1 # SPDX-License-Identifier: GPL-2.0
3 # General architecture dependent options
7 # Note: arch/$(SRCARCH)/Kconfig needs to be included first so that it can
8 # override the default values in this file.
10 source "arch/$(SRCARCH)/Kconfig"
12 menu "General architecture-dependent options"
34 tristate "OProfile system profiling"
36 depends on HAVE_OPROFILE
38 select RING_BUFFER_ALLOW_SWAP
40 OProfile is a profiling system capable of profiling the
41 whole system, include the kernel, kernel modules, libraries,
46 config OPROFILE_EVENT_MULTIPLEX
47 bool "OProfile multiplexing support (EXPERIMENTAL)"
49 depends on OPROFILE && X86
51 The number of hardware counters is limited. The multiplexing
52 feature enables OProfile to gather more events than counters
53 are provided by the hardware. This is realized by switching
54 between events at a user specified time interval.
61 config OPROFILE_NMI_TIMER
63 depends on PERF_EVENTS && HAVE_PERF_EVENTS_NMI && !PPC64
68 depends on HAVE_KPROBES
71 Kprobes allows you to trap at almost any kernel address and
72 execute a callback function. register_kprobe() establishes
73 a probepoint and specifies the callback. Kprobes is useful
74 for kernel debugging, non-intrusive instrumentation and testing.
78 bool "Optimize very unlikely/likely branches"
79 depends on HAVE_ARCH_JUMP_LABEL
80 depends on CC_HAS_ASM_GOTO
82 This option enables a transparent branch optimization that
83 makes certain almost-always-true or almost-always-false branch
84 conditions even cheaper to execute within the kernel.
86 Certain performance-sensitive kernel code, such as trace points,
87 scheduler functionality, networking code and KVM have such
88 branches and include support for this optimization technique.
90 If it is detected that the compiler has support for "asm goto",
91 the kernel will compile such branches with just a nop
92 instruction. When the condition flag is toggled to true, the
93 nop will be converted to a jump instruction to execute the
94 conditional block of instructions.
96 This technique lowers overhead and stress on the branch prediction
97 of the processor and generally makes the kernel faster. The update
98 of the condition is slower, but those are always very rare.
100 ( On 32-bit x86, the necessary options added to the compiler
101 flags may increase the size of the kernel slightly. )
103 config STATIC_KEYS_SELFTEST
104 bool "Static key selftest"
105 depends on JUMP_LABEL
107 Boot time self-test of the branch patching code.
111 depends on KPROBES && HAVE_OPTPROBES
112 select TASKS_RCU if PREEMPTION
114 config KPROBES_ON_FTRACE
116 depends on KPROBES && HAVE_KPROBES_ON_FTRACE
117 depends on DYNAMIC_FTRACE_WITH_REGS
119 If function tracer is enabled and the arch supports full
120 passing of pt_regs to function tracing, then kprobes can
121 optimize on top of function tracing.
125 depends on ARCH_SUPPORTS_UPROBES
127 Uprobes is the user-space counterpart to kprobes: they
128 enable instrumentation applications (such as 'perf probe')
129 to establish unintrusive probes in user-space binaries and
130 libraries, by executing handler functions when the probes
131 are hit by user-space applications.
133 ( These probes come in the form of single-byte breakpoints,
134 managed by the kernel and kept transparent to the probed
137 config HAVE_EFFICIENT_UNALIGNED_ACCESS
140 Some architectures are unable to perform unaligned accesses
141 without the use of get_unaligned/put_unaligned. Others are
142 unable to perform such accesses efficiently (e.g. trap on
143 unaligned access and require fixing it up in the exception
146 This symbol should be selected by an architecture if it can
147 perform unaligned accesses efficiently to allow different
148 code paths to be selected for these cases. Some network
149 drivers, for example, could opt to not fix up alignment
150 problems with received packets if doing so would not help
153 See Documentation/core-api/unaligned-memory-access.rst for more
154 information on the topic of unaligned memory accesses.
156 config ARCH_USE_BUILTIN_BSWAP
159 Modern versions of GCC (since 4.4) have builtin functions
160 for handling byte-swapping. Using these, instead of the old
161 inline assembler that the architecture code provides in the
162 __arch_bswapXX() macros, allows the compiler to see what's
163 happening and offers more opportunity for optimisation. In
164 particular, the compiler will be able to combine the byteswap
165 with a nearby load or store and use load-and-swap or
166 store-and-swap instructions if the architecture has them. It
167 should almost *never* result in code which is worse than the
168 hand-coded assembler in <asm/swab.h>. But just in case it
169 does, the use of the builtins is optional.
171 Any architecture with load-and-swap or store-and-swap
172 instructions should set this. And it shouldn't hurt to set it
173 on architectures that don't have such instructions.
177 depends on KPROBES && HAVE_KRETPROBES
179 config USER_RETURN_NOTIFIER
181 depends on HAVE_USER_RETURN_NOTIFIER
183 Provide a kernel-internal notification when a cpu is about to
186 config HAVE_IOREMAP_PROT
192 config HAVE_KRETPROBES
195 config HAVE_OPTPROBES
198 config HAVE_KPROBES_ON_FTRACE
201 config HAVE_FUNCTION_ERROR_INJECTION
208 # An arch should select this if it provides all these things:
210 # task_pt_regs() in asm/processor.h or asm/ptrace.h
211 # arch_has_single_step() if there is hardware single-step support
212 # arch_has_block_step() if there is hardware block-step support
213 # asm/syscall.h supplying asm-generic/syscall.h interface
214 # linux/regset.h user_regset interfaces
215 # CORE_DUMP_USE_REGSET #define'd in linux/elf.h
216 # TIF_SYSCALL_TRACE calls tracehook_report_syscall_{entry,exit}
217 # TIF_NOTIFY_RESUME calls tracehook_notify_resume()
218 # signal delivery calls tracehook_signal_handler()
220 config HAVE_ARCH_TRACEHOOK
223 config HAVE_DMA_CONTIGUOUS
226 config GENERIC_SMP_IDLE_THREAD
229 config GENERIC_IDLE_POLL_SETUP
232 config ARCH_HAS_FORTIFY_SOURCE
235 An architecture should select this when it can successfully
236 build and run with CONFIG_FORTIFY_SOURCE.
239 # Select if the arch provides a historic keepinit alias for the retain_initrd
240 # command line option
242 config ARCH_HAS_KEEPINITRD
245 # Select if arch has all set_memory_ro/rw/x/nx() functions in asm/cacheflush.h
246 config ARCH_HAS_SET_MEMORY
249 # Select if arch has all set_direct_map_invalid/default() functions
250 config ARCH_HAS_SET_DIRECT_MAP
254 # Select if the architecture provides the arch_dma_set_uncached symbol to
255 # either provide an uncached segement alias for a DMA allocation, or
256 # to remap the page tables in place.
258 config ARCH_HAS_DMA_SET_UNCACHED
262 # Select if the architectures provides the arch_dma_clear_uncached symbol
263 # to undo an in-place page table remap for uncached access.
265 config ARCH_HAS_DMA_CLEAR_UNCACHED
268 # Select if arch init_task must go in the __init_task_data section
269 config ARCH_TASK_STRUCT_ON_STACK
272 # Select if arch has its private alloc_task_struct() function
273 config ARCH_TASK_STRUCT_ALLOCATOR
276 config HAVE_ARCH_THREAD_STRUCT_WHITELIST
278 depends on !ARCH_TASK_STRUCT_ALLOCATOR
280 An architecture should select this to provide hardened usercopy
281 knowledge about what region of the thread_struct should be
282 whitelisted for copying to userspace. Normally this is only the
283 FPU registers. Specifically, arch_thread_struct_whitelist()
284 should be implemented. Without this, the entire thread_struct
285 field in task_struct will be left whitelisted.
287 # Select if arch has its private alloc_thread_stack() function
288 config ARCH_THREAD_STACK_ALLOCATOR
291 # Select if arch wants to size task_struct dynamically via arch_task_struct_size:
292 config ARCH_WANTS_DYNAMIC_TASK_STRUCT
295 config ARCH_32BIT_OFF_T
299 All new 32-bit architectures should have 64-bit off_t type on
300 userspace side which corresponds to the loff_t kernel type. This
301 is the requirement for modern ABIs. Some existing architectures
302 still support 32-bit off_t. This option is enabled for all such
303 architectures explicitly.
305 config HAVE_ASM_MODVERSIONS
308 This symbol should be selected by an architecure if it provides
309 <asm/asm-prototypes.h> to support the module versioning for symbols
310 exported from assembly code.
312 config HAVE_REGS_AND_STACK_ACCESS_API
315 This symbol should be selected by an architecure if it supports
316 the API needed to access registers and stack entries from pt_regs,
317 declared in asm/ptrace.h
318 For example the kprobes-based event tracer needs this API.
322 depends on HAVE_REGS_AND_STACK_ACCESS_API
324 This symbol should be selected by an architecture if it
325 supports an implementation of restartable sequences.
327 config HAVE_FUNCTION_ARG_ACCESS_API
330 This symbol should be selected by an architecure if it supports
331 the API needed to access function arguments from pt_regs,
332 declared in asm/ptrace.h
334 config HAVE_HW_BREAKPOINT
336 depends on PERF_EVENTS
338 config HAVE_MIXED_BREAKPOINTS_REGS
340 depends on HAVE_HW_BREAKPOINT
342 Depending on the arch implementation of hardware breakpoints,
343 some of them have separate registers for data and instruction
344 breakpoints addresses, others have mixed registers to store
345 them but define the access type in a control register.
346 Select this option if your arch implements breakpoints under the
349 config HAVE_USER_RETURN_NOTIFIER
352 config HAVE_PERF_EVENTS_NMI
355 System hardware can generate an NMI using the perf event
356 subsystem. Also has support for calculating CPU cycle events
357 to determine how many clock cycles in a given period.
359 config HAVE_HARDLOCKUP_DETECTOR_PERF
361 depends on HAVE_PERF_EVENTS_NMI
363 The arch chooses to use the generic perf-NMI-based hardlockup
364 detector. Must define HAVE_PERF_EVENTS_NMI.
366 config HAVE_NMI_WATCHDOG
370 The arch provides a low level NMI watchdog. It provides
371 asm/nmi.h, and defines its own arch_touch_nmi_watchdog().
373 config HAVE_HARDLOCKUP_DETECTOR_ARCH
375 select HAVE_NMI_WATCHDOG
377 The arch chooses to provide its own hardlockup detector, which is
378 a superset of the HAVE_NMI_WATCHDOG. It also conforms to config
379 interfaces and parameters provided by hardlockup detector subsystem.
381 config HAVE_PERF_REGS
384 Support selective register dumps for perf events. This includes
385 bit-mapping of each registers and a unique architecture id.
387 config HAVE_PERF_USER_STACK_DUMP
390 Support user stack dumps for perf event samples. This needs
391 access to the user stack pointer which is not unified across
394 config HAVE_ARCH_JUMP_LABEL
397 config HAVE_ARCH_JUMP_LABEL_RELATIVE
400 config MMU_GATHER_TABLE_FREE
403 config MMU_GATHER_RCU_TABLE_FREE
405 select MMU_GATHER_TABLE_FREE
407 config MMU_GATHER_PAGE_SIZE
410 config MMU_GATHER_NO_RANGE
413 config MMU_GATHER_NO_GATHER
415 depends on MMU_GATHER_TABLE_FREE
417 config ARCH_HAVE_NMI_SAFE_CMPXCHG
420 config HAVE_ALIGNED_STRUCT_PAGE
423 This makes sure that struct pages are double word aligned and that
424 e.g. the SLUB allocator can perform double word atomic operations
425 on a struct page for better performance. However selecting this
426 might increase the size of a struct page by a word.
428 config HAVE_CMPXCHG_LOCAL
431 config HAVE_CMPXCHG_DOUBLE
434 config ARCH_WEAK_RELEASE_ACQUIRE
437 config ARCH_WANT_IPC_PARSE_VERSION
440 config ARCH_WANT_COMPAT_IPC_PARSE_VERSION
443 config ARCH_WANT_OLD_COMPAT_IPC
444 select ARCH_WANT_COMPAT_IPC_PARSE_VERSION
447 config HAVE_ARCH_SECCOMP_FILTER
450 An arch should select this symbol if it provides all of these things:
452 - syscall_get_arguments()
454 - syscall_set_return_value()
455 - SIGSYS siginfo_t support
456 - secure_computing is called from a ptrace_event()-safe context
457 - secure_computing return value is checked and a return value of -1
458 results in the system call being skipped immediately.
459 - seccomp syscall wired up
461 config SECCOMP_FILTER
463 depends on HAVE_ARCH_SECCOMP_FILTER && SECCOMP && NET
465 Enable tasks to build secure computing environments defined
466 in terms of Berkeley Packet Filter programs which implement
467 task-defined system call filtering polices.
469 See Documentation/userspace-api/seccomp_filter.rst for details.
471 config HAVE_ARCH_STACKLEAK
474 An architecture should select this if it has the code which
475 fills the used part of the kernel stack with the STACKLEAK_POISON
476 value before returning from system calls.
478 config HAVE_STACKPROTECTOR
481 An arch should select this symbol if:
482 - it has implemented a stack canary (e.g. __stack_chk_guard)
484 config CC_HAS_STACKPROTECTOR_NONE
485 def_bool $(cc-option,-fno-stack-protector)
487 config STACKPROTECTOR
488 bool "Stack Protector buffer overflow detection"
489 depends on HAVE_STACKPROTECTOR
490 depends on $(cc-option,-fstack-protector)
493 This option turns on the "stack-protector" GCC feature. This
494 feature puts, at the beginning of functions, a canary value on
495 the stack just before the return address, and validates
496 the value just before actually returning. Stack based buffer
497 overflows (that need to overwrite this return address) now also
498 overwrite the canary, which gets detected and the attack is then
499 neutralized via a kernel panic.
501 Functions will have the stack-protector canary logic added if they
502 have an 8-byte or larger character array on the stack.
504 This feature requires gcc version 4.2 or above, or a distribution
505 gcc with the feature backported ("-fstack-protector").
507 On an x86 "defconfig" build, this feature adds canary checks to
508 about 3% of all kernel functions, which increases kernel code size
511 config STACKPROTECTOR_STRONG
512 bool "Strong Stack Protector"
513 depends on STACKPROTECTOR
514 depends on $(cc-option,-fstack-protector-strong)
517 Functions will have the stack-protector canary logic added in any
518 of the following conditions:
520 - local variable's address used as part of the right hand side of an
521 assignment or function argument
522 - local variable is an array (or union containing an array),
523 regardless of array type or length
524 - uses register local variables
526 This feature requires gcc version 4.9 or above, or a distribution
527 gcc with the feature backported ("-fstack-protector-strong").
529 On an x86 "defconfig" build, this feature adds canary checks to
530 about 20% of all kernel functions, which increases the kernel code
533 config ARCH_SUPPORTS_SHADOW_CALL_STACK
536 An architecture should select this if it supports Clang's Shadow
537 Call Stack and implements runtime support for shadow stack
540 config SHADOW_CALL_STACK
541 bool "Clang Shadow Call Stack"
542 depends on CC_IS_CLANG && ARCH_SUPPORTS_SHADOW_CALL_STACK
543 depends on DYNAMIC_FTRACE_WITH_REGS || !FUNCTION_GRAPH_TRACER
545 This option enables Clang's Shadow Call Stack, which uses a
546 shadow stack to protect function return addresses from being
547 overwritten by an attacker. More information can be found in
548 Clang's documentation:
550 https://clang.llvm.org/docs/ShadowCallStack.html
552 Note that security guarantees in the kernel differ from the
553 ones documented for user space. The kernel must store addresses
554 of shadow stacks in memory, which means an attacker capable of
555 reading and writing arbitrary memory may be able to locate them
556 and hijack control flow by modifying the stacks.
558 config HAVE_ARCH_WITHIN_STACK_FRAMES
561 An architecture should select this if it can walk the kernel stack
562 frames to determine if an object is part of either the arguments
563 or local variables (i.e. that it excludes saved return addresses,
564 and similar) by implementing an inline arch_within_stack_frames(),
565 which is used by CONFIG_HARDENED_USERCOPY.
567 config HAVE_CONTEXT_TRACKING
570 Provide kernel/user boundaries probes necessary for subsystems
571 that need it, such as userspace RCU extended quiescent state.
572 Syscalls need to be wrapped inside user_exit()-user_enter(), either
573 optimized behind static key or through the slow path using TIF_NOHZ
574 flag. Exceptions handlers must be wrapped as well. Irqs are already
575 protected inside rcu_irq_enter/rcu_irq_exit() but preemption or signal
576 handling on irq exit still need to be protected.
581 Arch relies on TIF_NOHZ and syscall slow path to implement context
582 tracking calls to user_enter()/user_exit().
584 config HAVE_VIRT_CPU_ACCOUNTING
587 config ARCH_HAS_SCALED_CPUTIME
590 config HAVE_VIRT_CPU_ACCOUNTING_GEN
594 With VIRT_CPU_ACCOUNTING_GEN, cputime_t becomes 64-bit.
595 Before enabling this option, arch code must be audited
596 to ensure there are no races in concurrent read/write of
597 cputime_t. For example, reading/writing 64-bit cputime_t on
598 some 32-bit arches may require multiple accesses, so proper
599 locking is needed to protect against concurrent accesses.
602 config HAVE_IRQ_TIME_ACCOUNTING
605 Archs need to ensure they use a high enough resolution clock to
606 support irq time accounting and then call enable_sched_clock_irqtime().
611 Archs that select this are able to move page tables at the PMD level.
613 config HAVE_ARCH_TRANSPARENT_HUGEPAGE
616 config HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD
619 config HAVE_ARCH_HUGE_VMAP
622 config ARCH_WANT_HUGE_PMD_SHARE
625 config HAVE_ARCH_SOFT_DIRTY
628 config HAVE_MOD_ARCH_SPECIFIC
631 The arch uses struct mod_arch_specific to store data. Many arches
632 just need a simple module loader without arch specific data - those
633 should not enable this.
635 config MODULES_USE_ELF_RELA
638 Modules only use ELF RELA relocations. Modules with ELF REL
639 relocations will give an error.
641 config MODULES_USE_ELF_REL
644 Modules only use ELF REL relocations. Modules with ELF RELA
645 relocations will give an error.
647 config HAVE_IRQ_EXIT_ON_IRQ_STACK
650 Architecture doesn't only execute the irq handler on the irq stack
651 but also irq_exit(). This way we can process softirqs on this irq
652 stack instead of switching to a new one when we call __do_softirq()
653 in the end of an hardirq.
654 This spares a stack switch and improves cache usage on softirq
657 config PGTABLE_LEVELS
661 config ARCH_HAS_ELF_RANDOMIZE
664 An architecture supports choosing randomized locations for
665 stack, mmap, brk, and ET_DYN. Defined functions:
667 - arch_randomize_brk()
669 config HAVE_ARCH_MMAP_RND_BITS
672 An arch should select this symbol if it supports setting a variable
673 number of bits for use in establishing the base address for mmap
674 allocations, has MMU enabled and provides values for both:
675 - ARCH_MMAP_RND_BITS_MIN
676 - ARCH_MMAP_RND_BITS_MAX
678 config HAVE_EXIT_THREAD
681 An architecture implements exit_thread.
683 config ARCH_MMAP_RND_BITS_MIN
686 config ARCH_MMAP_RND_BITS_MAX
689 config ARCH_MMAP_RND_BITS_DEFAULT
692 config ARCH_MMAP_RND_BITS
693 int "Number of bits to use for ASLR of mmap base address" if EXPERT
694 range ARCH_MMAP_RND_BITS_MIN ARCH_MMAP_RND_BITS_MAX
695 default ARCH_MMAP_RND_BITS_DEFAULT if ARCH_MMAP_RND_BITS_DEFAULT
696 default ARCH_MMAP_RND_BITS_MIN
697 depends on HAVE_ARCH_MMAP_RND_BITS
699 This value can be used to select the number of bits to use to
700 determine the random offset to the base address of vma regions
701 resulting from mmap allocations. This value will be bounded
702 by the architecture's minimum and maximum supported values.
704 This value can be changed after boot using the
705 /proc/sys/vm/mmap_rnd_bits tunable
707 config HAVE_ARCH_MMAP_RND_COMPAT_BITS
710 An arch should select this symbol if it supports running applications
711 in compatibility mode, supports setting a variable number of bits for
712 use in establishing the base address for mmap allocations, has MMU
713 enabled and provides values for both:
714 - ARCH_MMAP_RND_COMPAT_BITS_MIN
715 - ARCH_MMAP_RND_COMPAT_BITS_MAX
717 config ARCH_MMAP_RND_COMPAT_BITS_MIN
720 config ARCH_MMAP_RND_COMPAT_BITS_MAX
723 config ARCH_MMAP_RND_COMPAT_BITS_DEFAULT
726 config ARCH_MMAP_RND_COMPAT_BITS
727 int "Number of bits to use for ASLR of mmap base address for compatible applications" if EXPERT
728 range ARCH_MMAP_RND_COMPAT_BITS_MIN ARCH_MMAP_RND_COMPAT_BITS_MAX
729 default ARCH_MMAP_RND_COMPAT_BITS_DEFAULT if ARCH_MMAP_RND_COMPAT_BITS_DEFAULT
730 default ARCH_MMAP_RND_COMPAT_BITS_MIN
731 depends on HAVE_ARCH_MMAP_RND_COMPAT_BITS
733 This value can be used to select the number of bits to use to
734 determine the random offset to the base address of vma regions
735 resulting from mmap allocations for compatible applications This
736 value will be bounded by the architecture's minimum and maximum
739 This value can be changed after boot using the
740 /proc/sys/vm/mmap_rnd_compat_bits tunable
742 config HAVE_ARCH_COMPAT_MMAP_BASES
745 This allows 64bit applications to invoke 32-bit mmap() syscall
746 and vice-versa 32-bit applications to call 64-bit mmap().
747 Required for applications doing different bitness syscalls.
749 # This allows to use a set of generic functions to determine mmap base
750 # address by giving priority to top-down scheme only if the process
751 # is not in legacy mode (compat task, unlimited stack size or
752 # sysctl_legacy_va_layout).
753 # Architecture that selects this option can provide its own version of:
755 config ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT
758 select ARCH_HAS_ELF_RANDOMIZE
760 config HAVE_STACK_VALIDATION
763 Architecture supports the 'objtool check' host tool command, which
764 performs compile-time stack metadata validation.
766 config HAVE_RELIABLE_STACKTRACE
769 Architecture has either save_stack_trace_tsk_reliable() or
770 arch_stack_walk_reliable() function which only returns a stack trace
771 if it can guarantee the trace is reliable.
773 config HAVE_ARCH_HASH
777 If this is set, the architecture provides an <asm/hash.h>
778 file which provides platform-specific implementations of some
779 functions in <linux/hash.h> or fs/namei.c.
781 config HAVE_ARCH_NVRAM_OPS
790 config CLONE_BACKWARDS
793 Architecture has tls passed as the 4th argument of clone(2),
796 config CLONE_BACKWARDS2
799 Architecture has the first two arguments of clone(2) swapped.
801 config CLONE_BACKWARDS3
804 Architecture has tls passed as the 3rd argument of clone(2),
807 config ODD_RT_SIGACTION
810 Architecture has unusual rt_sigaction(2) arguments
812 config OLD_SIGSUSPEND
815 Architecture has old sigsuspend(2) syscall, of one-argument variety
817 config OLD_SIGSUSPEND3
820 Even weirder antique ABI - three-argument sigsuspend(2)
825 Architecture has old sigaction(2) syscall. Nope, not the same
826 as OLD_SIGSUSPEND | OLD_SIGSUSPEND3 - alpha has sigsuspend(2),
827 but fairly different variant of sigaction(2), thanks to OSF/1
830 config COMPAT_OLD_SIGACTION
833 config COMPAT_32BIT_TIME
834 bool "Provide system calls for 32-bit time_t"
835 default !64BIT || COMPAT
837 This enables 32 bit time_t support in addition to 64 bit time_t support.
838 This is relevant on all 32-bit architectures, and 64-bit architectures
839 as part of compat syscall handling.
841 config ARCH_NO_PREEMPT
844 config ARCH_SUPPORTS_RT
847 config CPU_NO_EFFICIENT_FFS
850 config HAVE_ARCH_VMAP_STACK
853 An arch should select this symbol if it can support kernel stacks
854 in vmalloc space. This means:
856 - vmalloc space must be large enough to hold many kernel stacks.
857 This may rule out many 32-bit architectures.
859 - Stacks in vmalloc space need to work reliably. For example, if
860 vmap page tables are created on demand, either this mechanism
861 needs to work while the stack points to a virtual address with
862 unpopulated page tables or arch code (switch_to() and switch_mm(),
863 most likely) needs to ensure that the stack's page table entries
864 are populated before running on a possibly unpopulated stack.
866 - If the stack overflows into a guard page, something reasonable
867 should happen. The definition of "reasonable" is flexible, but
868 instantly rebooting without logging anything would be unfriendly.
872 bool "Use a virtually-mapped stack"
873 depends on HAVE_ARCH_VMAP_STACK
874 depends on !KASAN || KASAN_VMALLOC
876 Enable this if you want the use virtually-mapped kernel stacks
877 with guard pages. This causes kernel stack overflows to be
878 caught immediately rather than causing difficult-to-diagnose
881 To use this with KASAN, the architecture must support backing
882 virtual mappings with real shadow memory, and KASAN_VMALLOC must
885 config ARCH_OPTIONAL_KERNEL_RWX
888 config ARCH_OPTIONAL_KERNEL_RWX_DEFAULT
891 config ARCH_HAS_STRICT_KERNEL_RWX
894 config STRICT_KERNEL_RWX
895 bool "Make kernel text and rodata read-only" if ARCH_OPTIONAL_KERNEL_RWX
896 depends on ARCH_HAS_STRICT_KERNEL_RWX
897 default !ARCH_OPTIONAL_KERNEL_RWX || ARCH_OPTIONAL_KERNEL_RWX_DEFAULT
899 If this is set, kernel text and rodata memory will be made read-only,
900 and non-text memory will be made non-executable. This provides
901 protection against certain security exploits (e.g. executing the heap
904 These features are considered standard security practice these days.
905 You should say Y here in almost all cases.
907 config ARCH_HAS_STRICT_MODULE_RWX
910 config STRICT_MODULE_RWX
911 bool "Set loadable kernel module data as NX and text as RO" if ARCH_OPTIONAL_KERNEL_RWX
912 depends on ARCH_HAS_STRICT_MODULE_RWX && MODULES
913 default !ARCH_OPTIONAL_KERNEL_RWX || ARCH_OPTIONAL_KERNEL_RWX_DEFAULT
915 If this is set, module text and rodata memory will be made read-only,
916 and non-text memory will be made non-executable. This provides
917 protection against certain security exploits (e.g. writing to text)
919 # select if the architecture provides an asm/dma-direct.h header
920 config ARCH_HAS_PHYS_TO_DMA
923 config HAVE_ARCH_COMPILER_H
926 An architecture can select this if it provides an
927 asm/compiler.h header that should be included after
928 linux/compiler-*.h in order to override macro definitions that those
929 headers generally provide.
931 config HAVE_ARCH_PREL32_RELOCATIONS
934 May be selected by an architecture if it supports place-relative
935 32-bit relocations, both in the toolchain and in the module loader,
936 in which case relative references can be used in special sections
937 for PCI fixup, initcalls etc which are only half the size on 64 bit
938 architectures, and don't require runtime relocation on relocatable
941 config ARCH_USE_MEMREMAP_PROT
944 config LOCK_EVENT_COUNTS
945 bool "Locking event counts collection"
948 Enable light-weight counting of various locking related events
949 in the system with minimal performance impact. This reduces
950 the chance of application behavior change because of timing
951 differences. The counts are reported via debugfs.
953 # Select if the architecture has support for applying RELR relocations.
958 bool "Use RELR relocation packing"
959 depends on ARCH_HAS_RELR && TOOLS_SUPPORT_RELR
962 Store the kernel's dynamic relocations in the RELR relocation packing
963 format. Requires a compatible linker (LLD supports this feature), as
964 well as compatible NM and OBJCOPY utilities (llvm-nm and llvm-objcopy
967 config ARCH_HAS_MEM_ENCRYPT
970 config HAVE_SPARSE_SYSCALL_NR
973 An architecture should select this if its syscall numbering is sparse
974 to save space. For example, MIPS architecture has a syscall array with
975 entries at 4000, 5000 and 6000 locations. This option turns on syscall
976 related optimizations for a given architecture.
978 source "kernel/gcov/Kconfig"
980 source "scripts/gcc-plugins/Kconfig"