF_JIT && BPF_JIT
1362 Enables BPF JIT and removes BPF interpreter to avoid
1363 speculative execution of BPF instructions by the interpreter
1366 bool "Use full shmem filesystem" if EXPERT
1370 The shmem is an internal filesystem used to manage shared memory.
1371 It is backed by swap and manages resource limits. It is also exported
1372 to userspace as tmpfs if TMPFS is enabled. Disabling this
1373 option replaces shmem and tmpfs with the much simpler ramfs code,
1374 which may be appropriate on small systems without swap.
1377 bool "Enable AIO support" if EXPERT
1380 This option enables POSIX asynchronous I/O which may by used
1381 by some high performance threaded applications. Disabling
1382 this option saves about 7k.
1384 config ADVISE_SYSCALLS
1385 bool "Enable madvise/fadvise syscalls" if EXPERT
1388 This option enables the madvise and fadvise syscalls, used by
1389 applications to advise the kernel about their future memory or file
1390 usage, improving performance. If building an embedded system where no
1391 applications use these syscalls, you can disable this option to save
1395 bool "Enable userfaultfd() system call"
1399 Enable the userfaultfd() system call that allows to intercept and
1400 handle page faults in userland.
1404 bool "Enable PCI quirk workarounds" if EXPERT
1407 This enables workarounds for various PCI chipset
1408 bugs/quirks. Disable this only if your target machine is
1409 unaffected by PCI quirks.
1412 bool "Enable membarrier() system call" if EXPERT
1415 Enable the membarrier() system call that allows issuing memory
1416 barriers across all running threads, which can be used to distribute
1417 the cost of user-space memory barriers asymmetrically by transforming
1418 pairs of memory barriers into pairs consisting of membarrier() and a
1424 bool "Embedded system"
1425 option allnoconfig_y
1428 This option should be enabled if compiling the kernel for
1429 an embedded system so certain expert options are available
1432 config HAVE_PERF_EVENTS
1435 See tools/perf/design.txt for details.
1437 config PERF_USE_VMALLOC
1440 See tools/perf/design.txt for details
1443 bool "PC/104 support"
1445 Expose PC/104 form factor device drivers and options available for
1446 selection and configuration. Enable this option if your target
1447 machine has a PC/104 bus.
1449 menu "Kernel Performance Events And Counters"
1452 bool "Kernel performance events and counters"
1453 default y if PROFILING
1454 depends on HAVE_PERF_EVENTS
1459 Enable kernel support for various performance events provided
1460 by software and hardware.
1462 Software events are supported either built-in or via the
1463 use of generic tracepoints.
1465 Most modern CPUs support performance events via performance
1466 counter registers. These registers count the number of certain
1467 types of hw events: such as instructions executed, cachemisses
1468 suffered, or branches mis-predicted - without slowing down the
1469 kernel or applications. These registers can also trigger interrupts
1470 when a threshold number of events have passed - and can thus be
1471 used to profile the code that runs on that CPU.
1473 The Linux Performance Event subsystem provides an abstraction of
1474 these software and hardware event capabilities, available via a
1475 system call and used by the "perf" utility in tools/perf/. It
1476 provides per task and per CPU counters, and it provides event
1477 capabilities on top of those.
1481 config DEBUG_PERF_USE_VMALLOC
1483 bool "Debug: use vmalloc to back perf mmap() buffers"
1484 depends on PERF_EVENTS && DEBUG_KERNEL && !PPC
1485 select PERF_USE_VMALLOC
1487 Use vmalloc memory to back perf mmap() buffers.
1489 Mostly useful for debugging the vmalloc code on platforms
1490 that don't require it.
1496 config VM_EVENT_COUNTERS
1498 bool "Enable VM event counters for /proc/vmstat" if EXPERT
1500 VM event counters are needed for event counts to be shown.
1501 This option allows the disabling of the VM event counters
1502 on EXPERT systems. /proc/vmstat will only show page counts
1503 if VM event counters are disabled.
1507 bool "Enable SLUB debugging support" if EXPERT
1508 depends on SLUB && SYSFS
1510 SLUB has extensive debug support features. Disabling these can
1511 result in significant savings in code size. This also disables
1512 SLUB sysfs support. /sys/slab will not exist and there will be
1513 no support for cache validation etc.
1515 config SLUB_MEMCG_SYSFS_ON
1517 bool "Enable memcg SLUB sysfs support by default" if EXPERT
1518 depends on SLUB && SYSFS && MEMCG
1520 SLUB creates a directory under /sys/kernel/slab for each
1521 allocation cache to host info and debug files. If memory
1522 cgroup is enabled, each cache can have per memory cgroup
1523 caches. SLUB can create the same sysfs directories for these
1524 caches under /sys/kernel/slab/CACHE/cgroup but it can lead
1525 to a very high number of debug files being created. This is
1526 controlled by slub_memcg_sysfs boot parameter and this
1527 config option determines the parameter's default value.
1530 bool "Disable heap randomization"
1533 Randomizing heap placement makes heap exploits harder, but it
1534 also breaks ancient binaries (including anything libc5 based).
1535 This option changes the bootup default to heap randomization
1536 disabled, and can be overridden at runtime by setting
1537 /proc/sys/kernel/randomize_va_space to 2.
1539 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1542 prompt "Choose SLAB allocator"
1545 This option allows to select a slab allocator.
1549 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1551 The regular slab allocator that is established and known to work
1552 well in all environments. It organizes cache hot objects in
1553 per cpu and per node queues.
1556 bool "SLUB (Unqueued Allocator)"
1557 select HAVE_HARDENED_USERCOPY_ALLOCATOR
1559 SLUB is a slab allocator that minimizes cache line usage
1560 instead of managing queues of cached objects (SLAB approach).
1561 Per cpu caching is realized using slabs of objects instead
1562 of queues of objects. SLUB can use memory efficiently
1563 and has enhanced diagnostics. SLUB is the default choice for
1568 bool "SLOB (Simple Allocator)"
1570 SLOB replaces the stock allocator with a drastically simpler
1571 allocator. SLOB is generally more space efficient but
1572 does not perform as well on large systems.
1576 config SLAB_MERGE_DEFAULT
1577 bool "Allow slab caches to be merged"
1580 For reduced kernel memory fragmentation, slab caches can be
1581 merged when they share the same size and other characteristics.
1582 This carries a risk of kernel heap overflows being able to
1583 overwrite objects from merged caches (and more easily control
1584 cache layout), which makes such heap attacks easier to exploit
1585 by attackers. By keeping caches unmerged, these kinds of exploits
1586 can usually only damage objects in the same cache. To disable
1587 merging at runtime, "slab_nomerge" can be passed on the kernel
1590 config SLAB_FREELIST_RANDOM
1592 depends on SLAB || SLUB
1593 bool "SLAB freelist randomization"
1595 Randomizes the freelist order used on creating new pages. This
1596 security feature reduces the predictability of the kernel slab
1597 allocator against heap overflows.
1599 config SLAB_FREELIST_HARDENED
1600 bool "Harden slab freelist metadata"
1603 Many kernel heap attacks try to target slab cache metadata and
1604 other infrastructure. This options makes minor performance
1605 sacrifies to harden the kernel slab allocator against common
1606 freelist exploit methods.
1608 config SLUB_CPU_PARTIAL
1610 depends on SLUB && SMP
1611 bool "SLUB per cpu partial cache"
1613 Per cpu partial caches accellerate objects allocation and freeing
1614 that is local to a processor at the price of more indeterminism
1615 in the latency of the free. On overflow these caches will be cleared
1616 which requires the taking of locks that may cause latency spikes.
1617 Typically one would choose no for a realtime system.
1619 config MMAP_ALLOW_UNINITIALIZED
1620 bool "Allow mmapped anonymous memory to be uninitialized"
1621 depends on EXPERT && !MMU
1624 Normally, and according to the Linux spec, anonymous memory obtained
1625 from mmap() has it's contents cleared before it is passed to
1626 userspace. Enabling this config option allows you to request that
1627 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
1628 providing a huge performance boost. If this option is not enabled,
1629 then the flag will be ignored.
1631 This is taken advantage of by uClibc's malloc(), and also by
1632 ELF-FDPIC binfmt's brk and stack allocator.
1634 Because of the obvious security issues, this option should only be
1635 enabled on embedded devices where you control what is run in
1636 userspace. Since that isn't generally a problem on no-MMU systems,
1637 it is normally safe to say Y here.
1639 See Documentation/nommu-mmap.txt for more information.
1641 config SYSTEM_DATA_VERIFICATION
1643 select SYSTEM_TRUSTED_KEYRING
1647 select ASYMMETRIC_KEY_TYPE
1648 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE
1651 select X509_CERTIFICATE_PARSER
1652 select PKCS7_MESSAGE_PARSER
1654 Provide PKCS#7 message verification using the contents of the system
1655 trusted keyring to provide public keys. This then can be used for
1656 module verification, kexec image verification and firmware blob
1660 bool "Profiling support"
1662 Say Y here to enable the extended profiling support mechanisms used
1663 by profilers such as OProfile.
1666 # Place an empty function call at each tracepoint site. Can be
1667 # dynamically changed for a probe function.
1672 source "arch/Kconfig"
1674 endmenu # General setup
1676 config HAVE_GENERIC_DMA_COHERENT
1683 depends on SLAB || SLUB_DEBUG
1691 default 0 if BASE_FULL
1692 default 1 if !BASE_FULL
1695 bool "Enable loadable module support"
1698 Kernel modules are small pieces of compiled code which can
1699 be inserted in the running kernel, rather than being
1700 permanently built into the kernel. You use the "modprobe"
1701 tool to add (and sometimes remove) them. If you say Y here,
1702 many parts of the kernel can be built as modules (by
1703 answering M instead of Y where indicated): this is most
1704 useful for infrequently used options which are not required
1705 for booting. For more information, see the man pages for
1706 modprobe, lsmod, modinfo, insmod and rmmod.
1708 If you say Y here, you will need to run "make
1709 modules_install" to put the modules under /lib/modules/
1710 where modprobe can find them (you may need to be root to do
1717 config MODULE_FORCE_LOAD
1718 bool "Forced module loading"
1721 Allow loading of modules without version information (ie. modprobe
1722 --force). Forced module loading sets the 'F' (forced) taint flag and
1723 is usually a really bad idea.
1725 config MODULE_UNLOAD
1726 bool "Module unloading"
1728 Without this option you will not be able to unload any
1729 modules (note that some modules may not be unloadable
1730 anyway), which makes your kernel smaller, faster
1731 and simpler. If unsure, say Y.
1733 config MODULE_FORCE_UNLOAD
1734 bool "Forced module unloading"
1735 depends on MODULE_UNLOAD
1737 This option allows you to force a module to unload, even if the
1738 kernel believes it is unsafe: the kernel will remove the module
1739 without waiting for anyone to stop using it (using the -f option to
1740 rmmod). This is mainly for kernel developers and desperate users.
1744 bool "Module versioning support"
1746 Usually, you have to use modules compiled with your kernel.
1747 Saying Y here makes it sometimes possible to use modules
1748 compiled for different kernels, by adding enough information
1749 to the modules to (hopefully) spot any changes which would
1750 make them incompatible with the kernel you are running. If
1753 config MODULE_REL_CRCS
1755 depends on MODVERSIONS
1757 config MODULE_SRCVERSION_ALL
1758 bool "Source checksum for all modules"
1760 Modules which contain a MODULE_VERSION get an extra "srcversion"
1761 field inserted into their modinfo section, which contains a
1762 sum of the source files which made it. This helps maintainers
1763 see exactly which source was used to build a module (since
1764 others sometimes change the module source without updating
1765 the version). With this option, such a "srcversion" field
1766 will be created for all modules. If unsure, say N.
1769 bool "Module signature verification"
1771 select SYSTEM_DATA_VERIFICATION
1773 Check modules for valid signatures upon load: the signature
1774 is simply appended to the module. For more information see
1775 Documentation/module-signing.txt.
1777 Note that this option adds the OpenSSL development packages as a
1778 kernel build dependency so that the signing tool can use its crypto
1781 !!!WARNING!!! If you enable this option, you MUST make sure that the
1782 module DOES NOT get stripped after being signed. This includes the
1783 debuginfo strip done by some packagers (such as rpmbuild) and
1784 inclusion into an initramfs that wants the module size reduced.
1786 config MODULE_SIG_FORCE
1787 bool "Require modules to be validly signed"
1788 depends on MODULE_SIG
1790 Reject unsigned modules or signed modules for which we don't have a
1791 key. Without this, such modules will simply taint the kernel.
1793 config MODULE_SIG_ALL
1794 bool "Automatically sign all modules"
1796 depends on MODULE_SIG
1798 Sign all modules during make modules_install. Without this option,
1799 modules must be signed manually, using the scripts/sign-file tool.
1801 comment "Do not forget to sign required modules with scripts/sign-file"
1802 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL
1805 prompt "Which hash algorithm should modules be signed with?"
1806 depends on MODULE_SIG
1808 This determines which sort of hashing algorithm will be used during
1809 signature generation. This algorithm _must_ be built into the kernel
1810 directly so that signature verification can take place. It is not
1811 possible to load a signed module containing the algorithm to check
1812 the signature on that module.
1814 config MODULE_SIG_SHA1
1815 bool "Sign modules with SHA-1"
1818 config MODULE_SIG_SHA224
1819 bool "Sign modules with SHA-224"
1820 select CRYPTO_SHA256
1822 config MODULE_SIG_SHA256
1823 bool "Sign modules with SHA-256"
1824 select CRYPTO_SHA256
1826 config MODULE_SIG_SHA384
1827 bool "Sign modules with SHA-384"
1828 select CRYPTO_SHA512
1830 config MODULE_SIG_SHA512
1831 bool "Sign modules with SHA-512"
1832 select CRYPTO_SHA512
1836 config MODULE_SIG_HASH
1838 depends on MODULE_SIG
1839 default "sha1" if MODULE_SIG_SHA1
1840 default "sha224" if MODULE_SIG_SHA224
1841 default "sha256" if MODULE_SIG_SHA256
1842 default "sha384" if MODULE_SIG_SHA384
1843 default "sha512" if MODULE_SIG_SHA512
1845 config MODULE_COMPRESS
1846 bool "Compress modules on installation"
1850 Compresses kernel modules when 'make modules_install' is run; gzip or
1851 xz depending on "Compression algorithm" below.
1853 module-init-tools MAY support gzip, and kmod MAY support gzip and xz.
1855 Out-of-tree kernel modules installed using Kbuild will also be
1856 compressed upon installation.
1858 Note: for modules inside an initrd or initramfs, it's more efficient
1859 to compress the whole initrd or initramfs instead.
1861 Note: This is fully compatible with signed modules.
1866 prompt "Compression algorithm"
1867 depends on MODULE_COMPRESS
1868 default MODULE_COMPRESS_GZIP
1870 This determines which sort of compression will be used during
1871 'make modules_install'.
1873 GZIP (default) and XZ are supported.
1875 config MODULE_COMPRESS_GZIP
1878 config MODULE_COMPRESS_XZ
1883 config TRIM_UNUSED_KSYMS
1884 bool "Trim unused exported kernel symbols"
1885 depends on MODULES && !UNUSED_SYMBOLS
1887 The kernel and some modules make many symbols available for
1888 other modules to use via EXPORT_SYMBOL() and variants. Depending
1889 on the set of modules being selected in your kernel configuration,
1890 many of those exported symbols might never be used.
1892 This option allows for unused exported symbols to be dropped from
1893 the build. In turn, this provides the compiler more opportunities
1894 (especially when using LTO) for optimizing the code and reducing
1895 binary size. This might have some security advantages as well.
1897 If unsure, or if you need to build out-of-tree modules, say N.
1901 config MODULES_TREE_LOOKUP
1903 depends on PERF_EVENTS || TRACING
1905 config INIT_ALL_POSSIBLE
1908 Back when each arch used to define their own cpu_online_mask and
1909 cpu_possible_mask, some of them chose to initialize cpu_possible_mask
1910 with all 1s, and others with all 0s. When they were centralised,
1911 it was better to provide this option than to break all the archs
1912 and have several arch maintainers pursuing me down dark alleys.
1914 source "block/Kconfig"
1916 config PREEMPT_NOTIFIERS
1926 Build a simple ASN.1 grammar compiler that produces a bytecode output
1927 that can be interpreted by the ASN.1 stream decoder and used to
1928 inform it as to what tags are to be expected in a stream and what
1929 functions to call on what tags.
1931 source "kernel/Kconfig.locks"