blob: 1455579791ec02514a5d616a4b6749cbec3185b0 [file] [log] [blame]
Mathieu Desnoyersfb32e032008-02-02 15:10:33 -05001#
2# General architecture dependent options
3#
Mathieu Desnoyers125e5642008-02-02 15:10:36 -05004
5config OPROFILE
Robert Richterb309a292010-02-26 15:01:23 +01006 tristate "OProfile system profiling"
Mathieu Desnoyers125e5642008-02-02 15:10:36 -05007 depends on PROFILING
8 depends on HAVE_OPROFILE
Ingo Molnard69d59f2008-12-12 09:38:57 +01009 select RING_BUFFER
Christian Borntraeger9a5963e2009-09-16 21:56:49 +020010 select RING_BUFFER_ALLOW_SWAP
Mathieu Desnoyers125e5642008-02-02 15:10:36 -050011 help
12 OProfile is a profiling system capable of profiling the
13 whole system, include the kernel, kernel modules, libraries,
14 and applications.
15
16 If unsure, say N.
17
Jason Yeh4d4036e2009-07-08 13:49:38 +020018config OPROFILE_EVENT_MULTIPLEX
19 bool "OProfile multiplexing support (EXPERIMENTAL)"
20 default n
21 depends on OPROFILE && X86
22 help
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.
27
28 If unsure, say N.
29
Mathieu Desnoyers125e5642008-02-02 15:10:36 -050030config HAVE_OPROFILE
Jan Beulich9ba16082008-10-15 22:01:38 -070031 bool
Mathieu Desnoyers125e5642008-02-02 15:10:36 -050032
Robert Richterdcfce4a2011-10-11 17:11:08 +020033config OPROFILE_NMI_TIMER
34 def_bool y
35 depends on PERF_EVENTS && HAVE_PERF_EVENTS_NMI
36
Mathieu Desnoyers125e5642008-02-02 15:10:36 -050037config KPROBES
38 bool "Kprobes"
Masami Hiramatsu05ed1602010-09-13 19:25:41 +090039 depends on MODULES
Mathieu Desnoyers125e5642008-02-02 15:10:36 -050040 depends on HAVE_KPROBES
Masami Hiramatsu05ed1602010-09-13 19:25:41 +090041 select KALLSYMS
Mathieu Desnoyers125e5642008-02-02 15:10:36 -050042 help
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.
47 If in doubt, say "N".
48
Steven Rostedt45f81b12010-10-29 12:33:43 -040049config JUMP_LABEL
Ingo Molnarc5905af2012-02-24 08:31:31 +010050 bool "Optimize very unlikely/likely branches"
Steven Rostedt45f81b12010-10-29 12:33:43 -040051 depends on HAVE_ARCH_JUMP_LABEL
52 help
Ingo Molnarc5905af2012-02-24 08:31:31 +010053 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.
Steven Rostedt45f81b12010-10-29 12:33:43 -040056
Ingo Molnarc5905af2012-02-24 08:31:31 +010057 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.
60
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.
66
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.
70
71 ( On 32-bit x86, the necessary options added to the compiler
72 flags may increase the size of the kernel slightly. )
Steven Rostedt45f81b12010-10-29 12:33:43 -040073
Masami Hiramatsuafd66252010-02-25 08:34:07 -050074config OPTPROBES
Masami Hiramatsu5cc718b2010-03-15 13:00:54 -040075 def_bool y
76 depends on KPROBES && HAVE_OPTPROBES
Masami Hiramatsuafd66252010-02-25 08:34:07 -050077 depends on !PREEMPT
Masami Hiramatsuafd66252010-02-25 08:34:07 -050078
Masami Hiramatsue7dbfe32012-09-28 17:15:20 +090079config KPROBES_ON_FTRACE
80 def_bool y
81 depends on KPROBES && HAVE_KPROBES_ON_FTRACE
82 depends on DYNAMIC_FTRACE_WITH_REGS
83 help
84 If function tracer is enabled and the arch supports full
85 passing of pt_regs to function tracing, then kprobes can
86 optimize on top of function tracing.
87
Srikar Dronamraju2b144492012-02-09 14:56:42 +053088config UPROBES
Ingo Molnar7b2d81d2012-02-17 09:27:41 +010089 bool "Transparent user-space probes (EXPERIMENTAL)"
Srikar Dronamrajuec83db02012-05-08 16:41:26 +053090 depends on UPROBE_EVENT && PERF_EVENTS
Srikar Dronamraju2b144492012-02-09 14:56:42 +053091 default n
Oleg Nesterov22b361d2012-12-17 16:01:39 -080092 select PERCPU_RWSEM
Srikar Dronamraju2b144492012-02-09 14:56:42 +053093 help
Ingo Molnar7b2d81d2012-02-17 09:27:41 +010094 Uprobes is the user-space counterpart to kprobes: they
95 enable instrumentation applications (such as 'perf probe')
96 to establish unintrusive probes in user-space binaries and
97 libraries, by executing handler functions when the probes
98 are hit by user-space applications.
99
100 ( These probes come in the form of single-byte breakpoints,
101 managed by the kernel and kept transparent to the probed
102 application. )
Srikar Dronamraju2b144492012-02-09 14:56:42 +0530103
104 If in doubt, say "N".
105
James Hoganc19fa942012-05-30 11:23:23 +0100106config HAVE_64BIT_ALIGNED_ACCESS
107 def_bool 64BIT && !HAVE_EFFICIENT_UNALIGNED_ACCESS
108 help
109 Some architectures require 64 bit accesses to be 64 bit
110 aligned, which also requires structs containing 64 bit values
111 to be 64 bit aligned too. This includes some 32 bit
112 architectures which can do 64 bit accesses, as well as 64 bit
113 architectures without unaligned access.
114
115 This symbol should be selected by an architecture if 64 bit
116 accesses are required to be 64 bit aligned in this way even
117 though it is not a 64 bit architecture.
118
119 See Documentation/unaligned-memory-access.txt for more
120 information on the topic of unaligned memory accesses.
121
Johannes Berg58340a02008-07-25 01:45:33 -0700122config HAVE_EFFICIENT_UNALIGNED_ACCESS
Jan Beulich9ba16082008-10-15 22:01:38 -0700123 bool
Johannes Berg58340a02008-07-25 01:45:33 -0700124 help
125 Some architectures are unable to perform unaligned accesses
126 without the use of get_unaligned/put_unaligned. Others are
127 unable to perform such accesses efficiently (e.g. trap on
128 unaligned access and require fixing it up in the exception
129 handler.)
130
131 This symbol should be selected by an architecture if it can
132 perform unaligned accesses efficiently to allow different
133 code paths to be selected for these cases. Some network
134 drivers, for example, could opt to not fix up alignment
135 problems with received packets if doing so would not help
136 much.
137
138 See Documentation/unaligned-memory-access.txt for more
139 information on the topic of unaligned memory accesses.
140
David Woodhousecf66bb92012-12-03 16:25:40 +0000141config ARCH_USE_BUILTIN_BSWAP
142 bool
143 help
144 Modern versions of GCC (since 4.4) have builtin functions
145 for handling byte-swapping. Using these, instead of the old
146 inline assembler that the architecture code provides in the
147 __arch_bswapXX() macros, allows the compiler to see what's
148 happening and offers more opportunity for optimisation. In
149 particular, the compiler will be able to combine the byteswap
150 with a nearby load or store and use load-and-swap or
151 store-and-swap instructions if the architecture has them. It
152 should almost *never* result in code which is worse than the
153 hand-coded assembler in <asm/swab.h>. But just in case it
154 does, the use of the builtins is optional.
155
156 Any architecture with load-and-swap or store-and-swap
157 instructions should set this. And it shouldn't hurt to set it
158 on architectures that don't have such instructions.
159
Heiko Carstens1a94bc32009-01-14 14:13:59 +0100160config HAVE_SYSCALL_WRAPPERS
161 bool
162
Ananth N Mavinakayanahalli9edddaa2008-03-04 14:28:37 -0800163config KRETPROBES
164 def_bool y
165 depends on KPROBES && HAVE_KRETPROBES
166
Avi Kivity7c68af62009-09-19 09:40:22 +0300167config USER_RETURN_NOTIFIER
168 bool
169 depends on HAVE_USER_RETURN_NOTIFIER
170 help
171 Provide a kernel-internal notification when a cpu is about to
172 switch to user mode.
173
Rik van Riel28b2ee22008-07-23 21:27:05 -0700174config HAVE_IOREMAP_PROT
Jan Beulich9ba16082008-10-15 22:01:38 -0700175 bool
Rik van Riel28b2ee22008-07-23 21:27:05 -0700176
Mathieu Desnoyers125e5642008-02-02 15:10:36 -0500177config HAVE_KPROBES
Jan Beulich9ba16082008-10-15 22:01:38 -0700178 bool
Ananth N Mavinakayanahalli9edddaa2008-03-04 14:28:37 -0800179
180config HAVE_KRETPROBES
Jan Beulich9ba16082008-10-15 22:01:38 -0700181 bool
Arthur Kepner74bc7ce2008-04-29 01:00:30 -0700182
Masami Hiramatsuafd66252010-02-25 08:34:07 -0500183config HAVE_OPTPROBES
184 bool
Cong Wangd314d742012-03-23 15:01:51 -0700185
Masami Hiramatsue7dbfe32012-09-28 17:15:20 +0900186config HAVE_KPROBES_ON_FTRACE
187 bool
188
Cong Wangd314d742012-03-23 15:01:51 -0700189config HAVE_NMI_WATCHDOG
190 bool
Roland McGrath1f5a4ad2008-07-25 19:45:57 -0700191#
192# An arch should select this if it provides all these things:
193#
194# task_pt_regs() in asm/processor.h or asm/ptrace.h
195# arch_has_single_step() if there is hardware single-step support
196# arch_has_block_step() if there is hardware block-step support
Roland McGrath1f5a4ad2008-07-25 19:45:57 -0700197# asm/syscall.h supplying asm-generic/syscall.h interface
198# linux/regset.h user_regset interfaces
199# CORE_DUMP_USE_REGSET #define'd in linux/elf.h
200# TIF_SYSCALL_TRACE calls tracehook_report_syscall_{entry,exit}
201# TIF_NOTIFY_RESUME calls tracehook_notify_resume()
202# signal delivery calls tracehook_signal_handler()
203#
204config HAVE_ARCH_TRACEHOOK
Jan Beulich9ba16082008-10-15 22:01:38 -0700205 bool
Roland McGrath1f5a4ad2008-07-25 19:45:57 -0700206
Arthur Kepner74bc7ce2008-04-29 01:00:30 -0700207config HAVE_DMA_ATTRS
Jan Beulich9ba16082008-10-15 22:01:38 -0700208 bool
Jens Axboe3d442232008-06-26 11:21:34 +0200209
Marek Szyprowskic64be2b2011-12-29 13:09:51 +0100210config HAVE_DMA_CONTIGUOUS
211 bool
212
Jens Axboe3d442232008-06-26 11:21:34 +0200213config USE_GENERIC_SMP_HELPERS
Jan Beulich9ba16082008-10-15 22:01:38 -0700214 bool
David Brownell9483a572008-07-23 21:26:48 -0700215
Thomas Gleixner29d5e042012-04-20 13:05:45 +0000216config GENERIC_SMP_IDLE_THREAD
217 bool
218
Thomas Gleixnera6359d12012-05-03 09:03:02 +0000219# Select if arch init_task initializer is different to init/init_task.c
220config ARCH_INIT_TASK
Thomas Gleixnera4a2eb42012-05-03 09:02:48 +0000221 bool
222
Thomas Gleixnerf5e10282012-05-05 15:05:48 +0000223# Select if arch has its private alloc_task_struct() function
224config ARCH_TASK_STRUCT_ALLOCATOR
225 bool
226
227# Select if arch has its private alloc_thread_info() function
228config ARCH_THREAD_INFO_ALLOCATOR
229 bool
230
Heiko Carstensf850c30c2010-02-10 17:25:17 +0100231config HAVE_REGS_AND_STACK_ACCESS_API
232 bool
Heiko Carstense01292b2010-02-18 14:25:21 +0100233 help
234 This symbol should be selected by an architecure if it supports
235 the API needed to access registers and stack entries from pt_regs,
236 declared in asm/ptrace.h
237 For example the kprobes-based event tracer needs this API.
Heiko Carstensf850c30c2010-02-10 17:25:17 +0100238
David Brownell9483a572008-07-23 21:26:48 -0700239config HAVE_CLK
Jan Beulich9ba16082008-10-15 22:01:38 -0700240 bool
David Brownell9483a572008-07-23 21:26:48 -0700241 help
242 The <linux/clk.h> calls support software clock gating and
243 thus are a key power management tool on many systems.
244
Joerg Roedel5ee00bd2009-01-09 12:14:24 +0100245config HAVE_DMA_API_DEBUG
246 bool
Heiko Carstens36cd3c92009-04-09 18:48:34 +0200247
K.Prasad62a038d2009-06-01 23:43:33 +0530248config HAVE_HW_BREAKPOINT
249 bool
Frederic Weisbecker99e8c5a2009-12-17 01:33:54 +0100250 depends on PERF_EVENTS
K.Prasad62a038d2009-06-01 23:43:33 +0530251
Frederic Weisbecker01027522010-04-11 18:55:56 +0200252config HAVE_MIXED_BREAKPOINTS_REGS
253 bool
254 depends on HAVE_HW_BREAKPOINT
255 help
256 Depending on the arch implementation of hardware breakpoints,
257 some of them have separate registers for data and instruction
258 breakpoints addresses, others have mixed registers to store
259 them but define the access type in a control register.
260 Select this option if your arch implements breakpoints under the
261 latter fashion.
262
Avi Kivity7c68af62009-09-19 09:40:22 +0300263config HAVE_USER_RETURN_NOTIFIER
264 bool
Ingo Molnara1922ed2009-09-07 08:19:51 +0200265
Frederic Weisbeckerc01d4322010-05-15 22:57:48 +0200266config HAVE_PERF_EVENTS_NMI
267 bool
Frederic Weisbecker23637d42010-05-15 23:15:20 +0200268 help
269 System hardware can generate an NMI using the perf event
270 subsystem. Also has support for calculating CPU cycle events
271 to determine how many clock cycles in a given period.
Frederic Weisbeckerc01d4322010-05-15 22:57:48 +0200272
Jiri Olsac5e63192012-08-07 15:20:36 +0200273config HAVE_PERF_REGS
274 bool
275 help
276 Support selective register dumps for perf events. This includes
277 bit-mapping of each registers and a unique architecture id.
278
Jiri Olsac5ebced2012-08-07 15:20:40 +0200279config HAVE_PERF_USER_STACK_DUMP
280 bool
281 help
282 Support user stack dumps for perf event samples. This needs
283 access to the user stack pointer which is not unified across
284 architectures.
285
Jason Baronbf5438fc2010-09-17 11:09:00 -0400286config HAVE_ARCH_JUMP_LABEL
287 bool
288
Gerald Schaefer335d7af2010-11-22 15:47:36 +0100289config HAVE_ARCH_MUTEX_CPU_RELAX
290 bool
291
Peter Zijlstra26723912011-05-24 17:12:00 -0700292config HAVE_RCU_TABLE_FREE
293 bool
294
Huang Yingdf013ff2011-07-13 13:14:22 +0800295config ARCH_HAVE_NMI_SAFE_CMPXCHG
296 bool
297
Heiko Carstens43570fd2012-01-12 17:17:27 -0800298config HAVE_ALIGNED_STRUCT_PAGE
299 bool
300 help
301 This makes sure that struct pages are double word aligned and that
302 e.g. the SLUB allocator can perform double word atomic operations
303 on a struct page for better performance. However selecting this
304 might increase the size of a struct page by a word.
305
Heiko Carstens41561532012-01-12 17:17:30 -0800306config HAVE_CMPXCHG_LOCAL
307 bool
308
Heiko Carstens25654092012-01-12 17:17:33 -0800309config HAVE_CMPXCHG_DOUBLE
310 bool
311
Will Deaconc1d7e012012-07-30 14:42:46 -0700312config ARCH_WANT_IPC_PARSE_VERSION
313 bool
314
315config ARCH_WANT_COMPAT_IPC_PARSE_VERSION
316 bool
317
Chris Metcalf48b25c42012-03-15 13:13:38 -0400318config ARCH_WANT_OLD_COMPAT_IPC
Will Deaconc1d7e012012-07-30 14:42:46 -0700319 select ARCH_WANT_COMPAT_IPC_PARSE_VERSION
Chris Metcalf48b25c42012-03-15 13:13:38 -0400320 bool
321
Will Drewrye2cfabdf2012-04-12 16:47:57 -0500322config HAVE_ARCH_SECCOMP_FILTER
323 bool
324 help
Will Drewryfb0fadf2012-04-12 16:48:02 -0500325 An arch should select this symbol if it provides all of these things:
Will Drewrybb6ea432012-04-12 16:48:01 -0500326 - syscall_get_arch()
327 - syscall_get_arguments()
328 - syscall_rollback()
329 - syscall_set_return_value()
Will Drewryfb0fadf2012-04-12 16:48:02 -0500330 - SIGSYS siginfo_t support
331 - secure_computing is called from a ptrace_event()-safe context
332 - secure_computing return value is checked and a return value of -1
333 results in the system call being skipped immediately.
Will Drewrye2cfabdf2012-04-12 16:47:57 -0500334
335config SECCOMP_FILTER
336 def_bool y
337 depends on HAVE_ARCH_SECCOMP_FILTER && SECCOMP && NET
338 help
339 Enable tasks to build secure computing environments defined
340 in terms of Berkeley Packet Filter programs which implement
341 task-defined system call filtering polices.
342
343 See Documentation/prctl/seccomp_filter.txt for details.
344
Frederic Weisbecker91d1aa432012-11-27 19:33:25 +0100345config HAVE_CONTEXT_TRACKING
Frederic Weisbecker2b1d5022012-07-11 20:26:30 +0200346 bool
347 help
Frederic Weisbecker91d1aa432012-11-27 19:33:25 +0100348 Provide kernel/user boundaries probes necessary for subsystems
349 that need it, such as userspace RCU extended quiescent state.
350 Syscalls need to be wrapped inside user_exit()-user_enter() through
351 the slow path using TIF_NOHZ flag. Exceptions handlers must be
352 wrapped as well. Irqs are already protected inside
353 rcu_irq_enter/rcu_irq_exit() but preemption or signal handling on
354 irq exit still need to be protected.
Frederic Weisbecker2b1d5022012-07-11 20:26:30 +0200355
Frederic Weisbeckerb9527412012-06-16 15:39:34 +0200356config HAVE_VIRT_CPU_ACCOUNTING
357 bool
358
Frederic Weisbeckerfdf9c352012-09-09 14:56:31 +0200359config HAVE_IRQ_TIME_ACCOUNTING
360 bool
361 help
362 Archs need to ensure they use a high enough resolution clock to
363 support irq time accounting and then call enable_sched_clock_irqtime().
364
Gerald Schaefer15626062012-10-08 16:30:04 -0700365config HAVE_ARCH_TRANSPARENT_HUGEPAGE
366 bool
367
David Howells786d35d2012-09-28 14:31:03 +0930368config HAVE_MOD_ARCH_SPECIFIC
369 bool
370 help
371 The arch uses struct mod_arch_specific to store data. Many arches
372 just need a simple module loader without arch specific data - those
373 should not enable this.
374
375config MODULES_USE_ELF_RELA
376 bool
377 help
378 Modules only use ELF RELA relocations. Modules with ELF REL
379 relocations will give an error.
380
381config MODULES_USE_ELF_REL
382 bool
383 help
384 Modules only use ELF REL relocations. Modules with ELF RELA
385 relocations will give an error.
386
Al Virod2125042012-10-23 13:17:59 -0400387#
388# ABI hall of shame
389#
390config CLONE_BACKWARDS
391 bool
392 help
393 Architecture has tls passed as the 4th argument of clone(2),
394 not the 5th one.
395
396config CLONE_BACKWARDS2
397 bool
398 help
399 Architecture has the first two arguments of clone(2) swapped.
400
Al Viroeaca6ea2012-11-25 23:12:10 -0500401config ODD_RT_SIGACTION
402 bool
403 help
404 Architecture has unusual rt_sigaction(2) arguments
405
Al Viro0a0e8cd2012-12-25 16:04:12 -0500406config OLD_SIGSUSPEND
407 bool
408 help
409 Architecture has old sigsuspend(2) syscall, of one-argument variety
410
411config OLD_SIGSUSPEND3
412 bool
413 help
414 Even weirder antique ABI - three-argument sigsuspend(2)
415
Al Viro495dfbf2012-12-25 19:09:45 -0500416config OLD_SIGACTION
417 bool
418 help
419 Architecture has old sigaction(2) syscall. Nope, not the same
420 as OLD_SIGSUSPEND | OLD_SIGSUSPEND3 - alpha has sigsuspend(2),
421 but fairly different variant of sigaction(2), thanks to OSF/1
422 compatibility...
423
424config COMPAT_OLD_SIGACTION
425 bool
426
Peter Oberparleiter2521f2c2009-06-17 16:28:08 -0700427source "kernel/gcov/Kconfig"