2 # General architecture dependent options
12 tristate "OProfile system profiling"
14 depends on HAVE_OPROFILE
16 select RING_BUFFER_ALLOW_SWAP
18 OProfile is a profiling system capable of profiling the
19 whole system, include the kernel, kernel modules, libraries,
24 config OPROFILE_EVENT_MULTIPLEX
25 bool "OProfile multiplexing support (EXPERIMENTAL)"
27 depends on OPROFILE && X86
29 The number of hardware counters is limited. The multiplexing
30 feature enables OProfile to gather more events than counters
31 are provided by the hardware. This is realized by switching
32 between events at an user specified time interval.
39 config OPROFILE_NMI_TIMER
41 depends on PERF_EVENTS && HAVE_PERF_EVENTS_NMI && !PPC64
46 depends on HAVE_KPROBES
49 Kprobes allows you to trap at almost any kernel address and
50 execute a callback function. register_kprobe() establishes
51 a probepoint and specifies the callback. Kprobes is useful
52 for kernel debugging, non-intrusive instrumentation and testing.
56 bool "Optimize very unlikely/likely branches"
57 depends on HAVE_ARCH_JUMP_LABEL
59 This option enables a transparent branch optimization that
60 makes certain almost-always-true or almost-always-false branch
61 conditions even cheaper to execute within the kernel.
63 Certain performance-sensitive kernel code, such as trace points,
64 scheduler functionality, networking code and KVM have such
65 branches and include support for this optimization technique.
67 If it is detected that the compiler has support for "asm goto",
68 the kernel will compile such branches with just a nop
69 instruction. When the condition flag is toggled to true, the
70 nop will be converted to a jump instruction to execute the
71 conditional block of instructions.
73 This technique lowers overhead and stress on the branch prediction
74 of the processor and generally makes the kernel faster. The update
75 of the condition is slower, but those are always very rare.
77 ( On 32-bit x86, the necessary options added to the compiler
78 flags may increase the size of the kernel slightly. )
80 config STATIC_KEYS_SELFTEST
81 bool "Static key selftest"
84 Boot time self-test of the branch patching code.
88 depends on KPROBES && HAVE_OPTPROBES
91 config KPROBES_ON_FTRACE
93 depends on KPROBES && HAVE_KPROBES_ON_FTRACE
94 depends on DYNAMIC_FTRACE_WITH_REGS
96 If function tracer is enabled and the arch supports full
97 passing of pt_regs to function tracing, then kprobes can
98 optimize on top of function tracing.
102 depends on ARCH_SUPPORTS_UPROBES
104 Uprobes is the user-space counterpart to kprobes: they
105 enable instrumentation applications (such as 'perf probe')
106 to establish unintrusive probes in user-space binaries and
107 libraries, by executing handler functions when the probes
108 are hit by user-space applications.
110 ( These probes come in the form of single-byte breakpoints,
111 managed by the kernel and kept transparent to the probed
114 config HAVE_64BIT_ALIGNED_ACCESS
115 def_bool 64BIT && !HAVE_EFFICIENT_UNALIGNED_ACCESS
117 Some architectures require 64 bit accesses to be 64 bit
118 aligned, which also requires structs containing 64 bit values
119 to be 64 bit aligned too. This includes some 32 bit
120 architectures which can do 64 bit accesses, as well as 64 bit
121 architectures without unaligned access.
123 This symbol should be selected by an architecture if 64 bit
124 accesses are required to be 64 bit aligned in this way even
125 though it is not a 64 bit architecture.
127 See Documentation/unaligned-memory-access.txt for more
128 information on the topic of unaligned memory accesses.
130 config HAVE_EFFICIENT_UNALIGNED_ACCESS
133 Some architectures are unable to perform unaligned accesses
134 without the use of get_unaligned/put_unaligned. Others are
135 unable to perform such accesses efficiently (e.g. trap on
136 unaligned access and require fixing it up in the exception
139 This symbol should be selected by an architecture if it can
140 perform unaligned accesses efficiently to allow different
141 code paths to be selected for these cases. Some network
142 drivers, for example, could opt to not fix up alignment
143 problems with received packets if doing so would not help
146 See Documentation/unaligned-memory-access.txt for more
147 information on the topic of unaligned memory accesses.
149 config ARCH_USE_BUILTIN_BSWAP
152 Modern versions of GCC (since 4.4) have builtin functions
153 for handling byte-swapping. Using these, instead of the old
154 inline assembler that the architecture code provides in the
155 __arch_bswapXX() macros, allows the compiler to see what's
156 happening and offers more opportunity for optimisation. In
157 particular, the compiler will be able to combine the byteswap
158 with a nearby load or store and use load-and-swap or
159 store-and-swap instructions if the architecture has them. It
160 should almost *never* result in code which is worse than the
161 hand-coded assembler in <asm/swab.h>. But just in case it
162 does, the use of the builtins is optional.
164 Any architecture with load-and-swap or store-and-swap
165 instructions should set this. And it shouldn't hurt to set it
166 on architectures that don't have such instructions.
170 depends on KPROBES && HAVE_KRETPROBES
172 config USER_RETURN_NOTIFIER
174 depends on HAVE_USER_RETURN_NOTIFIER
176 Provide a kernel-internal notification when a cpu is about to
179 config HAVE_IOREMAP_PROT
185 config HAVE_KRETPROBES
188 config HAVE_OPTPROBES
191 config HAVE_KPROBES_ON_FTRACE
197 config HAVE_NMI_WATCHDOG
201 # An arch should select this if it provides all these things:
203 # task_pt_regs() in asm/processor.h or asm/ptrace.h
204 # arch_has_single_step() if there is hardware single-step support
205 # arch_has_block_step() if there is hardware block-step support
206 # asm/syscall.h supplying asm-generic/syscall.h interface
207 # linux/regset.h user_regset interfaces
208 # CORE_DUMP_USE_REGSET #define'd in linux/elf.h
209 # TIF_SYSCALL_TRACE calls tracehook_report_syscall_{entry,exit}
210 # TIF_NOTIFY_RESUME calls tracehook_notify_resume()
211 # signal delivery calls tracehook_signal_handler()
213 config HAVE_ARCH_TRACEHOOK
216 config HAVE_DMA_CONTIGUOUS
219 config GENERIC_SMP_IDLE_THREAD
222 config GENERIC_IDLE_POLL_SETUP
225 # Select if arch init_task initializer is different to init/init_task.c
226 config ARCH_INIT_TASK
229 # Select if arch has its private alloc_task_struct() function
230 config ARCH_TASK_STRUCT_ALLOCATOR
233 # Select if arch has its private alloc_thread_stack() function
234 config ARCH_THREAD_STACK_ALLOCATOR
237 # Select if arch wants to size task_struct dynamically via arch_task_struct_size:
238 config ARCH_WANTS_DYNAMIC_TASK_STRUCT
241 config HAVE_REGS_AND_STACK_ACCESS_API
244 This symbol should be selected by an architecure if it supports
245 the API needed to access registers and stack entries from pt_regs,
246 declared in asm/ptrace.h
247 For example the kprobes-based event tracer needs this API.
252 The <linux/clk.h> calls support software clock gating and
253 thus are a key power management tool on many systems.
255 config HAVE_DMA_API_DEBUG
258 config HAVE_HW_BREAKPOINT
260 depends on PERF_EVENTS
262 config HAVE_MIXED_BREAKPOINTS_REGS
264 depends on HAVE_HW_BREAKPOINT
266 Depending on the arch implementation of hardware breakpoints,
267 some of them have separate registers for data and instruction
268 breakpoints addresses, others have mixed registers to store
269 them but define the access type in a control register.
270 Select this option if your arch implements breakpoints under the
273 config HAVE_USER_RETURN_NOTIFIER
276 config HAVE_PERF_EVENTS_NMI
279 System hardware can generate an NMI using the perf event
280 subsystem. Also has support for calculating CPU cycle events
281 to determine how many clock cycles in a given period.
283 config HAVE_PERF_REGS
286 Support selective register dumps for perf events. This includes
287 bit-mapping of each registers and a unique architecture id.
289 config HAVE_PERF_USER_STACK_DUMP
292 Support user stack dumps for perf event samples. This needs
293 access to the user stack pointer which is not unified across
296 config HAVE_ARCH_JUMP_LABEL
299 config HAVE_RCU_TABLE_FREE
302 config ARCH_HAVE_NMI_SAFE_CMPXCHG
305 config HAVE_ALIGNED_STRUCT_PAGE
308 This makes sure that struct pages are double word aligned and that
309 e.g. the SLUB allocator can perform double word atomic operations
310 on a struct page for better performance. However selecting this
311 might increase the size of a struct page by a word.
313 config HAVE_CMPXCHG_LOCAL
316 config HAVE_CMPXCHG_DOUBLE
319 config ARCH_WANT_IPC_PARSE_VERSION
322 config ARCH_WANT_COMPAT_IPC_PARSE_VERSION
325 config ARCH_WANT_OLD_COMPAT_IPC
326 select ARCH_WANT_COMPAT_IPC_PARSE_VERSION
329 config HAVE_ARCH_SECCOMP_FILTER
332 An arch should select this symbol if it provides all of these things:
334 - syscall_get_arguments()
336 - syscall_set_return_value()
337 - SIGSYS siginfo_t support
338 - secure_computing is called from a ptrace_event()-safe context
339 - secure_computing return value is checked and a return value of -1
340 results in the system call being skipped immediately.
341 - seccomp syscall wired up
343 config SECCOMP_FILTER
345 depends on HAVE_ARCH_SECCOMP_FILTER && SECCOMP && NET
347 Enable tasks to build secure computing environments defined
348 in terms of Berkeley Packet Filter programs which implement
349 task-defined system call filtering polices.
351 See Documentation/prctl/seccomp_filter.txt for details.
353 config HAVE_GCC_PLUGINS
356 An arch should select this symbol if it supports building with
359 menuconfig GCC_PLUGINS
361 depends on HAVE_GCC_PLUGINS
362 depends on !COMPILE_TEST
364 GCC plugins are loadable modules that provide extra features to the
365 compiler. They are useful for runtime instrumentation and static analysis.
367 See Documentation/gcc-plugins.txt for details.
369 config GCC_PLUGIN_CYC_COMPLEXITY
370 bool "Compute the cyclomatic complexity of a function" if EXPERT
371 depends on GCC_PLUGINS
372 depends on !COMPILE_TEST
374 The complexity M of a function's control flow graph is defined as:
378 E = the number of edges
379 N = the number of nodes
380 P = the number of connected components (exit nodes).
382 Enabling this plugin reports the complexity to stderr during the
383 build. It mainly serves as a simple example of how to create a
384 gcc plugin for the kernel.
386 config GCC_PLUGIN_SANCOV
388 depends on GCC_PLUGINS
390 This plugin inserts a __sanitizer_cov_trace_pc() call at the start of
391 basic blocks. It supports all gcc versions with plugin support (from
392 gcc-4.5 on). It is based on the commit "Add fuzzing coverage support"
393 by Dmitry Vyukov <dvyukov@google.com>.
395 config GCC_PLUGIN_LATENT_ENTROPY
396 bool "Generate some entropy during boot and runtime"
397 depends on GCC_PLUGINS
399 By saying Y here the kernel will instrument some kernel code to
400 extract some entropy from both original and artificially created
401 program state. This will help especially embedded systems where
402 there is little 'natural' source of entropy normally. The cost
403 is some slowdown of the boot process (about 0.5%) and fork and
406 Note that entropy extracted this way is not cryptographically
409 This plugin was ported from grsecurity/PaX. More information at:
410 * https://grsecurity.net/
411 * https://pax.grsecurity.net/
413 config HAVE_CC_STACKPROTECTOR
416 An arch should select this symbol if:
417 - its compiler supports the -fstack-protector option
418 - it has implemented a stack canary (e.g. __stack_chk_guard)
420 config CC_STACKPROTECTOR
423 Set when a stack-protector mode is enabled, so that the build
424 can enable kernel-side support for the GCC feature.
427 prompt "Stack Protector buffer overflow detection"
428 depends on HAVE_CC_STACKPROTECTOR
429 default CC_STACKPROTECTOR_NONE
431 This option turns on the "stack-protector" GCC feature. This
432 feature puts, at the beginning of functions, a canary value on
433 the stack just before the return address, and validates
434 the value just before actually returning. Stack based buffer
435 overflows (that need to overwrite this return address) now also
436 overwrite the canary, which gets detected and the attack is then
437 neutralized via a kernel panic.
439 config CC_STACKPROTECTOR_NONE
442 Disable "stack-protector" GCC feature.
444 config CC_STACKPROTECTOR_REGULAR
446 select CC_STACKPROTECTOR
448 Functions will have the stack-protector canary logic added if they
449 have an 8-byte or larger character array on the stack.
451 This feature requires gcc version 4.2 or above, or a distribution
452 gcc with the feature backported ("-fstack-protector").
454 On an x86 "defconfig" build, this feature adds canary checks to
455 about 3% of all kernel functions, which increases kernel code size
458 config CC_STACKPROTECTOR_STRONG
460 select CC_STACKPROTECTOR
462 Functions will have the stack-protector canary logic added in any
463 of the following conditions:
465 - local variable's address used as part of the right hand side of an
466 assignment or function argument
467 - local variable is an array (or union containing an array),
468 regardless of array type or length
469 - uses register local variables
471 This feature requires gcc version 4.9 or above, or a distribution
472 gcc with the feature backported ("-fstack-protector-strong").
474 On an x86 "defconfig" build, this feature adds canary checks to
475 about 20% of all kernel functions, which increases the kernel code
483 Select this if the architecture wants to use thin archives
484 instead of ld -r to create the built-in.o files.
486 config LD_DEAD_CODE_DATA_ELIMINATION
489 Select this if the architecture wants to do dead code and
490 data elimination with the linker by compiling with
491 -ffunction-sections -fdata-sections and linking with
494 This requires that the arch annotates or otherwise protects
495 its external entry points from being discarded. Linker scripts
496 must also merge .text.*, .data.*, and .bss.* correctly into
497 output sections. Care must be taken not to pull in unrelated
498 sections (e.g., '.text.init'). Typically '.' in section names
499 is used to distinguish them from label names / C identifiers.
501 config HAVE_ARCH_WITHIN_STACK_FRAMES
504 An architecture should select this if it can walk the kernel stack
505 frames to determine if an object is part of either the arguments
506 or local variables (i.e. that it excludes saved return addresses,
507 and similar) by implementing an inline arch_within_stack_frames(),
508 which is used by CONFIG_HARDENED_USERCOPY.
510 config HAVE_CONTEXT_TRACKING
513 Provide kernel/user boundaries probes necessary for subsystems
514 that need it, such as userspace RCU extended quiescent state.
515 Syscalls need to be wrapped inside user_exit()-user_enter() through
516 the slow path using TIF_NOHZ flag. Exceptions handlers must be
517 wrapped as well. Irqs are already protected inside
518 rcu_irq_enter/rcu_irq_exit() but preemption or signal handling on
519 irq exit still need to be protected.
521 config HAVE_VIRT_CPU_ACCOUNTING
524 config ARCH_HAS_SCALED_CPUTIME
527 config HAVE_VIRT_CPU_ACCOUNTING_GEN
531 With VIRT_CPU_ACCOUNTING_GEN, cputime_t becomes 64-bit.
532 Before enabling this option, arch code must be audited
533 to ensure there are no races in concurrent read/write of
534 cputime_t. For example, reading/writing 64-bit cputime_t on
535 some 32-bit arches may require multiple accesses, so proper
536 locking is needed to protect against concurrent accesses.
539 config HAVE_IRQ_TIME_ACCOUNTING
542 Archs need to ensure they use a high enough resolution clock to
543 support irq time accounting and then call enable_sched_clock_irqtime().
545 config HAVE_ARCH_TRANSPARENT_HUGEPAGE
548 config HAVE_ARCH_HUGE_VMAP
551 config HAVE_ARCH_SOFT_DIRTY
554 config HAVE_MOD_ARCH_SPECIFIC
557 The arch uses struct mod_arch_specific to store data. Many arches
558 just need a simple module loader without arch specific data - those
559 should not enable this.
561 config MODULES_USE_ELF_RELA
564 Modules only use ELF RELA relocations. Modules with ELF REL
565 relocations will give an error.
567 config MODULES_USE_ELF_REL
570 Modules only use ELF REL relocations. Modules with ELF RELA
571 relocations will give an error.
573 config HAVE_UNDERSCORE_SYMBOL_PREFIX
576 Some architectures generate an _ in front of C symbols; things like
577 module loading and assembly files need to know about this.
579 config HAVE_IRQ_EXIT_ON_IRQ_STACK
582 Architecture doesn't only execute the irq handler on the irq stack
583 but also irq_exit(). This way we can process softirqs on this irq
584 stack instead of switching to a new one when we call __do_softirq()
585 in the end of an hardirq.
586 This spares a stack switch and improves cache usage on softirq
589 config PGTABLE_LEVELS
593 config ARCH_HAS_ELF_RANDOMIZE
596 An architecture supports choosing randomized locations for
597 stack, mmap, brk, and ET_DYN. Defined functions:
599 - arch_randomize_brk()
601 config HAVE_ARCH_MMAP_RND_BITS
604 An arch should select this symbol if it supports setting a variable
605 number of bits for use in establishing the base address for mmap
606 allocations, has MMU enabled and provides values for both:
607 - ARCH_MMAP_RND_BITS_MIN
608 - ARCH_MMAP_RND_BITS_MAX
610 config HAVE_EXIT_THREAD
613 An architecture implements exit_thread.
615 config ARCH_MMAP_RND_BITS_MIN
618 config ARCH_MMAP_RND_BITS_MAX
621 config ARCH_MMAP_RND_BITS_DEFAULT
624 config ARCH_MMAP_RND_BITS
625 int "Number of bits to use for ASLR of mmap base address" if EXPERT
626 range ARCH_MMAP_RND_BITS_MIN ARCH_MMAP_RND_BITS_MAX
627 default ARCH_MMAP_RND_BITS_DEFAULT if ARCH_MMAP_RND_BITS_DEFAULT
628 default ARCH_MMAP_RND_BITS_MIN
629 depends on HAVE_ARCH_MMAP_RND_BITS
631 This value can be used to select the number of bits to use to
632 determine the random offset to the base address of vma regions
633 resulting from mmap allocations. This value will be bounded
634 by the architecture's minimum and maximum supported values.
636 This value can be changed after boot using the
637 /proc/sys/vm/mmap_rnd_bits tunable
639 config HAVE_ARCH_MMAP_RND_COMPAT_BITS
642 An arch should select this symbol if it supports running applications
643 in compatibility mode, supports setting a variable number of bits for
644 use in establishing the base address for mmap allocations, has MMU
645 enabled and provides values for both:
646 - ARCH_MMAP_RND_COMPAT_BITS_MIN
647 - ARCH_MMAP_RND_COMPAT_BITS_MAX
649 config ARCH_MMAP_RND_COMPAT_BITS_MIN
652 config ARCH_MMAP_RND_COMPAT_BITS_MAX
655 config ARCH_MMAP_RND_COMPAT_BITS_DEFAULT
658 config ARCH_MMAP_RND_COMPAT_BITS
659 int "Number of bits to use for ASLR of mmap base address for compatible applications" if EXPERT
660 range ARCH_MMAP_RND_COMPAT_BITS_MIN ARCH_MMAP_RND_COMPAT_BITS_MAX
661 default ARCH_MMAP_RND_COMPAT_BITS_DEFAULT if ARCH_MMAP_RND_COMPAT_BITS_DEFAULT
662 default ARCH_MMAP_RND_COMPAT_BITS_MIN
663 depends on HAVE_ARCH_MMAP_RND_COMPAT_BITS
665 This value can be used to select the number of bits to use to
666 determine the random offset to the base address of vma regions
667 resulting from mmap allocations for compatible applications This
668 value will be bounded by the architecture's minimum and maximum
671 This value can be changed after boot using the
672 /proc/sys/vm/mmap_rnd_compat_bits tunable
674 config HAVE_COPY_THREAD_TLS
677 Architecture provides copy_thread_tls to accept tls argument via
678 normal C parameter passing, rather than extracting the syscall
679 argument from pt_regs.
681 config HAVE_STACK_VALIDATION
684 Architecture supports the 'objtool check' host tool command, which
685 performs compile-time stack metadata validation.
687 config HAVE_ARCH_HASH
691 If this is set, the architecture provides an <asm/hash.h>
692 file which provides platform-specific implementations of some
693 functions in <linux/hash.h> or fs/namei.c.
701 config CLONE_BACKWARDS
704 Architecture has tls passed as the 4th argument of clone(2),
707 config CLONE_BACKWARDS2
710 Architecture has the first two arguments of clone(2) swapped.
712 config CLONE_BACKWARDS3
715 Architecture has tls passed as the 3rd argument of clone(2),
718 config ODD_RT_SIGACTION
721 Architecture has unusual rt_sigaction(2) arguments
723 config OLD_SIGSUSPEND
726 Architecture has old sigsuspend(2) syscall, of one-argument variety
728 config OLD_SIGSUSPEND3
731 Even weirder antique ABI - three-argument sigsuspend(2)
736 Architecture has old sigaction(2) syscall. Nope, not the same
737 as OLD_SIGSUSPEND | OLD_SIGSUSPEND3 - alpha has sigsuspend(2),
738 but fairly different variant of sigaction(2), thanks to OSF/1
741 config COMPAT_OLD_SIGACTION
744 config ARCH_NO_COHERENT_DMA_MMAP
747 config CPU_NO_EFFICIENT_FFS
750 config HAVE_ARCH_VMAP_STACK
753 An arch should select this symbol if it can support kernel stacks
754 in vmalloc space. This means:
756 - vmalloc space must be large enough to hold many kernel stacks.
757 This may rule out many 32-bit architectures.
759 - Stacks in vmalloc space need to work reliably. For example, if
760 vmap page tables are created on demand, either this mechanism
761 needs to work while the stack points to a virtual address with
762 unpopulated page tables or arch code (switch_to() and switch_mm(),
763 most likely) needs to ensure that the stack's page table entries
764 are populated before running on a possibly unpopulated stack.
766 - If the stack overflows into a guard page, something reasonable
767 should happen. The definition of "reasonable" is flexible, but
768 instantly rebooting without logging anything would be unfriendly.
772 bool "Use a virtually-mapped stack"
773 depends on HAVE_ARCH_VMAP_STACK && !KASAN
775 Enable this if you want the use virtually-mapped kernel stacks
776 with guard pages. This causes kernel stack overflows to be
777 caught immediately rather than causing difficult-to-diagnose
780 This is presently incompatible with KASAN because KASAN expects
781 the stack to map directly to the KASAN shadow map using a formula
782 that is incorrect if the stack is in vmalloc space.
784 config ARCH_WANT_RELAX_ORDER
787 source "kernel/gcov/Kconfig"