linux-6.15/arch/Kconfig

79 	  a probepoint and specifies the callback.  Kprobes is useful
96 	  If it is detected that the compiler has support for "asm goto",
98 	  instruction. When the condition flag is toggled to true, the
104 	  of the condition is slower, but those are always very rare.
131 	  If function tracer is enabled and the arch supports full
140 	  Uprobes is the user-space counterpart to kprobes: they
161 	  though it is not a 64 bit architecture.
196 	  should almost *never* result in code which is worse than the
198 	  does, the use of the builtins is optional.
218 	  Provide a kernel-internal notification when a cpu is about to
263 #	arch_has_single_step()	if there is hardware single-step support
264 #	arch_has_block_step()	if there is hardware block-step support
332 	  whitelisted for copying to userspace. Normally this is only the
344 	  An architecture should select this if the noinstr macro is being used on
346 	  functions and is required for correctness.
354 	  is the requirement for modern ABIs. Some existing architectures
355 	  still support 32-bit off_t. This option is enabled for all such
448 	  access to the user stack pointer which is not unified across
490 # to/from kernel threads when the same mm is running on a lot of CPUs (a large
508 # users before the mm is deallocated. __mmdrop() first IPIs all CPUs that may
510 # init_mm for their active mm. mm_cpumask(mm) is used to determine which CPUs
515 #   at least all possible CPUs in which the mm is lazy.
584 	  - secure_computing is called from a ptrace_event()-safe context
585 	  - secure_computing return value is checked and a return value of -1
590 	    COMPAT is supported, have the SECCOMP_ARCH_COMPAT* defines too.
597 	  This kernel feature is useful for number crunching applications
602 	  own address space using seccomp. Once seccomp is enabled via
604 	  disabled and the task is only allowed to execute a few safe
625 	  seccomp cache data. The file format is subject to change. Reading
628 	  This option is for debugging only. Enabling presents the risk that
657 	  overwrite the canary, which gets detected and the attack is then
681 	  - local variable is an array (or union containing an array),
723 	  shadow call stack push and pop instructions rather than on the
773 	  If unsure, select LTO_NONE. Note that LTO is very resource-intensive
824 	  an AutoFDO profile is specified in variable
828 	  If no profile is specified, AutoFDO options are
844 	  profiles is specified in variable CLANG_PROPELLER_PROFILE_PREFIX
848 	  If no profile is specified, Propeller options are still passed
870 	  indirect function call to ensure the target is a valid function with
887 	  The option is separate from CONFIG_RUST because it affects the ABI.
888 	  When working with build systems that care about the ABI, it is
889 	  convenient to be able to turn on this flag first, before Rust is
892 	  This option is necessary for using CFI with Rust. If unsure, say N.
922 	  frames to determine if an object is part of either the arguments
925 	  which is used by CONFIG_HARDENED_USERCOPY.
944 	  while context tracking is CT_STATE_USER. This feature reflects a sane
982 	  locking is needed to protect against concurrent accesses.
1063 	  boot, but rather require its initialization late when there is
1099 	  access_ok() check on a __user pointer is skipped.
1220 	  minimize memory consumption and is therefore recommended for low
1222 	  Some software that is written for x86 systems makes incorrect
1229 	  This option is the only supported page size on a few older
1236 	  This option is usually a good compromise between memory
1249 	  16KiB pages.	This option is available only on cnMIPS cores.
1260 	  This is not suitable for general-purpose workloads but the
1272 	  (the default is 64KiB or 4KiB on most architectures).
1296 # is not in legacy mode (compat task, unlimited stack size or
1332 	  if it can guarantee the trace is reliable.
1338 	  If this is set, the architecture provides an <asm/hash.h>
1399 	  This is relevant on all 32-bit architectures, and 64-bit architectures
1428 	    should happen.  The definition of "reasonable" is flexible, but
1469 	  The feature is controlled via the "randomize_kstack_offset=on/off"
1479 	  Kernel stack offset randomization is controlled by kernel boot param
1497 	  If this is set, kernel text and rodata memory will be made read-only,
1513 	  If this is set, module text and rodata memory will be made read-only,
1517 # select if the architecture provides an asm/dma-direct.h header
1582 	  An architecture should select this if its syscall numbering is sparse
1620 	  It is strongly advised to support inline static call to avoid any
1644 	  included, size-asserted, or discarded in the linker scripts. This is
1706 	  relocations preserved. This is used by some architectures to
1744 	# Set if the guaranteed alignment with -fmin-function-alignment is
1745 	# available or extra care is required in the kernel. Clang provides