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
37 # Selected by HOTPLUG_CORE_SYNC_DEAD or HOTPLUG_CORE_SYNC_FULL
38 config HOTPLUG_CORE_SYNC
41 # Basic CPU dead synchronization selected by architecture
42 config HOTPLUG_CORE_SYNC_DEAD
44 select HOTPLUG_CORE_SYNC
46 # Full CPU synchronization with alive state selected by architecture
47 config HOTPLUG_CORE_SYNC_FULL
49 select HOTPLUG_CORE_SYNC_DEAD if HOTPLUG_CPU
50 select HOTPLUG_CORE_SYNC
52 config HOTPLUG_SPLIT_STARTUP
54 select HOTPLUG_CORE_SYNC_FULL
56 config HOTPLUG_PARALLEL
58 select HOTPLUG_SPLIT_STARTUP
66 depends on HAVE_KPROBES
68 select TASKS_RCU if PREEMPTION
70 Kprobes allows you to trap at almost any kernel address and
71 execute a callback function. register_kprobe() establishes
72 a probepoint and specifies the callback. Kprobes is useful
73 for kernel debugging, non-intrusive instrumentation and testing.
77 bool "Optimize very unlikely/likely branches"
78 depends on HAVE_ARCH_JUMP_LABEL
79 select OBJTOOL if HAVE_JUMP_LABEL_HACK
81 This option enables a transparent branch optimization that
82 makes certain almost-always-true or almost-always-false branch
83 conditions even cheaper to execute within the kernel.
85 Certain performance-sensitive kernel code, such as trace points,
86 scheduler functionality, networking code and KVM have such
87 branches and include support for this optimization technique.
89 If it is detected that the compiler has support for "asm goto",
90 the kernel will compile such branches with just a nop
91 instruction. When the condition flag is toggled to true, the
92 nop will be converted to a jump instruction to execute the
93 conditional block of instructions.
95 This technique lowers overhead and stress on the branch prediction
96 of the processor and generally makes the kernel faster. The update
97 of the condition is slower, but those are always very rare.
99 ( On 32-bit x86, the necessary options added to the compiler
100 flags may increase the size of the kernel slightly. )
102 config STATIC_KEYS_SELFTEST
103 bool "Static key selftest"
104 depends on JUMP_LABEL
106 Boot time self-test of the branch patching code.
108 config STATIC_CALL_SELFTEST
109 bool "Static call selftest"
110 depends on HAVE_STATIC_CALL
112 Boot time self-test of the call patching code.
116 depends on KPROBES && HAVE_OPTPROBES
117 select TASKS_RCU if PREEMPTION
119 config KPROBES_ON_FTRACE
121 depends on KPROBES && HAVE_KPROBES_ON_FTRACE
122 depends on DYNAMIC_FTRACE_WITH_REGS
124 If function tracer is enabled and the arch supports full
125 passing of pt_regs to function tracing, then kprobes can
126 optimize on top of function tracing.
130 depends on ARCH_SUPPORTS_UPROBES
132 Uprobes is the user-space counterpart to kprobes: they
133 enable instrumentation applications (such as 'perf probe')
134 to establish unintrusive probes in user-space binaries and
135 libraries, by executing handler functions when the probes
136 are hit by user-space applications.
138 ( These probes come in the form of single-byte breakpoints,
139 managed by the kernel and kept transparent to the probed
142 config HAVE_64BIT_ALIGNED_ACCESS
143 def_bool 64BIT && !HAVE_EFFICIENT_UNALIGNED_ACCESS
145 Some architectures require 64 bit accesses to be 64 bit
146 aligned, which also requires structs containing 64 bit values
147 to be 64 bit aligned too. This includes some 32 bit
148 architectures which can do 64 bit accesses, as well as 64 bit
149 architectures without unaligned access.
151 This symbol should be selected by an architecture if 64 bit
152 accesses are required to be 64 bit aligned in this way even
153 though it is not a 64 bit architecture.
155 See Documentation/core-api/unaligned-memory-access.rst for
156 more information on the topic of unaligned memory accesses.
158 config HAVE_EFFICIENT_UNALIGNED_ACCESS
161 Some architectures are unable to perform unaligned accesses
162 without the use of get_unaligned/put_unaligned. Others are
163 unable to perform such accesses efficiently (e.g. trap on
164 unaligned access and require fixing it up in the exception
167 This symbol should be selected by an architecture if it can
168 perform unaligned accesses efficiently to allow different
169 code paths to be selected for these cases. Some network
170 drivers, for example, could opt to not fix up alignment
171 problems with received packets if doing so would not help
174 See Documentation/core-api/unaligned-memory-access.rst for more
175 information on the topic of unaligned memory accesses.
177 config ARCH_USE_BUILTIN_BSWAP
180 Modern versions of GCC (since 4.4) have builtin functions
181 for handling byte-swapping. Using these, instead of the old
182 inline assembler that the architecture code provides in the
183 __arch_bswapXX() macros, allows the compiler to see what's
184 happening and offers more opportunity for optimisation. In
185 particular, the compiler will be able to combine the byteswap
186 with a nearby load or store and use load-and-swap or
187 store-and-swap instructions if the architecture has them. It
188 should almost *never* result in code which is worse than the
189 hand-coded assembler in <asm/swab.h>. But just in case it
190 does, the use of the builtins is optional.
192 Any architecture with load-and-swap or store-and-swap
193 instructions should set this. And it shouldn't hurt to set it
194 on architectures that don't have such instructions.
198 depends on KPROBES && (HAVE_KRETPROBES || HAVE_RETHOOK)
200 config KRETPROBE_ON_RETHOOK
202 depends on HAVE_RETHOOK
203 depends on KRETPROBES
206 config USER_RETURN_NOTIFIER
208 depends on HAVE_USER_RETURN_NOTIFIER
210 Provide a kernel-internal notification when a cpu is about to
213 config HAVE_IOREMAP_PROT
219 config HAVE_KRETPROBES
222 config HAVE_OPTPROBES
225 config HAVE_KPROBES_ON_FTRACE
228 config ARCH_CORRECT_STACKTRACE_ON_KRETPROBE
231 Since kretprobes modifies return address on the stack, the
232 stacktrace may see the kretprobe trampoline address instead
233 of correct one. If the architecture stacktrace code and
234 unwinder can adjust such entries, select this configuration.
236 config HAVE_FUNCTION_ERROR_INJECTION
242 config HAVE_FUNCTION_DESCRIPTORS
245 config TRACE_IRQFLAGS_SUPPORT
248 config TRACE_IRQFLAGS_NMI_SUPPORT
252 # An arch should select this if it provides all these things:
254 # task_pt_regs() in asm/processor.h or asm/ptrace.h
255 # arch_has_single_step() if there is hardware single-step support
256 # arch_has_block_step() if there is hardware block-step support
257 # asm/syscall.h supplying asm-generic/syscall.h interface
258 # linux/regset.h user_regset interfaces
259 # CORE_DUMP_USE_REGSET #define'd in linux/elf.h
260 # TIF_SYSCALL_TRACE calls ptrace_report_syscall_{entry,exit}
261 # TIF_NOTIFY_RESUME calls resume_user_mode_work()
263 config HAVE_ARCH_TRACEHOOK
266 config HAVE_DMA_CONTIGUOUS
269 config GENERIC_SMP_IDLE_THREAD
272 config GENERIC_IDLE_POLL_SETUP
275 config ARCH_HAS_FORTIFY_SOURCE
278 An architecture should select this when it can successfully
279 build and run with CONFIG_FORTIFY_SOURCE.
282 # Select if the arch provides a historic keepinit alias for the retain_initrd
283 # command line option
285 config ARCH_HAS_KEEPINITRD
288 # Select if arch has all set_memory_ro/rw/x/nx() functions in asm/cacheflush.h
289 config ARCH_HAS_SET_MEMORY
292 # Select if arch has all set_direct_map_invalid/default() functions
293 config ARCH_HAS_SET_DIRECT_MAP
297 # Select if the architecture provides the arch_dma_set_uncached symbol to
298 # either provide an uncached segment alias for a DMA allocation, or
299 # to remap the page tables in place.
301 config ARCH_HAS_DMA_SET_UNCACHED
305 # Select if the architectures provides the arch_dma_clear_uncached symbol
306 # to undo an in-place page table remap for uncached access.
308 config ARCH_HAS_DMA_CLEAR_UNCACHED
311 config ARCH_HAS_CPU_FINALIZE_INIT
314 # Select if arch init_task must go in the __init_task_data section
315 config ARCH_TASK_STRUCT_ON_STACK
318 # Select if arch has its private alloc_task_struct() function
319 config ARCH_TASK_STRUCT_ALLOCATOR
322 config HAVE_ARCH_THREAD_STRUCT_WHITELIST
324 depends on !ARCH_TASK_STRUCT_ALLOCATOR
326 An architecture should select this to provide hardened usercopy
327 knowledge about what region of the thread_struct should be
328 whitelisted for copying to userspace. Normally this is only the
329 FPU registers. Specifically, arch_thread_struct_whitelist()
330 should be implemented. Without this, the entire thread_struct
331 field in task_struct will be left whitelisted.
333 # Select if arch has its private alloc_thread_stack() function
334 config ARCH_THREAD_STACK_ALLOCATOR
337 # Select if arch wants to size task_struct dynamically via arch_task_struct_size:
338 config ARCH_WANTS_DYNAMIC_TASK_STRUCT
341 config ARCH_WANTS_NO_INSTR
344 An architecture should select this if the noinstr macro is being used on
345 functions to denote that the toolchain should avoid instrumenting such
346 functions and is required for correctness.
348 config ARCH_32BIT_OFF_T
352 All new 32-bit architectures should have 64-bit off_t type on
353 userspace side which corresponds to the loff_t kernel type. This
354 is the requirement for modern ABIs. Some existing architectures
355 still support 32-bit off_t. This option is enabled for all such
356 architectures explicitly.
358 # Selected by 64 bit architectures which have a 32 bit f_tinode in struct ustat
359 config ARCH_32BIT_USTAT_F_TINODE
362 config HAVE_ASM_MODVERSIONS
365 This symbol should be selected by an architecture if it provides
366 <asm/asm-prototypes.h> to support the module versioning for symbols
367 exported from assembly code.
369 config HAVE_REGS_AND_STACK_ACCESS_API
372 This symbol should be selected by an architecture if it supports
373 the API needed to access registers and stack entries from pt_regs,
374 declared in asm/ptrace.h
375 For example the kprobes-based event tracer needs this API.
379 depends on HAVE_REGS_AND_STACK_ACCESS_API
381 This symbol should be selected by an architecture if it
382 supports an implementation of restartable sequences.
387 This symbol should be selected by an architecture if it
390 config HAVE_FUNCTION_ARG_ACCESS_API
393 This symbol should be selected by an architecture if it supports
394 the API needed to access function arguments from pt_regs,
395 declared in asm/ptrace.h
397 config HAVE_HW_BREAKPOINT
399 depends on PERF_EVENTS
401 config HAVE_MIXED_BREAKPOINTS_REGS
403 depends on HAVE_HW_BREAKPOINT
405 Depending on the arch implementation of hardware breakpoints,
406 some of them have separate registers for data and instruction
407 breakpoints addresses, others have mixed registers to store
408 them but define the access type in a control register.
409 Select this option if your arch implements breakpoints under the
412 config HAVE_USER_RETURN_NOTIFIER
415 config HAVE_PERF_EVENTS_NMI
418 System hardware can generate an NMI using the perf event
419 subsystem. Also has support for calculating CPU cycle events
420 to determine how many clock cycles in a given period.
422 config HAVE_HARDLOCKUP_DETECTOR_PERF
424 depends on HAVE_PERF_EVENTS_NMI
426 The arch chooses to use the generic perf-NMI-based hardlockup
427 detector. Must define HAVE_PERF_EVENTS_NMI.
429 config HAVE_HARDLOCKUP_DETECTOR_ARCH
432 The arch provides its own hardlockup detector implementation instead
435 It uses the same command line parameters, and sysctl interface,
436 as the generic hardlockup detectors.
438 config HAVE_PERF_REGS
441 Support selective register dumps for perf events. This includes
442 bit-mapping of each registers and a unique architecture id.
444 config HAVE_PERF_USER_STACK_DUMP
447 Support user stack dumps for perf event samples. This needs
448 access to the user stack pointer which is not unified across
451 config HAVE_ARCH_JUMP_LABEL
454 config HAVE_ARCH_JUMP_LABEL_RELATIVE
457 config MMU_GATHER_TABLE_FREE
460 config MMU_GATHER_RCU_TABLE_FREE
462 select MMU_GATHER_TABLE_FREE
464 config MMU_GATHER_PAGE_SIZE
467 config MMU_GATHER_NO_RANGE
469 select MMU_GATHER_MERGE_VMAS
471 config MMU_GATHER_NO_FLUSH_CACHE
474 config MMU_GATHER_MERGE_VMAS
477 config MMU_GATHER_NO_GATHER
479 depends on MMU_GATHER_TABLE_FREE
481 config ARCH_WANT_IRQS_OFF_ACTIVATE_MM
484 Temporary select until all architectures can be converted to have
485 irqs disabled over activate_mm. Architectures that do IPI based TLB
486 shootdowns should enable this.
488 # Use normal mm refcounting for MMU_LAZY_TLB kernel thread references.
489 # MMU_LAZY_TLB_REFCOUNT=n can improve the scalability of context switching
490 # to/from kernel threads when the same mm is running on a lot of CPUs (a large
491 # multi-threaded application), by reducing contention on the mm refcount.
493 # This can be disabled if the architecture ensures no CPUs are using an mm as a
494 # "lazy tlb" beyond its final refcount (i.e., by the time __mmdrop frees the mm
495 # or its kernel page tables). This could be arranged by arch_exit_mmap(), or
496 # final exit(2) TLB flush, for example.
498 # To implement this, an arch *must*:
499 # Ensure the _lazy_tlb variants of mmgrab/mmdrop are used when manipulating
500 # the lazy tlb reference of a kthread's ->active_mm (non-arch code has been
501 # converted already).
502 config MMU_LAZY_TLB_REFCOUNT
504 depends on !MMU_LAZY_TLB_SHOOTDOWN
506 # This option allows MMU_LAZY_TLB_REFCOUNT=n. It ensures no CPUs are using an
507 # mm as a lazy tlb beyond its last reference count, by shooting down these
508 # users before the mm is deallocated. __mmdrop() first IPIs all CPUs that may
509 # be using the mm as a lazy tlb, so that they may switch themselves to using
510 # init_mm for their active mm. mm_cpumask(mm) is used to determine which CPUs
511 # may be using mm as a lazy tlb mm.
513 # To implement this, an arch *must*:
514 # - At the time of the final mmdrop of the mm, ensure mm_cpumask(mm) contains
515 # at least all possible CPUs in which the mm is lazy.
516 # - It must meet the requirements for MMU_LAZY_TLB_REFCOUNT=n (see above).
517 config MMU_LAZY_TLB_SHOOTDOWN
520 config ARCH_HAVE_NMI_SAFE_CMPXCHG
523 config ARCH_HAS_NMI_SAFE_THIS_CPU_OPS
526 config HAVE_ALIGNED_STRUCT_PAGE
529 This makes sure that struct pages are double word aligned and that
530 e.g. the SLUB allocator can perform double word atomic operations
531 on a struct page for better performance. However selecting this
532 might increase the size of a struct page by a word.
534 config HAVE_CMPXCHG_LOCAL
537 config HAVE_CMPXCHG_DOUBLE
540 config ARCH_WEAK_RELEASE_ACQUIRE
543 config ARCH_WANT_IPC_PARSE_VERSION
546 config ARCH_WANT_COMPAT_IPC_PARSE_VERSION
549 config ARCH_WANT_OLD_COMPAT_IPC
550 select ARCH_WANT_COMPAT_IPC_PARSE_VERSION
553 config HAVE_ARCH_SECCOMP
556 An arch should select this symbol to support seccomp mode 1 (the fixed
557 syscall policy), and must provide an overrides for __NR_seccomp_sigreturn,
558 and compat syscalls if the asm-generic/seccomp.h defaults need adjustment:
559 - __NR_seccomp_read_32
560 - __NR_seccomp_write_32
561 - __NR_seccomp_exit_32
562 - __NR_seccomp_sigreturn_32
564 config HAVE_ARCH_SECCOMP_FILTER
566 select HAVE_ARCH_SECCOMP
568 An arch should select this symbol if it provides all of these things:
569 - all the requirements for HAVE_ARCH_SECCOMP
571 - syscall_get_arguments()
573 - syscall_set_return_value()
574 - SIGSYS siginfo_t support
575 - secure_computing is called from a ptrace_event()-safe context
576 - secure_computing return value is checked and a return value of -1
577 results in the system call being skipped immediately.
578 - seccomp syscall wired up
579 - if !HAVE_SPARSE_SYSCALL_NR, have SECCOMP_ARCH_NATIVE,
580 SECCOMP_ARCH_NATIVE_NR, SECCOMP_ARCH_NATIVE_NAME defined. If
581 COMPAT is supported, have the SECCOMP_ARCH_COMPAT* defines too.
584 prompt "Enable seccomp to safely execute untrusted bytecode"
586 depends on HAVE_ARCH_SECCOMP
588 This kernel feature is useful for number crunching applications
589 that may need to handle untrusted bytecode during their
590 execution. By using pipes or other transports made available
591 to the process as file descriptors supporting the read/write
592 syscalls, it's possible to isolate those applications in their
593 own address space using seccomp. Once seccomp is enabled via
594 prctl(PR_SET_SECCOMP) or the seccomp() syscall, it cannot be
595 disabled and the task is only allowed to execute a few safe
596 syscalls defined by each seccomp mode.
600 config SECCOMP_FILTER
602 depends on HAVE_ARCH_SECCOMP_FILTER && SECCOMP && NET
604 Enable tasks to build secure computing environments defined
605 in terms of Berkeley Packet Filter programs which implement
606 task-defined system call filtering polices.
608 See Documentation/userspace-api/seccomp_filter.rst for details.
610 config SECCOMP_CACHE_DEBUG
611 bool "Show seccomp filter cache status in /proc/pid/seccomp_cache"
612 depends on SECCOMP_FILTER && !HAVE_SPARSE_SYSCALL_NR
615 This enables the /proc/pid/seccomp_cache interface to monitor
616 seccomp cache data. The file format is subject to change. Reading
617 the file requires CAP_SYS_ADMIN.
619 This option is for debugging only. Enabling presents the risk that
620 an adversary may be able to infer the seccomp filter logic.
624 config HAVE_ARCH_STACKLEAK
627 An architecture should select this if it has the code which
628 fills the used part of the kernel stack with the STACKLEAK_POISON
629 value before returning from system calls.
631 config HAVE_STACKPROTECTOR
634 An arch should select this symbol if:
635 - it has implemented a stack canary (e.g. __stack_chk_guard)
637 config STACKPROTECTOR
638 bool "Stack Protector buffer overflow detection"
639 depends on HAVE_STACKPROTECTOR
640 depends on $(cc-option,-fstack-protector)
643 This option turns on the "stack-protector" GCC feature. This
644 feature puts, at the beginning of functions, a canary value on
645 the stack just before the return address, and validates
646 the value just before actually returning. Stack based buffer
647 overflows (that need to overwrite this return address) now also
648 overwrite the canary, which gets detected and the attack is then
649 neutralized via a kernel panic.
651 Functions will have the stack-protector canary logic added if they
652 have an 8-byte or larger character array on the stack.
654 This feature requires gcc version 4.2 or above, or a distribution
655 gcc with the feature backported ("-fstack-protector").
657 On an x86 "defconfig" build, this feature adds canary checks to
658 about 3% of all kernel functions, which increases kernel code size
661 config STACKPROTECTOR_STRONG
662 bool "Strong Stack Protector"
663 depends on STACKPROTECTOR
664 depends on $(cc-option,-fstack-protector-strong)
667 Functions will have the stack-protector canary logic added in any
668 of the following conditions:
670 - local variable's address used as part of the right hand side of an
671 assignment or function argument
672 - local variable is an array (or union containing an array),
673 regardless of array type or length
674 - uses register local variables
676 This feature requires gcc version 4.9 or above, or a distribution
677 gcc with the feature backported ("-fstack-protector-strong").
679 On an x86 "defconfig" build, this feature adds canary checks to
680 about 20% of all kernel functions, which increases the kernel code
683 config ARCH_SUPPORTS_SHADOW_CALL_STACK
686 An architecture should select this if it supports the compiler's
687 Shadow Call Stack and implements runtime support for shadow stack
690 config SHADOW_CALL_STACK
691 bool "Shadow Call Stack"
692 depends on ARCH_SUPPORTS_SHADOW_CALL_STACK
693 depends on DYNAMIC_FTRACE_WITH_ARGS || DYNAMIC_FTRACE_WITH_REGS || !FUNCTION_GRAPH_TRACER
695 This option enables the compiler's Shadow Call Stack, which
696 uses a shadow stack to protect function return addresses from
697 being overwritten by an attacker. More information can be found
698 in the compiler's documentation:
700 - Clang: https://clang.llvm.org/docs/ShadowCallStack.html
701 - GCC: https://gcc.gnu.org/onlinedocs/gcc/Instrumentation-Options.html#Instrumentation-Options
703 Note that security guarantees in the kernel differ from the
704 ones documented for user space. The kernel must store addresses
705 of shadow stacks in memory, which means an attacker capable of
706 reading and writing arbitrary memory may be able to locate them
707 and hijack control flow by modifying the stacks.
712 Set by the arch code if it relies on code patching to insert the
713 shadow call stack push and pop instructions rather than on the
719 Selected if the kernel will be built using the compiler's LTO feature.
725 Selected if the kernel will be built using Clang's LTO feature.
727 config ARCH_SUPPORTS_LTO_CLANG
730 An architecture should select this option if it supports:
731 - compiling with Clang,
732 - compiling inline assembly with Clang's integrated assembler,
733 - and linking with LLD.
735 config ARCH_SUPPORTS_LTO_CLANG_THIN
738 An architecture should select this option if it can support Clang's
743 depends on CC_IS_CLANG && LD_IS_LLD && AS_IS_LLVM
744 depends on $(success,$(NM) --help | head -n 1 | grep -qi llvm)
745 depends on $(success,$(AR) --help | head -n 1 | grep -qi llvm)
746 depends on ARCH_SUPPORTS_LTO_CLANG
747 depends on !FTRACE_MCOUNT_USE_RECORDMCOUNT
748 depends on !KASAN || KASAN_HW_TAGS
749 depends on !GCOV_KERNEL
751 The compiler and Kconfig options support building with Clang's
755 prompt "Link Time Optimization (LTO)"
758 This option enables Link Time Optimization (LTO), which allows the
759 compiler to optimize binaries globally.
761 If unsure, select LTO_NONE. Note that LTO is very resource-intensive
762 so it's disabled by default.
767 Build the kernel normally, without Link Time Optimization (LTO).
769 config LTO_CLANG_FULL
770 bool "Clang Full LTO (EXPERIMENTAL)"
771 depends on HAS_LTO_CLANG
772 depends on !COMPILE_TEST
775 This option enables Clang's full Link Time Optimization (LTO), which
776 allows the compiler to optimize the kernel globally. If you enable
777 this option, the compiler generates LLVM bitcode instead of ELF
778 object files, and the actual compilation from bitcode happens at
779 the LTO link step, which may take several minutes depending on the
780 kernel configuration. More information can be found from LLVM's
783 https://llvm.org/docs/LinkTimeOptimization.html
785 During link time, this option can use a large amount of RAM, and
786 may take much longer than the ThinLTO option.
788 config LTO_CLANG_THIN
789 bool "Clang ThinLTO (EXPERIMENTAL)"
790 depends on HAS_LTO_CLANG && ARCH_SUPPORTS_LTO_CLANG_THIN
793 This option enables Clang's ThinLTO, which allows for parallel
794 optimization and faster incremental compiles compared to the
795 CONFIG_LTO_CLANG_FULL option. More information can be found
796 from Clang's documentation:
798 https://clang.llvm.org/docs/ThinLTO.html
803 config ARCH_SUPPORTS_CFI_CLANG
806 An architecture should select this option if it can support Clang's
807 Control-Flow Integrity (CFI) checking.
809 config ARCH_USES_CFI_TRAPS
813 bool "Use Clang's Control Flow Integrity (CFI)"
814 depends on ARCH_SUPPORTS_CFI_CLANG
815 depends on $(cc-option,-fsanitize=kcfi)
817 This option enables Clang’s forward-edge Control Flow Integrity
818 (CFI) checking, where the compiler injects a runtime check to each
819 indirect function call to ensure the target is a valid function with
820 the correct static type. This restricts possible call targets and
821 makes it more difficult for an attacker to exploit bugs that allow
822 the modification of stored function pointers. More information can be
823 found from Clang's documentation:
825 https://clang.llvm.org/docs/ControlFlowIntegrity.html
827 config CFI_PERMISSIVE
828 bool "Use CFI in permissive mode"
831 When selected, Control Flow Integrity (CFI) violations result in a
832 warning instead of a kernel panic. This option should only be used
833 for finding indirect call type mismatches during development.
837 config HAVE_ARCH_WITHIN_STACK_FRAMES
840 An architecture should select this if it can walk the kernel stack
841 frames to determine if an object is part of either the arguments
842 or local variables (i.e. that it excludes saved return addresses,
843 and similar) by implementing an inline arch_within_stack_frames(),
844 which is used by CONFIG_HARDENED_USERCOPY.
846 config HAVE_CONTEXT_TRACKING_USER
849 Provide kernel/user boundaries probes necessary for subsystems
850 that need it, such as userspace RCU extended quiescent state.
851 Syscalls need to be wrapped inside user_exit()-user_enter(), either
852 optimized behind static key or through the slow path using TIF_NOHZ
853 flag. Exceptions handlers must be wrapped as well. Irqs are already
854 protected inside ct_irq_enter/ct_irq_exit() but preemption or signal
855 handling on irq exit still need to be protected.
857 config HAVE_CONTEXT_TRACKING_USER_OFFSTACK
860 Architecture neither relies on exception_enter()/exception_exit()
861 nor on schedule_user(). Also preempt_schedule_notrace() and
862 preempt_schedule_irq() can't be called in a preemptible section
863 while context tracking is CONTEXT_USER. This feature reflects a sane
864 entry implementation where the following requirements are met on
865 critical entry code, ie: before user_exit() or after user_enter():
867 - Critical entry code isn't preemptible (or better yet:
869 - No use of RCU read side critical sections, unless ct_nmi_enter()
871 - No use of instrumentation, unless instrumentation_begin() got
877 Arch relies on TIF_NOHZ and syscall slow path to implement context
878 tracking calls to user_enter()/user_exit().
880 config HAVE_VIRT_CPU_ACCOUNTING
883 config HAVE_VIRT_CPU_ACCOUNTING_IDLE
886 Architecture has its own way to account idle CPU time and therefore
887 doesn't implement vtime_account_idle().
889 config ARCH_HAS_SCALED_CPUTIME
892 config HAVE_VIRT_CPU_ACCOUNTING_GEN
896 With VIRT_CPU_ACCOUNTING_GEN, cputime_t becomes 64-bit.
897 Before enabling this option, arch code must be audited
898 to ensure there are no races in concurrent read/write of
899 cputime_t. For example, reading/writing 64-bit cputime_t on
900 some 32-bit arches may require multiple accesses, so proper
901 locking is needed to protect against concurrent accesses.
903 config HAVE_IRQ_TIME_ACCOUNTING
906 Archs need to ensure they use a high enough resolution clock to
907 support irq time accounting and then call enable_sched_clock_irqtime().
912 Architectures that select this are able to move page tables at the
913 PUD level. If there are only 3 page table levels, the move effectively
914 happens at the PGD level.
919 Archs that select this are able to move page tables at the PMD level.
921 config HAVE_ARCH_TRANSPARENT_HUGEPAGE
924 config HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD
927 config HAVE_ARCH_HUGE_VMAP
931 # Archs that select this would be capable of PMD-sized vmaps (i.e.,
932 # arch_vmap_pmd_supported() returns true). The VM_ALLOW_HUGE_VMAP flag
933 # must be used to enable allocations to use hugepages.
935 config HAVE_ARCH_HUGE_VMALLOC
936 depends on HAVE_ARCH_HUGE_VMAP
939 config ARCH_WANT_HUGE_PMD_SHARE
942 config HAVE_ARCH_SOFT_DIRTY
945 config HAVE_MOD_ARCH_SPECIFIC
948 The arch uses struct mod_arch_specific to store data. Many arches
949 just need a simple module loader without arch specific data - those
950 should not enable this.
952 config MODULES_USE_ELF_RELA
955 Modules only use ELF RELA relocations. Modules with ELF REL
956 relocations will give an error.
958 config MODULES_USE_ELF_REL
961 Modules only use ELF REL relocations. Modules with ELF RELA
962 relocations will give an error.
964 config ARCH_WANTS_MODULES_DATA_IN_VMALLOC
967 For architectures like powerpc/32 which have constraints on module
968 allocation and need to allocate module data outside of module area.
970 config HAVE_IRQ_EXIT_ON_IRQ_STACK
973 Architecture doesn't only execute the irq handler on the irq stack
974 but also irq_exit(). This way we can process softirqs on this irq
975 stack instead of switching to a new one when we call __do_softirq()
976 in the end of an hardirq.
977 This spares a stack switch and improves cache usage on softirq
980 config HAVE_SOFTIRQ_ON_OWN_STACK
983 Architecture provides a function to run __do_softirq() on a
986 config SOFTIRQ_ON_OWN_STACK
987 def_bool HAVE_SOFTIRQ_ON_OWN_STACK && !PREEMPT_RT
989 config ALTERNATE_USER_ADDRESS_SPACE
992 Architectures set this when the CPU uses separate address
993 spaces for kernel and user space pointers. In this case, the
994 access_ok() check on a __user pointer is skipped.
996 config PGTABLE_LEVELS
1000 config ARCH_HAS_ELF_RANDOMIZE
1003 An architecture supports choosing randomized locations for
1004 stack, mmap, brk, and ET_DYN. Defined functions:
1006 - arch_randomize_brk()
1008 config HAVE_ARCH_MMAP_RND_BITS
1011 An arch should select this symbol if it supports setting a variable
1012 number of bits for use in establishing the base address for mmap
1013 allocations, has MMU enabled and provides values for both:
1014 - ARCH_MMAP_RND_BITS_MIN
1015 - ARCH_MMAP_RND_BITS_MAX
1017 config HAVE_EXIT_THREAD
1020 An architecture implements exit_thread.
1022 config ARCH_MMAP_RND_BITS_MIN
1025 config ARCH_MMAP_RND_BITS_MAX
1028 config ARCH_MMAP_RND_BITS_DEFAULT
1031 config ARCH_MMAP_RND_BITS
1032 int "Number of bits to use for ASLR of mmap base address" if EXPERT
1033 range ARCH_MMAP_RND_BITS_MIN ARCH_MMAP_RND_BITS_MAX
1034 default ARCH_MMAP_RND_BITS_DEFAULT if ARCH_MMAP_RND_BITS_DEFAULT
1035 default ARCH_MMAP_RND_BITS_MIN
1036 depends on HAVE_ARCH_MMAP_RND_BITS
1038 This value can be used to select the number of bits to use to
1039 determine the random offset to the base address of vma regions
1040 resulting from mmap allocations. This value will be bounded
1041 by the architecture's minimum and maximum supported values.
1043 This value can be changed after boot using the
1044 /proc/sys/vm/mmap_rnd_bits tunable
1046 config HAVE_ARCH_MMAP_RND_COMPAT_BITS
1049 An arch should select this symbol if it supports running applications
1050 in compatibility mode, supports setting a variable number of bits for
1051 use in establishing the base address for mmap allocations, has MMU
1052 enabled and provides values for both:
1053 - ARCH_MMAP_RND_COMPAT_BITS_MIN
1054 - ARCH_MMAP_RND_COMPAT_BITS_MAX
1056 config ARCH_MMAP_RND_COMPAT_BITS_MIN
1059 config ARCH_MMAP_RND_COMPAT_BITS_MAX
1062 config ARCH_MMAP_RND_COMPAT_BITS_DEFAULT
1065 config ARCH_MMAP_RND_COMPAT_BITS
1066 int "Number of bits to use for ASLR of mmap base address for compatible applications" if EXPERT
1067 range ARCH_MMAP_RND_COMPAT_BITS_MIN ARCH_MMAP_RND_COMPAT_BITS_MAX
1068 default ARCH_MMAP_RND_COMPAT_BITS_DEFAULT if ARCH_MMAP_RND_COMPAT_BITS_DEFAULT
1069 default ARCH_MMAP_RND_COMPAT_BITS_MIN
1070 depends on HAVE_ARCH_MMAP_RND_COMPAT_BITS
1072 This value can be used to select the number of bits to use to
1073 determine the random offset to the base address of vma regions
1074 resulting from mmap allocations for compatible applications This
1075 value will be bounded by the architecture's minimum and maximum
1078 This value can be changed after boot using the
1079 /proc/sys/vm/mmap_rnd_compat_bits tunable
1081 config HAVE_ARCH_COMPAT_MMAP_BASES
1084 This allows 64bit applications to invoke 32-bit mmap() syscall
1085 and vice-versa 32-bit applications to call 64-bit mmap().
1086 Required for applications doing different bitness syscalls.
1088 config PAGE_SIZE_LESS_THAN_64KB
1090 depends on !ARM64_64K_PAGES
1091 depends on !IA64_PAGE_SIZE_64KB
1092 depends on !PAGE_SIZE_64KB
1093 depends on !PARISC_PAGE_SIZE_64KB
1094 depends on PAGE_SIZE_LESS_THAN_256KB
1096 config PAGE_SIZE_LESS_THAN_256KB
1098 depends on !PAGE_SIZE_256KB
1100 # This allows to use a set of generic functions to determine mmap base
1101 # address by giving priority to top-down scheme only if the process
1102 # is not in legacy mode (compat task, unlimited stack size or
1103 # sysctl_legacy_va_layout).
1104 # Architecture that selects this option can provide its own version of:
1106 config ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT
1109 select ARCH_HAS_ELF_RANDOMIZE
1114 config HAVE_JUMP_LABEL_HACK
1117 config HAVE_NOINSTR_HACK
1120 config HAVE_NOINSTR_VALIDATION
1123 config HAVE_UACCESS_VALIDATION
1127 config HAVE_STACK_VALIDATION
1130 Architecture supports objtool compile-time frame pointer rule
1133 config HAVE_RELIABLE_STACKTRACE
1136 Architecture has either save_stack_trace_tsk_reliable() or
1137 arch_stack_walk_reliable() function which only returns a stack trace
1138 if it can guarantee the trace is reliable.
1140 config HAVE_ARCH_HASH
1144 If this is set, the architecture provides an <asm/hash.h>
1145 file which provides platform-specific implementations of some
1146 functions in <linux/hash.h> or fs/namei.c.
1148 config HAVE_ARCH_NVRAM_OPS
1157 config CLONE_BACKWARDS
1160 Architecture has tls passed as the 4th argument of clone(2),
1163 config CLONE_BACKWARDS2
1166 Architecture has the first two arguments of clone(2) swapped.
1168 config CLONE_BACKWARDS3
1171 Architecture has tls passed as the 3rd argument of clone(2),
1174 config ODD_RT_SIGACTION
1177 Architecture has unusual rt_sigaction(2) arguments
1179 config OLD_SIGSUSPEND
1182 Architecture has old sigsuspend(2) syscall, of one-argument variety
1184 config OLD_SIGSUSPEND3
1187 Even weirder antique ABI - three-argument sigsuspend(2)
1189 config OLD_SIGACTION
1192 Architecture has old sigaction(2) syscall. Nope, not the same
1193 as OLD_SIGSUSPEND | OLD_SIGSUSPEND3 - alpha has sigsuspend(2),
1194 but fairly different variant of sigaction(2), thanks to OSF/1
1197 config COMPAT_OLD_SIGACTION
1200 config COMPAT_32BIT_TIME
1201 bool "Provide system calls for 32-bit time_t"
1202 default !64BIT || COMPAT
1204 This enables 32 bit time_t support in addition to 64 bit time_t support.
1205 This is relevant on all 32-bit architectures, and 64-bit architectures
1206 as part of compat syscall handling.
1208 config ARCH_NO_PREEMPT
1211 config ARCH_SUPPORTS_RT
1214 config CPU_NO_EFFICIENT_FFS
1217 config HAVE_ARCH_VMAP_STACK
1220 An arch should select this symbol if it can support kernel stacks
1221 in vmalloc space. This means:
1223 - vmalloc space must be large enough to hold many kernel stacks.
1224 This may rule out many 32-bit architectures.
1226 - Stacks in vmalloc space need to work reliably. For example, if
1227 vmap page tables are created on demand, either this mechanism
1228 needs to work while the stack points to a virtual address with
1229 unpopulated page tables or arch code (switch_to() and switch_mm(),
1230 most likely) needs to ensure that the stack's page table entries
1231 are populated before running on a possibly unpopulated stack.
1233 - If the stack overflows into a guard page, something reasonable
1234 should happen. The definition of "reasonable" is flexible, but
1235 instantly rebooting without logging anything would be unfriendly.
1239 bool "Use a virtually-mapped stack"
1240 depends on HAVE_ARCH_VMAP_STACK
1241 depends on !KASAN || KASAN_HW_TAGS || KASAN_VMALLOC
1243 Enable this if you want the use virtually-mapped kernel stacks
1244 with guard pages. This causes kernel stack overflows to be
1245 caught immediately rather than causing difficult-to-diagnose
1248 To use this with software KASAN modes, the architecture must support
1249 backing virtual mappings with real shadow memory, and KASAN_VMALLOC
1252 config HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET
1255 An arch should select this symbol if it can support kernel stack
1256 offset randomization with calls to add_random_kstack_offset()
1257 during syscall entry and choose_random_kstack_offset() during
1258 syscall exit. Careful removal of -fstack-protector-strong and
1259 -fstack-protector should also be applied to the entry code and
1260 closely examined, as the artificial stack bump looks like an array
1261 to the compiler, so it will attempt to add canary checks regardless
1262 of the static branch state.
1264 config RANDOMIZE_KSTACK_OFFSET
1265 bool "Support for randomizing kernel stack offset on syscall entry" if EXPERT
1267 depends on HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET
1268 depends on INIT_STACK_NONE || !CC_IS_CLANG || CLANG_VERSION >= 140000
1270 The kernel stack offset can be randomized (after pt_regs) by
1271 roughly 5 bits of entropy, frustrating memory corruption
1272 attacks that depend on stack address determinism or
1273 cross-syscall address exposures.
1275 The feature is controlled via the "randomize_kstack_offset=on/off"
1276 kernel boot param, and if turned off has zero overhead due to its use
1277 of static branches (see JUMP_LABEL).
1281 config RANDOMIZE_KSTACK_OFFSET_DEFAULT
1282 bool "Default state of kernel stack offset randomization"
1283 depends on RANDOMIZE_KSTACK_OFFSET
1285 Kernel stack offset randomization is controlled by kernel boot param
1286 "randomize_kstack_offset=on/off", and this config chooses the default
1289 config ARCH_OPTIONAL_KERNEL_RWX
1292 config ARCH_OPTIONAL_KERNEL_RWX_DEFAULT
1295 config ARCH_HAS_STRICT_KERNEL_RWX
1298 config STRICT_KERNEL_RWX
1299 bool "Make kernel text and rodata read-only" if ARCH_OPTIONAL_KERNEL_RWX
1300 depends on ARCH_HAS_STRICT_KERNEL_RWX
1301 default !ARCH_OPTIONAL_KERNEL_RWX || ARCH_OPTIONAL_KERNEL_RWX_DEFAULT
1303 If this is set, kernel text and rodata memory will be made read-only,
1304 and non-text memory will be made non-executable. This provides
1305 protection against certain security exploits (e.g. executing the heap
1308 These features are considered standard security practice these days.
1309 You should say Y here in almost all cases.
1311 config ARCH_HAS_STRICT_MODULE_RWX
1314 config STRICT_MODULE_RWX
1315 bool "Set loadable kernel module data as NX and text as RO" if ARCH_OPTIONAL_KERNEL_RWX
1316 depends on ARCH_HAS_STRICT_MODULE_RWX && MODULES
1317 default !ARCH_OPTIONAL_KERNEL_RWX || ARCH_OPTIONAL_KERNEL_RWX_DEFAULT
1319 If this is set, module text and rodata memory will be made read-only,
1320 and non-text memory will be made non-executable. This provides
1321 protection against certain security exploits (e.g. writing to text)
1323 # select if the architecture provides an asm/dma-direct.h header
1324 config ARCH_HAS_PHYS_TO_DMA
1327 config HAVE_ARCH_COMPILER_H
1330 An architecture can select this if it provides an
1331 asm/compiler.h header that should be included after
1332 linux/compiler-*.h in order to override macro definitions that those
1333 headers generally provide.
1335 config HAVE_ARCH_PREL32_RELOCATIONS
1338 May be selected by an architecture if it supports place-relative
1339 32-bit relocations, both in the toolchain and in the module loader,
1340 in which case relative references can be used in special sections
1341 for PCI fixup, initcalls etc which are only half the size on 64 bit
1342 architectures, and don't require runtime relocation on relocatable
1345 config ARCH_USE_MEMREMAP_PROT
1348 config LOCK_EVENT_COUNTS
1349 bool "Locking event counts collection"
1352 Enable light-weight counting of various locking related events
1353 in the system with minimal performance impact. This reduces
1354 the chance of application behavior change because of timing
1355 differences. The counts are reported via debugfs.
1357 # Select if the architecture has support for applying RELR relocations.
1358 config ARCH_HAS_RELR
1362 bool "Use RELR relocation packing"
1363 depends on ARCH_HAS_RELR && TOOLS_SUPPORT_RELR
1366 Store the kernel's dynamic relocations in the RELR relocation packing
1367 format. Requires a compatible linker (LLD supports this feature), as
1368 well as compatible NM and OBJCOPY utilities (llvm-nm and llvm-objcopy
1371 config ARCH_HAS_MEM_ENCRYPT
1374 config ARCH_HAS_CC_PLATFORM
1377 config HAVE_SPARSE_SYSCALL_NR
1380 An architecture should select this if its syscall numbering is sparse
1381 to save space. For example, MIPS architecture has a syscall array with
1382 entries at 4000, 5000 and 6000 locations. This option turns on syscall
1383 related optimizations for a given architecture.
1385 config ARCH_HAS_VDSO_DATA
1388 config HAVE_STATIC_CALL
1391 config HAVE_STATIC_CALL_INLINE
1393 depends on HAVE_STATIC_CALL
1396 config HAVE_PREEMPT_DYNAMIC
1399 config HAVE_PREEMPT_DYNAMIC_CALL
1401 depends on HAVE_STATIC_CALL
1402 select HAVE_PREEMPT_DYNAMIC
1404 An architecture should select this if it can handle the preemption
1405 model being selected at boot time using static calls.
1407 Where an architecture selects HAVE_STATIC_CALL_INLINE, any call to a
1408 preemption function will be patched directly.
1410 Where an architecture does not select HAVE_STATIC_CALL_INLINE, any
1411 call to a preemption function will go through a trampoline, and the
1412 trampoline will be patched.
1414 It is strongly advised to support inline static call to avoid any
1417 config HAVE_PREEMPT_DYNAMIC_KEY
1419 depends on HAVE_ARCH_JUMP_LABEL
1420 select HAVE_PREEMPT_DYNAMIC
1422 An architecture should select this if it can handle the preemption
1423 model being selected at boot time using static keys.
1425 Each preemption function will be given an early return based on a
1426 static key. This should have slightly lower overhead than non-inline
1427 static calls, as this effectively inlines each trampoline into the
1428 start of its callee. This may avoid redundant work, and may
1429 integrate better with CFI schemes.
1431 This will have greater overhead than using inline static calls as
1432 the call to the preemption function cannot be entirely elided.
1434 config ARCH_WANT_LD_ORPHAN_WARN
1437 An arch should select this symbol once all linker sections are explicitly
1438 included, size-asserted, or discarded in the linker scripts. This is
1439 important because we never want expected sections to be placed heuristically
1440 by the linker, since the locations of such sections can change between linker
1443 config HAVE_ARCH_PFN_VALID
1446 config ARCH_SUPPORTS_DEBUG_PAGEALLOC
1449 config ARCH_SUPPORTS_PAGE_TABLE_CHECK
1452 config ARCH_SPLIT_ARG64
1455 If a 32-bit architecture requires 64-bit arguments to be split into
1456 pairs of 32-bit arguments, select this option.
1458 config ARCH_HAS_ELFCORE_COMPAT
1461 config ARCH_HAS_PARANOID_L1D_FLUSH
1464 config ARCH_HAVE_TRACE_MMIO_ACCESS
1467 config DYNAMIC_SIGFRAME
1470 # Select, if arch has a named attribute group bound to NUMA device nodes.
1471 config HAVE_ARCH_NODE_DEV_GROUP
1474 config ARCH_HAS_NONLEAF_PMD_YOUNG
1477 Architectures that select this option are capable of setting the
1478 accessed bit in non-leaf PMD entries when using them as part of linear
1479 address translations. Page table walkers that clear the accessed bit
1480 may use this capability to reduce their search space.
1482 source "kernel/gcov/Kconfig"
1484 source "scripts/gcc-plugins/Kconfig"
1486 config FUNCTION_ALIGNMENT_4B
1489 config FUNCTION_ALIGNMENT_8B
1492 config FUNCTION_ALIGNMENT_16B
1495 config FUNCTION_ALIGNMENT_32B
1498 config FUNCTION_ALIGNMENT_64B
1501 config FUNCTION_ALIGNMENT
1503 default 64 if FUNCTION_ALIGNMENT_64B
1504 default 32 if FUNCTION_ALIGNMENT_32B
1505 default 16 if FUNCTION_ALIGNMENT_16B
1506 default 8 if FUNCTION_ALIGNMENT_8B
1507 default 4 if FUNCTION_ALIGNMENT_4B