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"
27 config ARCH_HAS_SUBPAGE_FAULTS
30 Select if the architecture can check permissions at sub-page
31 granularity (e.g. arm64 MTE). The probe_user_*() functions
43 depends on HAVE_KPROBES
45 select TASKS_RCU if PREEMPTION
47 Kprobes allows you to trap at almost any kernel address and
48 execute a callback function. register_kprobe() establishes
49 a probepoint and specifies the callback. Kprobes is useful
50 for kernel debugging, non-intrusive instrumentation and testing.
54 bool "Optimize very unlikely/likely branches"
55 depends on HAVE_ARCH_JUMP_LABEL
56 select OBJTOOL if HAVE_JUMP_LABEL_HACK
58 This option enables a transparent branch optimization that
59 makes certain almost-always-true or almost-always-false branch
60 conditions even cheaper to execute within the kernel.
62 Certain performance-sensitive kernel code, such as trace points,
63 scheduler functionality, networking code and KVM have such
64 branches and include support for this optimization technique.
66 If it is detected that the compiler has support for "asm goto",
67 the kernel will compile such branches with just a nop
68 instruction. When the condition flag is toggled to true, the
69 nop will be converted to a jump instruction to execute the
70 conditional block of instructions.
72 This technique lowers overhead and stress on the branch prediction
73 of the processor and generally makes the kernel faster. The update
74 of the condition is slower, but those are always very rare.
76 ( On 32-bit x86, the necessary options added to the compiler
77 flags may increase the size of the kernel slightly. )
79 config STATIC_KEYS_SELFTEST
80 bool "Static key selftest"
83 Boot time self-test of the branch patching code.
85 config STATIC_CALL_SELFTEST
86 bool "Static call selftest"
87 depends on HAVE_STATIC_CALL
89 Boot time self-test of the call patching code.
93 depends on KPROBES && HAVE_OPTPROBES
94 select TASKS_RCU if PREEMPTION
96 config KPROBES_ON_FTRACE
98 depends on KPROBES && HAVE_KPROBES_ON_FTRACE
99 depends on DYNAMIC_FTRACE_WITH_REGS
101 If function tracer is enabled and the arch supports full
102 passing of pt_regs to function tracing, then kprobes can
103 optimize on top of function tracing.
107 depends on ARCH_SUPPORTS_UPROBES
109 Uprobes is the user-space counterpart to kprobes: they
110 enable instrumentation applications (such as 'perf probe')
111 to establish unintrusive probes in user-space binaries and
112 libraries, by executing handler functions when the probes
113 are hit by user-space applications.
115 ( These probes come in the form of single-byte breakpoints,
116 managed by the kernel and kept transparent to the probed
119 config HAVE_64BIT_ALIGNED_ACCESS
120 def_bool 64BIT && !HAVE_EFFICIENT_UNALIGNED_ACCESS
122 Some architectures require 64 bit accesses to be 64 bit
123 aligned, which also requires structs containing 64 bit values
124 to be 64 bit aligned too. This includes some 32 bit
125 architectures which can do 64 bit accesses, as well as 64 bit
126 architectures without unaligned access.
128 This symbol should be selected by an architecture if 64 bit
129 accesses are required to be 64 bit aligned in this way even
130 though it is not a 64 bit architecture.
132 See Documentation/core-api/unaligned-memory-access.rst for
133 more information on the topic of unaligned memory accesses.
135 config HAVE_EFFICIENT_UNALIGNED_ACCESS
138 Some architectures are unable to perform unaligned accesses
139 without the use of get_unaligned/put_unaligned. Others are
140 unable to perform such accesses efficiently (e.g. trap on
141 unaligned access and require fixing it up in the exception
144 This symbol should be selected by an architecture if it can
145 perform unaligned accesses efficiently to allow different
146 code paths to be selected for these cases. Some network
147 drivers, for example, could opt to not fix up alignment
148 problems with received packets if doing so would not help
151 See Documentation/core-api/unaligned-memory-access.rst for more
152 information on the topic of unaligned memory accesses.
154 config ARCH_USE_BUILTIN_BSWAP
157 Modern versions of GCC (since 4.4) have builtin functions
158 for handling byte-swapping. Using these, instead of the old
159 inline assembler that the architecture code provides in the
160 __arch_bswapXX() macros, allows the compiler to see what's
161 happening and offers more opportunity for optimisation. In
162 particular, the compiler will be able to combine the byteswap
163 with a nearby load or store and use load-and-swap or
164 store-and-swap instructions if the architecture has them. It
165 should almost *never* result in code which is worse than the
166 hand-coded assembler in <asm/swab.h>. But just in case it
167 does, the use of the builtins is optional.
169 Any architecture with load-and-swap or store-and-swap
170 instructions should set this. And it shouldn't hurt to set it
171 on architectures that don't have such instructions.
175 depends on KPROBES && (HAVE_KRETPROBES || HAVE_RETHOOK)
177 config KRETPROBE_ON_RETHOOK
179 depends on HAVE_RETHOOK
180 depends on KRETPROBES
183 config USER_RETURN_NOTIFIER
185 depends on HAVE_USER_RETURN_NOTIFIER
187 Provide a kernel-internal notification when a cpu is about to
190 config HAVE_IOREMAP_PROT
196 config HAVE_KRETPROBES
199 config HAVE_OPTPROBES
202 config HAVE_KPROBES_ON_FTRACE
205 config ARCH_CORRECT_STACKTRACE_ON_KRETPROBE
208 Since kretprobes modifies return address on the stack, the
209 stacktrace may see the kretprobe trampoline address instead
210 of correct one. If the architecture stacktrace code and
211 unwinder can adjust such entries, select this configuration.
213 config HAVE_FUNCTION_ERROR_INJECTION
219 config HAVE_FUNCTION_DESCRIPTORS
222 config TRACE_IRQFLAGS_SUPPORT
225 config TRACE_IRQFLAGS_NMI_SUPPORT
229 # An arch should select this if it provides all these things:
231 # task_pt_regs() in asm/processor.h or asm/ptrace.h
232 # arch_has_single_step() if there is hardware single-step support
233 # arch_has_block_step() if there is hardware block-step support
234 # asm/syscall.h supplying asm-generic/syscall.h interface
235 # linux/regset.h user_regset interfaces
236 # CORE_DUMP_USE_REGSET #define'd in linux/elf.h
237 # TIF_SYSCALL_TRACE calls ptrace_report_syscall_{entry,exit}
238 # TIF_NOTIFY_RESUME calls resume_user_mode_work()
240 config HAVE_ARCH_TRACEHOOK
243 config HAVE_DMA_CONTIGUOUS
246 config GENERIC_SMP_IDLE_THREAD
249 config GENERIC_IDLE_POLL_SETUP
252 config ARCH_HAS_FORTIFY_SOURCE
255 An architecture should select this when it can successfully
256 build and run with CONFIG_FORTIFY_SOURCE.
259 # Select if the arch provides a historic keepinit alias for the retain_initrd
260 # command line option
262 config ARCH_HAS_KEEPINITRD
265 # Select if arch has all set_memory_ro/rw/x/nx() functions in asm/cacheflush.h
266 config ARCH_HAS_SET_MEMORY
269 # Select if arch has all set_direct_map_invalid/default() functions
270 config ARCH_HAS_SET_DIRECT_MAP
274 # Select if the architecture provides the arch_dma_set_uncached symbol to
275 # either provide an uncached segment alias for a DMA allocation, or
276 # to remap the page tables in place.
278 config ARCH_HAS_DMA_SET_UNCACHED
282 # Select if the architectures provides the arch_dma_clear_uncached symbol
283 # to undo an in-place page table remap for uncached access.
285 config ARCH_HAS_DMA_CLEAR_UNCACHED
288 # Select if arch init_task must go in the __init_task_data section
289 config ARCH_TASK_STRUCT_ON_STACK
292 # Select if arch has its private alloc_task_struct() function
293 config ARCH_TASK_STRUCT_ALLOCATOR
296 config HAVE_ARCH_THREAD_STRUCT_WHITELIST
298 depends on !ARCH_TASK_STRUCT_ALLOCATOR
300 An architecture should select this to provide hardened usercopy
301 knowledge about what region of the thread_struct should be
302 whitelisted for copying to userspace. Normally this is only the
303 FPU registers. Specifically, arch_thread_struct_whitelist()
304 should be implemented. Without this, the entire thread_struct
305 field in task_struct will be left whitelisted.
307 # Select if arch has its private alloc_thread_stack() function
308 config ARCH_THREAD_STACK_ALLOCATOR
311 # Select if arch wants to size task_struct dynamically via arch_task_struct_size:
312 config ARCH_WANTS_DYNAMIC_TASK_STRUCT
315 config ARCH_WANTS_NO_INSTR
318 An architecture should select this if the noinstr macro is being used on
319 functions to denote that the toolchain should avoid instrumenting such
320 functions and is required for correctness.
322 config ARCH_32BIT_OFF_T
326 All new 32-bit architectures should have 64-bit off_t type on
327 userspace side which corresponds to the loff_t kernel type. This
328 is the requirement for modern ABIs. Some existing architectures
329 still support 32-bit off_t. This option is enabled for all such
330 architectures explicitly.
332 # Selected by 64 bit architectures which have a 32 bit f_tinode in struct ustat
333 config ARCH_32BIT_USTAT_F_TINODE
336 config HAVE_ASM_MODVERSIONS
339 This symbol should be selected by an architecture if it provides
340 <asm/asm-prototypes.h> to support the module versioning for symbols
341 exported from assembly code.
343 config HAVE_REGS_AND_STACK_ACCESS_API
346 This symbol should be selected by an architecture if it supports
347 the API needed to access registers and stack entries from pt_regs,
348 declared in asm/ptrace.h
349 For example the kprobes-based event tracer needs this API.
353 depends on HAVE_REGS_AND_STACK_ACCESS_API
355 This symbol should be selected by an architecture if it
356 supports an implementation of restartable sequences.
358 config HAVE_FUNCTION_ARG_ACCESS_API
361 This symbol should be selected by an architecture if it supports
362 the API needed to access function arguments from pt_regs,
363 declared in asm/ptrace.h
365 config HAVE_HW_BREAKPOINT
367 depends on PERF_EVENTS
369 config HAVE_MIXED_BREAKPOINTS_REGS
371 depends on HAVE_HW_BREAKPOINT
373 Depending on the arch implementation of hardware breakpoints,
374 some of them have separate registers for data and instruction
375 breakpoints addresses, others have mixed registers to store
376 them but define the access type in a control register.
377 Select this option if your arch implements breakpoints under the
380 config HAVE_USER_RETURN_NOTIFIER
383 config HAVE_PERF_EVENTS_NMI
386 System hardware can generate an NMI using the perf event
387 subsystem. Also has support for calculating CPU cycle events
388 to determine how many clock cycles in a given period.
390 config HAVE_HARDLOCKUP_DETECTOR_PERF
392 depends on HAVE_PERF_EVENTS_NMI
394 The arch chooses to use the generic perf-NMI-based hardlockup
395 detector. Must define HAVE_PERF_EVENTS_NMI.
397 config HAVE_NMI_WATCHDOG
401 The arch provides a low level NMI watchdog. It provides
402 asm/nmi.h, and defines its own arch_touch_nmi_watchdog().
404 config HAVE_HARDLOCKUP_DETECTOR_ARCH
406 select HAVE_NMI_WATCHDOG
408 The arch chooses to provide its own hardlockup detector, which is
409 a superset of the HAVE_NMI_WATCHDOG. It also conforms to config
410 interfaces and parameters provided by hardlockup detector subsystem.
412 config HAVE_PERF_REGS
415 Support selective register dumps for perf events. This includes
416 bit-mapping of each registers and a unique architecture id.
418 config HAVE_PERF_USER_STACK_DUMP
421 Support user stack dumps for perf event samples. This needs
422 access to the user stack pointer which is not unified across
425 config HAVE_ARCH_JUMP_LABEL
428 config HAVE_ARCH_JUMP_LABEL_RELATIVE
431 config MMU_GATHER_TABLE_FREE
434 config MMU_GATHER_RCU_TABLE_FREE
436 select MMU_GATHER_TABLE_FREE
438 config MMU_GATHER_PAGE_SIZE
441 config MMU_GATHER_NO_RANGE
443 select MMU_GATHER_MERGE_VMAS
445 config MMU_GATHER_NO_FLUSH_CACHE
448 config MMU_GATHER_MERGE_VMAS
451 config MMU_GATHER_NO_GATHER
453 depends on MMU_GATHER_TABLE_FREE
455 config ARCH_WANT_IRQS_OFF_ACTIVATE_MM
458 Temporary select until all architectures can be converted to have
459 irqs disabled over activate_mm. Architectures that do IPI based TLB
460 shootdowns should enable this.
462 config ARCH_HAVE_NMI_SAFE_CMPXCHG
465 config HAVE_ALIGNED_STRUCT_PAGE
468 This makes sure that struct pages are double word aligned and that
469 e.g. the SLUB allocator can perform double word atomic operations
470 on a struct page for better performance. However selecting this
471 might increase the size of a struct page by a word.
473 config HAVE_CMPXCHG_LOCAL
476 config HAVE_CMPXCHG_DOUBLE
479 config ARCH_WEAK_RELEASE_ACQUIRE
482 config ARCH_WANT_IPC_PARSE_VERSION
485 config ARCH_WANT_COMPAT_IPC_PARSE_VERSION
488 config ARCH_WANT_OLD_COMPAT_IPC
489 select ARCH_WANT_COMPAT_IPC_PARSE_VERSION
492 config HAVE_ARCH_SECCOMP
495 An arch should select this symbol to support seccomp mode 1 (the fixed
496 syscall policy), and must provide an overrides for __NR_seccomp_sigreturn,
497 and compat syscalls if the asm-generic/seccomp.h defaults need adjustment:
498 - __NR_seccomp_read_32
499 - __NR_seccomp_write_32
500 - __NR_seccomp_exit_32
501 - __NR_seccomp_sigreturn_32
503 config HAVE_ARCH_SECCOMP_FILTER
505 select HAVE_ARCH_SECCOMP
507 An arch should select this symbol if it provides all of these things:
508 - all the requirements for HAVE_ARCH_SECCOMP
510 - syscall_get_arguments()
512 - syscall_set_return_value()
513 - SIGSYS siginfo_t support
514 - secure_computing is called from a ptrace_event()-safe context
515 - secure_computing return value is checked and a return value of -1
516 results in the system call being skipped immediately.
517 - seccomp syscall wired up
518 - if !HAVE_SPARSE_SYSCALL_NR, have SECCOMP_ARCH_NATIVE,
519 SECCOMP_ARCH_NATIVE_NR, SECCOMP_ARCH_NATIVE_NAME defined. If
520 COMPAT is supported, have the SECCOMP_ARCH_COMPAT* defines too.
523 prompt "Enable seccomp to safely execute untrusted bytecode"
525 depends on HAVE_ARCH_SECCOMP
527 This kernel feature is useful for number crunching applications
528 that may need to handle untrusted bytecode during their
529 execution. By using pipes or other transports made available
530 to the process as file descriptors supporting the read/write
531 syscalls, it's possible to isolate those applications in their
532 own address space using seccomp. Once seccomp is enabled via
533 prctl(PR_SET_SECCOMP) or the seccomp() syscall, it cannot be
534 disabled and the task is only allowed to execute a few safe
535 syscalls defined by each seccomp mode.
539 config SECCOMP_FILTER
541 depends on HAVE_ARCH_SECCOMP_FILTER && SECCOMP && NET
543 Enable tasks to build secure computing environments defined
544 in terms of Berkeley Packet Filter programs which implement
545 task-defined system call filtering polices.
547 See Documentation/userspace-api/seccomp_filter.rst for details.
549 config SECCOMP_CACHE_DEBUG
550 bool "Show seccomp filter cache status in /proc/pid/seccomp_cache"
551 depends on SECCOMP_FILTER && !HAVE_SPARSE_SYSCALL_NR
554 This enables the /proc/pid/seccomp_cache interface to monitor
555 seccomp cache data. The file format is subject to change. Reading
556 the file requires CAP_SYS_ADMIN.
558 This option is for debugging only. Enabling presents the risk that
559 an adversary may be able to infer the seccomp filter logic.
563 config HAVE_ARCH_STACKLEAK
566 An architecture should select this if it has the code which
567 fills the used part of the kernel stack with the STACKLEAK_POISON
568 value before returning from system calls.
570 config HAVE_STACKPROTECTOR
573 An arch should select this symbol if:
574 - it has implemented a stack canary (e.g. __stack_chk_guard)
576 config STACKPROTECTOR
577 bool "Stack Protector buffer overflow detection"
578 depends on HAVE_STACKPROTECTOR
579 depends on $(cc-option,-fstack-protector)
582 This option turns on the "stack-protector" GCC feature. This
583 feature puts, at the beginning of functions, a canary value on
584 the stack just before the return address, and validates
585 the value just before actually returning. Stack based buffer
586 overflows (that need to overwrite this return address) now also
587 overwrite the canary, which gets detected and the attack is then
588 neutralized via a kernel panic.
590 Functions will have the stack-protector canary logic added if they
591 have an 8-byte or larger character array on the stack.
593 This feature requires gcc version 4.2 or above, or a distribution
594 gcc with the feature backported ("-fstack-protector").
596 On an x86 "defconfig" build, this feature adds canary checks to
597 about 3% of all kernel functions, which increases kernel code size
600 config STACKPROTECTOR_STRONG
601 bool "Strong Stack Protector"
602 depends on STACKPROTECTOR
603 depends on $(cc-option,-fstack-protector-strong)
606 Functions will have the stack-protector canary logic added in any
607 of the following conditions:
609 - local variable's address used as part of the right hand side of an
610 assignment or function argument
611 - local variable is an array (or union containing an array),
612 regardless of array type or length
613 - uses register local variables
615 This feature requires gcc version 4.9 or above, or a distribution
616 gcc with the feature backported ("-fstack-protector-strong").
618 On an x86 "defconfig" build, this feature adds canary checks to
619 about 20% of all kernel functions, which increases the kernel code
622 config ARCH_SUPPORTS_SHADOW_CALL_STACK
625 An architecture should select this if it supports the compiler's
626 Shadow Call Stack and implements runtime support for shadow stack
629 config SHADOW_CALL_STACK
630 bool "Shadow Call Stack"
631 depends on ARCH_SUPPORTS_SHADOW_CALL_STACK
632 depends on DYNAMIC_FTRACE_WITH_REGS || !FUNCTION_GRAPH_TRACER
634 This option enables the compiler's Shadow Call Stack, which
635 uses a shadow stack to protect function return addresses from
636 being overwritten by an attacker. More information can be found
637 in the compiler's documentation:
639 - Clang: https://clang.llvm.org/docs/ShadowCallStack.html
640 - GCC: https://gcc.gnu.org/onlinedocs/gcc/Instrumentation-Options.html#Instrumentation-Options
642 Note that security guarantees in the kernel differ from the
643 ones documented for user space. The kernel must store addresses
644 of shadow stacks in memory, which means an attacker capable of
645 reading and writing arbitrary memory may be able to locate them
646 and hijack control flow by modifying the stacks.
651 Selected if the kernel will be built using the compiler's LTO feature.
657 Selected if the kernel will be built using Clang's LTO feature.
659 config ARCH_SUPPORTS_LTO_CLANG
662 An architecture should select this option if it supports:
663 - compiling with Clang,
664 - compiling inline assembly with Clang's integrated assembler,
665 - and linking with LLD.
667 config ARCH_SUPPORTS_LTO_CLANG_THIN
670 An architecture should select this option if it can support Clang's
675 depends on CC_IS_CLANG && LD_IS_LLD && AS_IS_LLVM
676 depends on $(success,$(NM) --help | head -n 1 | grep -qi llvm)
677 depends on $(success,$(AR) --help | head -n 1 | grep -qi llvm)
678 depends on ARCH_SUPPORTS_LTO_CLANG
679 depends on !FTRACE_MCOUNT_USE_RECORDMCOUNT
680 depends on !KASAN || KASAN_HW_TAGS
681 depends on !GCOV_KERNEL
683 The compiler and Kconfig options support building with Clang's
687 prompt "Link Time Optimization (LTO)"
690 This option enables Link Time Optimization (LTO), which allows the
691 compiler to optimize binaries globally.
693 If unsure, select LTO_NONE. Note that LTO is very resource-intensive
694 so it's disabled by default.
699 Build the kernel normally, without Link Time Optimization (LTO).
701 config LTO_CLANG_FULL
702 bool "Clang Full LTO (EXPERIMENTAL)"
703 depends on HAS_LTO_CLANG
704 depends on !COMPILE_TEST
707 This option enables Clang's full Link Time Optimization (LTO), which
708 allows the compiler to optimize the kernel globally. If you enable
709 this option, the compiler generates LLVM bitcode instead of ELF
710 object files, and the actual compilation from bitcode happens at
711 the LTO link step, which may take several minutes depending on the
712 kernel configuration. More information can be found from LLVM's
715 https://llvm.org/docs/LinkTimeOptimization.html
717 During link time, this option can use a large amount of RAM, and
718 may take much longer than the ThinLTO option.
720 config LTO_CLANG_THIN
721 bool "Clang ThinLTO (EXPERIMENTAL)"
722 depends on HAS_LTO_CLANG && ARCH_SUPPORTS_LTO_CLANG_THIN
725 This option enables Clang's ThinLTO, which allows for parallel
726 optimization and faster incremental compiles compared to the
727 CONFIG_LTO_CLANG_FULL option. More information can be found
728 from Clang's documentation:
730 https://clang.llvm.org/docs/ThinLTO.html
735 config ARCH_SUPPORTS_CFI_CLANG
738 An architecture should select this option if it can support Clang's
739 Control-Flow Integrity (CFI) checking.
742 bool "Use Clang's Control Flow Integrity (CFI)"
743 depends on LTO_CLANG && ARCH_SUPPORTS_CFI_CLANG
744 depends on CLANG_VERSION >= 140000
747 This option enables Clang’s forward-edge Control Flow Integrity
748 (CFI) checking, where the compiler injects a runtime check to each
749 indirect function call to ensure the target is a valid function with
750 the correct static type. This restricts possible call targets and
751 makes it more difficult for an attacker to exploit bugs that allow
752 the modification of stored function pointers. More information can be
753 found from Clang's documentation:
755 https://clang.llvm.org/docs/ControlFlowIntegrity.html
757 config CFI_CLANG_SHADOW
758 bool "Use CFI shadow to speed up cross-module checks"
760 depends on CFI_CLANG && MODULES
762 If you select this option, the kernel builds a fast look-up table of
763 CFI check functions in loaded modules to reduce performance overhead.
767 config CFI_PERMISSIVE
768 bool "Use CFI in permissive mode"
771 When selected, Control Flow Integrity (CFI) violations result in a
772 warning instead of a kernel panic. This option should only be used
773 for finding indirect call type mismatches during development.
777 config HAVE_ARCH_WITHIN_STACK_FRAMES
780 An architecture should select this if it can walk the kernel stack
781 frames to determine if an object is part of either the arguments
782 or local variables (i.e. that it excludes saved return addresses,
783 and similar) by implementing an inline arch_within_stack_frames(),
784 which is used by CONFIG_HARDENED_USERCOPY.
786 config HAVE_CONTEXT_TRACKING_USER
789 Provide kernel/user boundaries probes necessary for subsystems
790 that need it, such as userspace RCU extended quiescent state.
791 Syscalls need to be wrapped inside user_exit()-user_enter(), either
792 optimized behind static key or through the slow path using TIF_NOHZ
793 flag. Exceptions handlers must be wrapped as well. Irqs are already
794 protected inside ct_irq_enter/ct_irq_exit() but preemption or signal
795 handling on irq exit still need to be protected.
797 config HAVE_CONTEXT_TRACKING_USER_OFFSTACK
800 Architecture neither relies on exception_enter()/exception_exit()
801 nor on schedule_user(). Also preempt_schedule_notrace() and
802 preempt_schedule_irq() can't be called in a preemptible section
803 while context tracking is CONTEXT_USER. This feature reflects a sane
804 entry implementation where the following requirements are met on
805 critical entry code, ie: before user_exit() or after user_enter():
807 - Critical entry code isn't preemptible (or better yet:
809 - No use of RCU read side critical sections, unless ct_nmi_enter()
811 - No use of instrumentation, unless instrumentation_begin() got
817 Arch relies on TIF_NOHZ and syscall slow path to implement context
818 tracking calls to user_enter()/user_exit().
820 config HAVE_VIRT_CPU_ACCOUNTING
823 config HAVE_VIRT_CPU_ACCOUNTING_IDLE
826 Architecture has its own way to account idle CPU time and therefore
827 doesn't implement vtime_account_idle().
829 config ARCH_HAS_SCALED_CPUTIME
832 config HAVE_VIRT_CPU_ACCOUNTING_GEN
836 With VIRT_CPU_ACCOUNTING_GEN, cputime_t becomes 64-bit.
837 Before enabling this option, arch code must be audited
838 to ensure there are no races in concurrent read/write of
839 cputime_t. For example, reading/writing 64-bit cputime_t on
840 some 32-bit arches may require multiple accesses, so proper
841 locking is needed to protect against concurrent accesses.
843 config HAVE_IRQ_TIME_ACCOUNTING
846 Archs need to ensure they use a high enough resolution clock to
847 support irq time accounting and then call enable_sched_clock_irqtime().
852 Architectures that select this are able to move page tables at the
853 PUD level. If there are only 3 page table levels, the move effectively
854 happens at the PGD level.
859 Archs that select this are able to move page tables at the PMD level.
861 config HAVE_ARCH_TRANSPARENT_HUGEPAGE
864 config HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD
867 config HAVE_ARCH_HUGE_VMAP
871 # Archs that select this would be capable of PMD-sized vmaps (i.e.,
872 # arch_vmap_pmd_supported() returns true). The VM_ALLOW_HUGE_VMAP flag
873 # must be used to enable allocations to use hugepages.
875 config HAVE_ARCH_HUGE_VMALLOC
876 depends on HAVE_ARCH_HUGE_VMAP
879 config ARCH_WANT_HUGE_PMD_SHARE
882 config HAVE_ARCH_SOFT_DIRTY
885 config HAVE_MOD_ARCH_SPECIFIC
888 The arch uses struct mod_arch_specific to store data. Many arches
889 just need a simple module loader without arch specific data - those
890 should not enable this.
892 config MODULES_USE_ELF_RELA
895 Modules only use ELF RELA relocations. Modules with ELF REL
896 relocations will give an error.
898 config MODULES_USE_ELF_REL
901 Modules only use ELF REL relocations. Modules with ELF RELA
902 relocations will give an error.
904 config ARCH_WANTS_MODULES_DATA_IN_VMALLOC
907 For architectures like powerpc/32 which have constraints on module
908 allocation and need to allocate module data outside of module area.
910 config HAVE_IRQ_EXIT_ON_IRQ_STACK
913 Architecture doesn't only execute the irq handler on the irq stack
914 but also irq_exit(). This way we can process softirqs on this irq
915 stack instead of switching to a new one when we call __do_softirq()
916 in the end of an hardirq.
917 This spares a stack switch and improves cache usage on softirq
920 config HAVE_SOFTIRQ_ON_OWN_STACK
923 Architecture provides a function to run __do_softirq() on a
926 config ALTERNATE_USER_ADDRESS_SPACE
929 Architectures set this when the CPU uses separate address
930 spaces for kernel and user space pointers. In this case, the
931 access_ok() check on a __user pointer is skipped.
933 config PGTABLE_LEVELS
937 config ARCH_HAS_ELF_RANDOMIZE
940 An architecture supports choosing randomized locations for
941 stack, mmap, brk, and ET_DYN. Defined functions:
943 - arch_randomize_brk()
945 config HAVE_ARCH_MMAP_RND_BITS
948 An arch should select this symbol if it supports setting a variable
949 number of bits for use in establishing the base address for mmap
950 allocations, has MMU enabled and provides values for both:
951 - ARCH_MMAP_RND_BITS_MIN
952 - ARCH_MMAP_RND_BITS_MAX
954 config HAVE_EXIT_THREAD
957 An architecture implements exit_thread.
959 config ARCH_MMAP_RND_BITS_MIN
962 config ARCH_MMAP_RND_BITS_MAX
965 config ARCH_MMAP_RND_BITS_DEFAULT
968 config ARCH_MMAP_RND_BITS
969 int "Number of bits to use for ASLR of mmap base address" if EXPERT
970 range ARCH_MMAP_RND_BITS_MIN ARCH_MMAP_RND_BITS_MAX
971 default ARCH_MMAP_RND_BITS_DEFAULT if ARCH_MMAP_RND_BITS_DEFAULT
972 default ARCH_MMAP_RND_BITS_MIN
973 depends on HAVE_ARCH_MMAP_RND_BITS
975 This value can be used to select the number of bits to use to
976 determine the random offset to the base address of vma regions
977 resulting from mmap allocations. This value will be bounded
978 by the architecture's minimum and maximum supported values.
980 This value can be changed after boot using the
981 /proc/sys/vm/mmap_rnd_bits tunable
983 config HAVE_ARCH_MMAP_RND_COMPAT_BITS
986 An arch should select this symbol if it supports running applications
987 in compatibility mode, supports setting a variable number of bits for
988 use in establishing the base address for mmap allocations, has MMU
989 enabled and provides values for both:
990 - ARCH_MMAP_RND_COMPAT_BITS_MIN
991 - ARCH_MMAP_RND_COMPAT_BITS_MAX
993 config ARCH_MMAP_RND_COMPAT_BITS_MIN
996 config ARCH_MMAP_RND_COMPAT_BITS_MAX
999 config ARCH_MMAP_RND_COMPAT_BITS_DEFAULT
1002 config ARCH_MMAP_RND_COMPAT_BITS
1003 int "Number of bits to use for ASLR of mmap base address for compatible applications" if EXPERT
1004 range ARCH_MMAP_RND_COMPAT_BITS_MIN ARCH_MMAP_RND_COMPAT_BITS_MAX
1005 default ARCH_MMAP_RND_COMPAT_BITS_DEFAULT if ARCH_MMAP_RND_COMPAT_BITS_DEFAULT
1006 default ARCH_MMAP_RND_COMPAT_BITS_MIN
1007 depends on HAVE_ARCH_MMAP_RND_COMPAT_BITS
1009 This value can be used to select the number of bits to use to
1010 determine the random offset to the base address of vma regions
1011 resulting from mmap allocations for compatible applications This
1012 value will be bounded by the architecture's minimum and maximum
1015 This value can be changed after boot using the
1016 /proc/sys/vm/mmap_rnd_compat_bits tunable
1018 config HAVE_ARCH_COMPAT_MMAP_BASES
1021 This allows 64bit applications to invoke 32-bit mmap() syscall
1022 and vice-versa 32-bit applications to call 64-bit mmap().
1023 Required for applications doing different bitness syscalls.
1025 config PAGE_SIZE_LESS_THAN_64KB
1027 depends on !ARM64_64K_PAGES
1028 depends on !IA64_PAGE_SIZE_64KB
1029 depends on !PAGE_SIZE_64KB
1030 depends on !PARISC_PAGE_SIZE_64KB
1031 depends on PAGE_SIZE_LESS_THAN_256KB
1033 config PAGE_SIZE_LESS_THAN_256KB
1035 depends on !PAGE_SIZE_256KB
1037 # This allows to use a set of generic functions to determine mmap base
1038 # address by giving priority to top-down scheme only if the process
1039 # is not in legacy mode (compat task, unlimited stack size or
1040 # sysctl_legacy_va_layout).
1041 # Architecture that selects this option can provide its own version of:
1043 config ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT
1046 select ARCH_HAS_ELF_RANDOMIZE
1051 config HAVE_JUMP_LABEL_HACK
1054 config HAVE_NOINSTR_HACK
1057 config HAVE_NOINSTR_VALIDATION
1060 config HAVE_UACCESS_VALIDATION
1064 config HAVE_STACK_VALIDATION
1067 Architecture supports objtool compile-time frame pointer rule
1070 config HAVE_RELIABLE_STACKTRACE
1073 Architecture has either save_stack_trace_tsk_reliable() or
1074 arch_stack_walk_reliable() function which only returns a stack trace
1075 if it can guarantee the trace is reliable.
1077 config HAVE_ARCH_HASH
1081 If this is set, the architecture provides an <asm/hash.h>
1082 file which provides platform-specific implementations of some
1083 functions in <linux/hash.h> or fs/namei.c.
1085 config HAVE_ARCH_NVRAM_OPS
1094 config CLONE_BACKWARDS
1097 Architecture has tls passed as the 4th argument of clone(2),
1100 config CLONE_BACKWARDS2
1103 Architecture has the first two arguments of clone(2) swapped.
1105 config CLONE_BACKWARDS3
1108 Architecture has tls passed as the 3rd argument of clone(2),
1111 config ODD_RT_SIGACTION
1114 Architecture has unusual rt_sigaction(2) arguments
1116 config OLD_SIGSUSPEND
1119 Architecture has old sigsuspend(2) syscall, of one-argument variety
1121 config OLD_SIGSUSPEND3
1124 Even weirder antique ABI - three-argument sigsuspend(2)
1126 config OLD_SIGACTION
1129 Architecture has old sigaction(2) syscall. Nope, not the same
1130 as OLD_SIGSUSPEND | OLD_SIGSUSPEND3 - alpha has sigsuspend(2),
1131 but fairly different variant of sigaction(2), thanks to OSF/1
1134 config COMPAT_OLD_SIGACTION
1137 config COMPAT_32BIT_TIME
1138 bool "Provide system calls for 32-bit time_t"
1139 default !64BIT || COMPAT
1141 This enables 32 bit time_t support in addition to 64 bit time_t support.
1142 This is relevant on all 32-bit architectures, and 64-bit architectures
1143 as part of compat syscall handling.
1145 config ARCH_NO_PREEMPT
1148 config ARCH_EPHEMERAL_INODES
1151 An arch should select this symbol if it doesn't keep track of inode
1152 instances on its own, but instead relies on something else (e.g. the
1153 host kernel for an UML kernel).
1155 config ARCH_SUPPORTS_RT
1158 config CPU_NO_EFFICIENT_FFS
1161 config HAVE_ARCH_VMAP_STACK
1164 An arch should select this symbol if it can support kernel stacks
1165 in vmalloc space. This means:
1167 - vmalloc space must be large enough to hold many kernel stacks.
1168 This may rule out many 32-bit architectures.
1170 - Stacks in vmalloc space need to work reliably. For example, if
1171 vmap page tables are created on demand, either this mechanism
1172 needs to work while the stack points to a virtual address with
1173 unpopulated page tables or arch code (switch_to() and switch_mm(),
1174 most likely) needs to ensure that the stack's page table entries
1175 are populated before running on a possibly unpopulated stack.
1177 - If the stack overflows into a guard page, something reasonable
1178 should happen. The definition of "reasonable" is flexible, but
1179 instantly rebooting without logging anything would be unfriendly.
1183 bool "Use a virtually-mapped stack"
1184 depends on HAVE_ARCH_VMAP_STACK
1185 depends on !KASAN || KASAN_HW_TAGS || KASAN_VMALLOC
1187 Enable this if you want the use virtually-mapped kernel stacks
1188 with guard pages. This causes kernel stack overflows to be
1189 caught immediately rather than causing difficult-to-diagnose
1192 To use this with software KASAN modes, the architecture must support
1193 backing virtual mappings with real shadow memory, and KASAN_VMALLOC
1196 config HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET
1199 An arch should select this symbol if it can support kernel stack
1200 offset randomization with calls to add_random_kstack_offset()
1201 during syscall entry and choose_random_kstack_offset() during
1202 syscall exit. Careful removal of -fstack-protector-strong and
1203 -fstack-protector should also be applied to the entry code and
1204 closely examined, as the artificial stack bump looks like an array
1205 to the compiler, so it will attempt to add canary checks regardless
1206 of the static branch state.
1208 config RANDOMIZE_KSTACK_OFFSET
1209 bool "Support for randomizing kernel stack offset on syscall entry" if EXPERT
1211 depends on HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET
1212 depends on INIT_STACK_NONE || !CC_IS_CLANG || CLANG_VERSION >= 140000
1214 The kernel stack offset can be randomized (after pt_regs) by
1215 roughly 5 bits of entropy, frustrating memory corruption
1216 attacks that depend on stack address determinism or
1217 cross-syscall address exposures.
1219 The feature is controlled via the "randomize_kstack_offset=on/off"
1220 kernel boot param, and if turned off has zero overhead due to its use
1221 of static branches (see JUMP_LABEL).
1225 config RANDOMIZE_KSTACK_OFFSET_DEFAULT
1226 bool "Default state of kernel stack offset randomization"
1227 depends on RANDOMIZE_KSTACK_OFFSET
1229 Kernel stack offset randomization is controlled by kernel boot param
1230 "randomize_kstack_offset=on/off", and this config chooses the default
1233 config ARCH_OPTIONAL_KERNEL_RWX
1236 config ARCH_OPTIONAL_KERNEL_RWX_DEFAULT
1239 config ARCH_HAS_STRICT_KERNEL_RWX
1242 config STRICT_KERNEL_RWX
1243 bool "Make kernel text and rodata read-only" if ARCH_OPTIONAL_KERNEL_RWX
1244 depends on ARCH_HAS_STRICT_KERNEL_RWX
1245 default !ARCH_OPTIONAL_KERNEL_RWX || ARCH_OPTIONAL_KERNEL_RWX_DEFAULT
1247 If this is set, kernel text and rodata memory will be made read-only,
1248 and non-text memory will be made non-executable. This provides
1249 protection against certain security exploits (e.g. executing the heap
1252 These features are considered standard security practice these days.
1253 You should say Y here in almost all cases.
1255 config ARCH_HAS_STRICT_MODULE_RWX
1258 config STRICT_MODULE_RWX
1259 bool "Set loadable kernel module data as NX and text as RO" if ARCH_OPTIONAL_KERNEL_RWX
1260 depends on ARCH_HAS_STRICT_MODULE_RWX && MODULES
1261 default !ARCH_OPTIONAL_KERNEL_RWX || ARCH_OPTIONAL_KERNEL_RWX_DEFAULT
1263 If this is set, module text and rodata memory will be made read-only,
1264 and non-text memory will be made non-executable. This provides
1265 protection against certain security exploits (e.g. writing to text)
1267 # select if the architecture provides an asm/dma-direct.h header
1268 config ARCH_HAS_PHYS_TO_DMA
1271 config HAVE_ARCH_COMPILER_H
1274 An architecture can select this if it provides an
1275 asm/compiler.h header that should be included after
1276 linux/compiler-*.h in order to override macro definitions that those
1277 headers generally provide.
1279 config HAVE_ARCH_PREL32_RELOCATIONS
1282 May be selected by an architecture if it supports place-relative
1283 32-bit relocations, both in the toolchain and in the module loader,
1284 in which case relative references can be used in special sections
1285 for PCI fixup, initcalls etc which are only half the size on 64 bit
1286 architectures, and don't require runtime relocation on relocatable
1289 config ARCH_USE_MEMREMAP_PROT
1292 config LOCK_EVENT_COUNTS
1293 bool "Locking event counts collection"
1296 Enable light-weight counting of various locking related events
1297 in the system with minimal performance impact. This reduces
1298 the chance of application behavior change because of timing
1299 differences. The counts are reported via debugfs.
1301 # Select if the architecture has support for applying RELR relocations.
1302 config ARCH_HAS_RELR
1306 bool "Use RELR relocation packing"
1307 depends on ARCH_HAS_RELR && TOOLS_SUPPORT_RELR
1310 Store the kernel's dynamic relocations in the RELR relocation packing
1311 format. Requires a compatible linker (LLD supports this feature), as
1312 well as compatible NM and OBJCOPY utilities (llvm-nm and llvm-objcopy
1315 config ARCH_HAS_MEM_ENCRYPT
1318 config ARCH_HAS_CC_PLATFORM
1321 config HAVE_SPARSE_SYSCALL_NR
1324 An architecture should select this if its syscall numbering is sparse
1325 to save space. For example, MIPS architecture has a syscall array with
1326 entries at 4000, 5000 and 6000 locations. This option turns on syscall
1327 related optimizations for a given architecture.
1329 config ARCH_HAS_VDSO_DATA
1332 config HAVE_STATIC_CALL
1335 config HAVE_STATIC_CALL_INLINE
1337 depends on HAVE_STATIC_CALL
1340 config HAVE_PREEMPT_DYNAMIC
1343 config HAVE_PREEMPT_DYNAMIC_CALL
1345 depends on HAVE_STATIC_CALL
1346 select HAVE_PREEMPT_DYNAMIC
1348 An architecture should select this if it can handle the preemption
1349 model being selected at boot time using static calls.
1351 Where an architecture selects HAVE_STATIC_CALL_INLINE, any call to a
1352 preemption function will be patched directly.
1354 Where an architecture does not select HAVE_STATIC_CALL_INLINE, any
1355 call to a preemption function will go through a trampoline, and the
1356 trampoline will be patched.
1358 It is strongly advised to support inline static call to avoid any
1361 config HAVE_PREEMPT_DYNAMIC_KEY
1363 depends on HAVE_ARCH_JUMP_LABEL
1364 select HAVE_PREEMPT_DYNAMIC
1366 An architecture should select this if it can handle the preemption
1367 model being selected at boot time using static keys.
1369 Each preemption function will be given an early return based on a
1370 static key. This should have slightly lower overhead than non-inline
1371 static calls, as this effectively inlines each trampoline into the
1372 start of its callee. This may avoid redundant work, and may
1373 integrate better with CFI schemes.
1375 This will have greater overhead than using inline static calls as
1376 the call to the preemption function cannot be entirely elided.
1378 config ARCH_WANT_LD_ORPHAN_WARN
1381 An arch should select this symbol once all linker sections are explicitly
1382 included, size-asserted, or discarded in the linker scripts. This is
1383 important because we never want expected sections to be placed heuristically
1384 by the linker, since the locations of such sections can change between linker
1387 config HAVE_ARCH_PFN_VALID
1390 config ARCH_SUPPORTS_DEBUG_PAGEALLOC
1393 config ARCH_SUPPORTS_PAGE_TABLE_CHECK
1396 config ARCH_SPLIT_ARG64
1399 If a 32-bit architecture requires 64-bit arguments to be split into
1400 pairs of 32-bit arguments, select this option.
1402 config ARCH_HAS_ELFCORE_COMPAT
1405 config ARCH_HAS_PARANOID_L1D_FLUSH
1408 config ARCH_HAVE_TRACE_MMIO_ACCESS
1411 config DYNAMIC_SIGFRAME
1414 # Select, if arch has a named attribute group bound to NUMA device nodes.
1415 config HAVE_ARCH_NODE_DEV_GROUP
1418 source "kernel/gcov/Kconfig"
1420 source "scripts/gcc-plugins/Kconfig"