2 # General architecture dependent options
6 tristate "OProfile system profiling"
8 depends on HAVE_OPROFILE
10 select RING_BUFFER_ALLOW_SWAP
12 OProfile is a profiling system capable of profiling the
13 whole system, include the kernel, kernel modules, libraries,
18 config OPROFILE_EVENT_MULTIPLEX
19 bool "OProfile multiplexing support (EXPERIMENTAL)"
21 depends on OPROFILE && X86
23 The number of hardware counters is limited. The multiplexing
24 feature enables OProfile to gather more events than counters
25 are provided by the hardware. This is realized by switching
26 between events at an user specified time interval.
33 config OPROFILE_NMI_TIMER
35 depends on PERF_EVENTS && HAVE_PERF_EVENTS_NMI
40 depends on HAVE_KPROBES
43 Kprobes allows you to trap at almost any kernel address and
44 execute a callback function. register_kprobe() establishes
45 a probepoint and specifies the callback. Kprobes is useful
46 for kernel debugging, non-intrusive instrumentation and testing.
50 bool "Optimize very unlikely/likely branches"
51 depends on HAVE_ARCH_JUMP_LABEL
53 This option enables a transparent branch optimization that
54 makes certain almost-always-true or almost-always-false branch
55 conditions even cheaper to execute within the kernel.
57 Certain performance-sensitive kernel code, such as trace points,
58 scheduler functionality, networking code and KVM have such
59 branches and include support for this optimization technique.
61 If it is detected that the compiler has support for "asm goto",
62 the kernel will compile such branches with just a nop
63 instruction. When the condition flag is toggled to true, the
64 nop will be converted to a jump instruction to execute the
65 conditional block of instructions.
67 This technique lowers overhead and stress on the branch prediction
68 of the processor and generally makes the kernel faster. The update
69 of the condition is slower, but those are always very rare.
71 ( On 32-bit x86, the necessary options added to the compiler
72 flags may increase the size of the kernel slightly. )
76 depends on KPROBES && HAVE_OPTPROBES
80 bool "Transparent user-space probes (EXPERIMENTAL)"
81 depends on UPROBE_EVENT && PERF_EVENTS
85 Uprobes is the user-space counterpart to kprobes: they
86 enable instrumentation applications (such as 'perf probe')
87 to establish unintrusive probes in user-space binaries and
88 libraries, by executing handler functions when the probes
89 are hit by user-space applications.
91 ( These probes come in the form of single-byte breakpoints,
92 managed by the kernel and kept transparent to the probed
97 config HAVE_64BIT_ALIGNED_ACCESS
98 def_bool 64BIT && !HAVE_EFFICIENT_UNALIGNED_ACCESS
100 Some architectures require 64 bit accesses to be 64 bit
101 aligned, which also requires structs containing 64 bit values
102 to be 64 bit aligned too. This includes some 32 bit
103 architectures which can do 64 bit accesses, as well as 64 bit
104 architectures without unaligned access.
106 This symbol should be selected by an architecture if 64 bit
107 accesses are required to be 64 bit aligned in this way even
108 though it is not a 64 bit architecture.
110 See Documentation/unaligned-memory-access.txt for more
111 information on the topic of unaligned memory accesses.
113 config HAVE_EFFICIENT_UNALIGNED_ACCESS
116 Some architectures are unable to perform unaligned accesses
117 without the use of get_unaligned/put_unaligned. Others are
118 unable to perform such accesses efficiently (e.g. trap on
119 unaligned access and require fixing it up in the exception
122 This symbol should be selected by an architecture if it can
123 perform unaligned accesses efficiently to allow different
124 code paths to be selected for these cases. Some network
125 drivers, for example, could opt to not fix up alignment
126 problems with received packets if doing so would not help
129 See Documentation/unaligned-memory-access.txt for more
130 information on the topic of unaligned memory accesses.
132 config ARCH_USE_BUILTIN_BSWAP
135 Modern versions of GCC (since 4.4) have builtin functions
136 for handling byte-swapping. Using these, instead of the old
137 inline assembler that the architecture code provides in the
138 __arch_bswapXX() macros, allows the compiler to see what's
139 happening and offers more opportunity for optimisation. In
140 particular, the compiler will be able to combine the byteswap
141 with a nearby load or store and use load-and-swap or
142 store-and-swap instructions if the architecture has them. It
143 should almost *never* result in code which is worse than the
144 hand-coded assembler in <asm/swab.h>. But just in case it
145 does, the use of the builtins is optional.
147 Any architecture with load-and-swap or store-and-swap
148 instructions should set this. And it shouldn't hurt to set it
149 on architectures that don't have such instructions.
151 config HAVE_SYSCALL_WRAPPERS
156 depends on KPROBES && HAVE_KRETPROBES
158 config USER_RETURN_NOTIFIER
160 depends on HAVE_USER_RETURN_NOTIFIER
162 Provide a kernel-internal notification when a cpu is about to
165 config HAVE_IOREMAP_PROT
171 config HAVE_KRETPROBES
174 config HAVE_OPTPROBES
177 config HAVE_NMI_WATCHDOG
180 # An arch should select this if it provides all these things:
182 # task_pt_regs() in asm/processor.h or asm/ptrace.h
183 # arch_has_single_step() if there is hardware single-step support
184 # arch_has_block_step() if there is hardware block-step support
185 # asm/syscall.h supplying asm-generic/syscall.h interface
186 # linux/regset.h user_regset interfaces
187 # CORE_DUMP_USE_REGSET #define'd in linux/elf.h
188 # TIF_SYSCALL_TRACE calls tracehook_report_syscall_{entry,exit}
189 # TIF_NOTIFY_RESUME calls tracehook_notify_resume()
190 # signal delivery calls tracehook_signal_handler()
192 config HAVE_ARCH_TRACEHOOK
195 config HAVE_DMA_ATTRS
198 config HAVE_DMA_CONTIGUOUS
201 config USE_GENERIC_SMP_HELPERS
204 config GENERIC_SMP_IDLE_THREAD
207 # Select if arch init_task initializer is different to init/init_task.c
208 config ARCH_INIT_TASK
211 # Select if arch has its private alloc_task_struct() function
212 config ARCH_TASK_STRUCT_ALLOCATOR
215 # Select if arch has its private alloc_thread_info() function
216 config ARCH_THREAD_INFO_ALLOCATOR
219 config HAVE_REGS_AND_STACK_ACCESS_API
222 This symbol should be selected by an architecure if it supports
223 the API needed to access registers and stack entries from pt_regs,
224 declared in asm/ptrace.h
225 For example the kprobes-based event tracer needs this API.
230 The <linux/clk.h> calls support software clock gating and
231 thus are a key power management tool on many systems.
233 config HAVE_DMA_API_DEBUG
236 config HAVE_HW_BREAKPOINT
238 depends on PERF_EVENTS
240 config HAVE_MIXED_BREAKPOINTS_REGS
242 depends on HAVE_HW_BREAKPOINT
244 Depending on the arch implementation of hardware breakpoints,
245 some of them have separate registers for data and instruction
246 breakpoints addresses, others have mixed registers to store
247 them but define the access type in a control register.
248 Select this option if your arch implements breakpoints under the
251 config HAVE_USER_RETURN_NOTIFIER
254 config HAVE_PERF_EVENTS_NMI
257 System hardware can generate an NMI using the perf event
258 subsystem. Also has support for calculating CPU cycle events
259 to determine how many clock cycles in a given period.
261 config HAVE_PERF_REGS
264 Support selective register dumps for perf events. This includes
265 bit-mapping of each registers and a unique architecture id.
267 config HAVE_PERF_USER_STACK_DUMP
270 Support user stack dumps for perf event samples. This needs
271 access to the user stack pointer which is not unified across
274 config HAVE_ARCH_JUMP_LABEL
277 config HAVE_ARCH_MUTEX_CPU_RELAX
280 config HAVE_RCU_TABLE_FREE
283 config ARCH_HAVE_NMI_SAFE_CMPXCHG
286 config HAVE_ALIGNED_STRUCT_PAGE
289 This makes sure that struct pages are double word aligned and that
290 e.g. the SLUB allocator can perform double word atomic operations
291 on a struct page for better performance. However selecting this
292 might increase the size of a struct page by a word.
294 config HAVE_CMPXCHG_LOCAL
297 config HAVE_CMPXCHG_DOUBLE
300 config ARCH_WANT_IPC_PARSE_VERSION
303 config ARCH_WANT_COMPAT_IPC_PARSE_VERSION
306 config ARCH_WANT_OLD_COMPAT_IPC
307 select ARCH_WANT_COMPAT_IPC_PARSE_VERSION
310 config HAVE_ARCH_SECCOMP_FILTER
313 An arch should select this symbol if it provides all of these things:
315 - syscall_get_arguments()
317 - syscall_set_return_value()
318 - SIGSYS siginfo_t support
319 - secure_computing is called from a ptrace_event()-safe context
320 - secure_computing return value is checked and a return value of -1
321 results in the system call being skipped immediately.
323 config SECCOMP_FILTER
325 depends on HAVE_ARCH_SECCOMP_FILTER && SECCOMP && NET
327 Enable tasks to build secure computing environments defined
328 in terms of Berkeley Packet Filter programs which implement
329 task-defined system call filtering polices.
331 See Documentation/prctl/seccomp_filter.txt for details.
333 config HAVE_CONTEXT_TRACKING
336 Provide kernel/user boundaries probes necessary for subsystems
337 that need it, such as userspace RCU extended quiescent state.
338 Syscalls need to be wrapped inside user_exit()-user_enter() through
339 the slow path using TIF_NOHZ flag. Exceptions handlers must be
340 wrapped as well. Irqs are already protected inside
341 rcu_irq_enter/rcu_irq_exit() but preemption or signal handling on
342 irq exit still need to be protected.
344 config HAVE_VIRT_CPU_ACCOUNTING
347 config HAVE_IRQ_TIME_ACCOUNTING
350 Archs need to ensure they use a high enough resolution clock to
351 support irq time accounting and then call enable_sched_clock_irqtime().
353 config HAVE_ARCH_TRANSPARENT_HUGEPAGE
356 config HAVE_MOD_ARCH_SPECIFIC
359 The arch uses struct mod_arch_specific to store data. Many arches
360 just need a simple module loader without arch specific data - those
361 should not enable this.
363 config MODULES_USE_ELF_RELA
366 Modules only use ELF RELA relocations. Modules with ELF REL
367 relocations will give an error.
369 config MODULES_USE_ELF_REL
372 Modules only use ELF REL relocations. Modules with ELF RELA
373 relocations will give an error.
375 config GENERIC_SIGALTSTACK
381 config CLONE_BACKWARDS
384 Architecture has tls passed as the 4th argument of clone(2),
387 config CLONE_BACKWARDS2
390 Architecture has the first two arguments of clone(2) swapped.
392 source "kernel/gcov/Kconfig"