]>
Commit | Line | Data |
---|---|---|
b2441318 | 1 | # SPDX-License-Identifier: GPL-2.0 |
fb32e03f MD |
2 | # |
3 | # General architecture dependent options | |
4 | # | |
125e5645 | 5 | |
692f66f2 HB |
6 | config CRASH_CORE |
7 | bool | |
8 | ||
2965faa5 | 9 | config KEXEC_CORE |
692f66f2 | 10 | select CRASH_CORE |
2965faa5 DY |
11 | bool |
12 | ||
467d2782 TJB |
13 | config HAVE_IMA_KEXEC |
14 | bool | |
15 | ||
125e5645 | 16 | config OPROFILE |
b309a294 | 17 | tristate "OProfile system profiling" |
125e5645 MD |
18 | depends on PROFILING |
19 | depends on HAVE_OPROFILE | |
d69d59f4 | 20 | select RING_BUFFER |
9a5963eb | 21 | select RING_BUFFER_ALLOW_SWAP |
125e5645 MD |
22 | help |
23 | OProfile is a profiling system capable of profiling the | |
24 | whole system, include the kernel, kernel modules, libraries, | |
25 | and applications. | |
26 | ||
27 | If unsure, say N. | |
28 | ||
4d4036e0 JY |
29 | config OPROFILE_EVENT_MULTIPLEX |
30 | bool "OProfile multiplexing support (EXPERIMENTAL)" | |
31 | default n | |
32 | depends on OPROFILE && X86 | |
33 | help | |
34 | The number of hardware counters is limited. The multiplexing | |
35 | feature enables OProfile to gather more events than counters | |
36 | are provided by the hardware. This is realized by switching | |
9332ef9d | 37 | between events at a user specified time interval. |
4d4036e0 JY |
38 | |
39 | If unsure, say N. | |
40 | ||
125e5645 | 41 | config HAVE_OPROFILE |
9ba16087 | 42 | bool |
125e5645 | 43 | |
dcfce4a0 RR |
44 | config OPROFILE_NMI_TIMER |
45 | def_bool y | |
af9feebe | 46 | depends on PERF_EVENTS && HAVE_PERF_EVENTS_NMI && !PPC64 |
dcfce4a0 | 47 | |
125e5645 MD |
48 | config KPROBES |
49 | bool "Kprobes" | |
05ed160e | 50 | depends on MODULES |
125e5645 | 51 | depends on HAVE_KPROBES |
05ed160e | 52 | select KALLSYMS |
125e5645 MD |
53 | help |
54 | Kprobes allows you to trap at almost any kernel address and | |
55 | execute a callback function. register_kprobe() establishes | |
56 | a probepoint and specifies the callback. Kprobes is useful | |
57 | for kernel debugging, non-intrusive instrumentation and testing. | |
58 | If in doubt, say "N". | |
59 | ||
45f81b1c | 60 | config JUMP_LABEL |
c5905afb | 61 | bool "Optimize very unlikely/likely branches" |
45f81b1c SR |
62 | depends on HAVE_ARCH_JUMP_LABEL |
63 | help | |
c5905afb IM |
64 | This option enables a transparent branch optimization that |
65 | makes certain almost-always-true or almost-always-false branch | |
66 | conditions even cheaper to execute within the kernel. | |
67 | ||
68 | Certain performance-sensitive kernel code, such as trace points, | |
69 | scheduler functionality, networking code and KVM have such | |
70 | branches and include support for this optimization technique. | |
71 | ||
45f81b1c | 72 | If it is detected that the compiler has support for "asm goto", |
c5905afb IM |
73 | the kernel will compile such branches with just a nop |
74 | instruction. When the condition flag is toggled to true, the | |
75 | nop will be converted to a jump instruction to execute the | |
76 | conditional block of instructions. | |
77 | ||
78 | This technique lowers overhead and stress on the branch prediction | |
79 | of the processor and generally makes the kernel faster. The update | |
80 | of the condition is slower, but those are always very rare. | |
45f81b1c | 81 | |
c5905afb IM |
82 | ( On 32-bit x86, the necessary options added to the compiler |
83 | flags may increase the size of the kernel slightly. ) | |
45f81b1c | 84 | |
1987c947 PZ |
85 | config STATIC_KEYS_SELFTEST |
86 | bool "Static key selftest" | |
87 | depends on JUMP_LABEL | |
88 | help | |
89 | Boot time self-test of the branch patching code. | |
90 | ||
afd66255 | 91 | config OPTPROBES |
5cc718b9 MH |
92 | def_bool y |
93 | depends on KPROBES && HAVE_OPTPROBES | |
a30b85df | 94 | select TASKS_RCU if PREEMPT |
afd66255 | 95 | |
e7dbfe34 MH |
96 | config KPROBES_ON_FTRACE |
97 | def_bool y | |
98 | depends on KPROBES && HAVE_KPROBES_ON_FTRACE | |
99 | depends on DYNAMIC_FTRACE_WITH_REGS | |
100 | help | |
101 | If function tracer is enabled and the arch supports full | |
102 | passing of pt_regs to function tracing, then kprobes can | |
103 | optimize on top of function tracing. | |
104 | ||
2b144498 | 105 | config UPROBES |
09294e31 | 106 | def_bool n |
e8f4aa60 | 107 | depends on ARCH_SUPPORTS_UPROBES |
2b144498 | 108 | help |
7b2d81d4 IM |
109 | Uprobes is the user-space counterpart to kprobes: they |
110 | enable instrumentation applications (such as 'perf probe') | |
111 | to establish unintrusive probes in user-space binaries and | |
112 | libraries, by executing handler functions when the probes | |
113 | are hit by user-space applications. | |
114 | ||
115 | ( These probes come in the form of single-byte breakpoints, | |
116 | managed by the kernel and kept transparent to the probed | |
117 | application. ) | |
2b144498 | 118 | |
c19fa94a JH |
119 | config HAVE_64BIT_ALIGNED_ACCESS |
120 | def_bool 64BIT && !HAVE_EFFICIENT_UNALIGNED_ACCESS | |
121 | help | |
122 | Some architectures require 64 bit accesses to be 64 bit | |
123 | aligned, which also requires structs containing 64 bit values | |
124 | to be 64 bit aligned too. This includes some 32 bit | |
125 | architectures which can do 64 bit accesses, as well as 64 bit | |
126 | architectures without unaligned access. | |
127 | ||
128 | This symbol should be selected by an architecture if 64 bit | |
129 | accesses are required to be 64 bit aligned in this way even | |
130 | though it is not a 64 bit architecture. | |
131 | ||
132 | See Documentation/unaligned-memory-access.txt for more | |
133 | information on the topic of unaligned memory accesses. | |
134 | ||
58340a07 | 135 | config HAVE_EFFICIENT_UNALIGNED_ACCESS |
9ba16087 | 136 | bool |
58340a07 JB |
137 | help |
138 | Some architectures are unable to perform unaligned accesses | |
139 | without the use of get_unaligned/put_unaligned. Others are | |
140 | unable to perform such accesses efficiently (e.g. trap on | |
141 | unaligned access and require fixing it up in the exception | |
142 | handler.) | |
143 | ||
144 | This symbol should be selected by an architecture if it can | |
145 | perform unaligned accesses efficiently to allow different | |
146 | code paths to be selected for these cases. Some network | |
147 | drivers, for example, could opt to not fix up alignment | |
148 | problems with received packets if doing so would not help | |
149 | much. | |
150 | ||
151 | See Documentation/unaligned-memory-access.txt for more | |
152 | information on the topic of unaligned memory accesses. | |
153 | ||
cf66bb93 DW |
154 | config ARCH_USE_BUILTIN_BSWAP |
155 | bool | |
156 | help | |
157 | Modern versions of GCC (since 4.4) have builtin functions | |
158 | for handling byte-swapping. Using these, instead of the old | |
159 | inline assembler that the architecture code provides in the | |
160 | __arch_bswapXX() macros, allows the compiler to see what's | |
161 | happening and offers more opportunity for optimisation. In | |
162 | particular, the compiler will be able to combine the byteswap | |
163 | with a nearby load or store and use load-and-swap or | |
164 | store-and-swap instructions if the architecture has them. It | |
165 | should almost *never* result in code which is worse than the | |
166 | hand-coded assembler in <asm/swab.h>. But just in case it | |
167 | does, the use of the builtins is optional. | |
168 | ||
169 | Any architecture with load-and-swap or store-and-swap | |
170 | instructions should set this. And it shouldn't hurt to set it | |
171 | on architectures that don't have such instructions. | |
172 | ||
9edddaa2 AM |
173 | config KRETPROBES |
174 | def_bool y | |
175 | depends on KPROBES && HAVE_KRETPROBES | |
176 | ||
7c68af6e AK |
177 | config USER_RETURN_NOTIFIER |
178 | bool | |
179 | depends on HAVE_USER_RETURN_NOTIFIER | |
180 | help | |
181 | Provide a kernel-internal notification when a cpu is about to | |
182 | switch to user mode. | |
183 | ||
28b2ee20 | 184 | config HAVE_IOREMAP_PROT |
9ba16087 | 185 | bool |
28b2ee20 | 186 | |
125e5645 | 187 | config HAVE_KPROBES |
9ba16087 | 188 | bool |
9edddaa2 AM |
189 | |
190 | config HAVE_KRETPROBES | |
9ba16087 | 191 | bool |
74bc7cee | 192 | |
afd66255 MH |
193 | config HAVE_OPTPROBES |
194 | bool | |
d314d74c | 195 | |
e7dbfe34 MH |
196 | config HAVE_KPROBES_ON_FTRACE |
197 | bool | |
198 | ||
540adea3 | 199 | config HAVE_FUNCTION_ERROR_INJECTION |
9802d865 JB |
200 | bool |
201 | ||
42a0bb3f PM |
202 | config HAVE_NMI |
203 | bool | |
204 | ||
1f5a4ad9 RM |
205 | # |
206 | # An arch should select this if it provides all these things: | |
207 | # | |
208 | # task_pt_regs() in asm/processor.h or asm/ptrace.h | |
209 | # arch_has_single_step() if there is hardware single-step support | |
210 | # arch_has_block_step() if there is hardware block-step support | |
1f5a4ad9 RM |
211 | # asm/syscall.h supplying asm-generic/syscall.h interface |
212 | # linux/regset.h user_regset interfaces | |
213 | # CORE_DUMP_USE_REGSET #define'd in linux/elf.h | |
214 | # TIF_SYSCALL_TRACE calls tracehook_report_syscall_{entry,exit} | |
215 | # TIF_NOTIFY_RESUME calls tracehook_notify_resume() | |
216 | # signal delivery calls tracehook_signal_handler() | |
217 | # | |
218 | config HAVE_ARCH_TRACEHOOK | |
9ba16087 | 219 | bool |
1f5a4ad9 | 220 | |
c64be2bb MS |
221 | config HAVE_DMA_CONTIGUOUS |
222 | bool | |
223 | ||
29d5e047 TG |
224 | config GENERIC_SMP_IDLE_THREAD |
225 | bool | |
226 | ||
485cf5da KH |
227 | config GENERIC_IDLE_POLL_SETUP |
228 | bool | |
229 | ||
6974f0c4 DM |
230 | config ARCH_HAS_FORTIFY_SOURCE |
231 | bool | |
232 | help | |
233 | An architecture should select this when it can successfully | |
234 | build and run with CONFIG_FORTIFY_SOURCE. | |
235 | ||
d2852a22 DB |
236 | # Select if arch has all set_memory_ro/rw/x/nx() functions in asm/cacheflush.h |
237 | config ARCH_HAS_SET_MEMORY | |
238 | bool | |
239 | ||
0500871f DH |
240 | # Select if arch init_task must go in the __init_task_data section |
241 | config ARCH_TASK_STRUCT_ON_STACK | |
a4a2eb49 TG |
242 | bool |
243 | ||
f5e10287 TG |
244 | # Select if arch has its private alloc_task_struct() function |
245 | config ARCH_TASK_STRUCT_ALLOCATOR | |
246 | bool | |
247 | ||
5905429a KC |
248 | config HAVE_ARCH_THREAD_STRUCT_WHITELIST |
249 | bool | |
250 | depends on !ARCH_TASK_STRUCT_ALLOCATOR | |
251 | help | |
252 | An architecture should select this to provide hardened usercopy | |
253 | knowledge about what region of the thread_struct should be | |
254 | whitelisted for copying to userspace. Normally this is only the | |
255 | FPU registers. Specifically, arch_thread_struct_whitelist() | |
256 | should be implemented. Without this, the entire thread_struct | |
257 | field in task_struct will be left whitelisted. | |
258 | ||
b235beea LT |
259 | # Select if arch has its private alloc_thread_stack() function |
260 | config ARCH_THREAD_STACK_ALLOCATOR | |
f5e10287 TG |
261 | bool |
262 | ||
5aaeb5c0 IM |
263 | # Select if arch wants to size task_struct dynamically via arch_task_struct_size: |
264 | config ARCH_WANTS_DYNAMIC_TASK_STRUCT | |
265 | bool | |
266 | ||
f850c30c HC |
267 | config HAVE_REGS_AND_STACK_ACCESS_API |
268 | bool | |
e01292b1 HC |
269 | help |
270 | This symbol should be selected by an architecure if it supports | |
271 | the API needed to access registers and stack entries from pt_regs, | |
272 | declared in asm/ptrace.h | |
273 | For example the kprobes-based event tracer needs this API. | |
f850c30c | 274 | |
9483a578 | 275 | config HAVE_CLK |
9ba16087 | 276 | bool |
9483a578 DB |
277 | help |
278 | The <linux/clk.h> calls support software clock gating and | |
279 | thus are a key power management tool on many systems. | |
280 | ||
62a038d3 P |
281 | config HAVE_HW_BREAKPOINT |
282 | bool | |
99e8c5a3 | 283 | depends on PERF_EVENTS |
62a038d3 | 284 | |
0102752e FW |
285 | config HAVE_MIXED_BREAKPOINTS_REGS |
286 | bool | |
287 | depends on HAVE_HW_BREAKPOINT | |
288 | help | |
289 | Depending on the arch implementation of hardware breakpoints, | |
290 | some of them have separate registers for data and instruction | |
291 | breakpoints addresses, others have mixed registers to store | |
292 | them but define the access type in a control register. | |
293 | Select this option if your arch implements breakpoints under the | |
294 | latter fashion. | |
295 | ||
7c68af6e AK |
296 | config HAVE_USER_RETURN_NOTIFIER |
297 | bool | |
a1922ed6 | 298 | |
c01d4323 FW |
299 | config HAVE_PERF_EVENTS_NMI |
300 | bool | |
23637d47 FW |
301 | help |
302 | System hardware can generate an NMI using the perf event | |
303 | subsystem. Also has support for calculating CPU cycle events | |
304 | to determine how many clock cycles in a given period. | |
c01d4323 | 305 | |
05a4a952 NP |
306 | config HAVE_HARDLOCKUP_DETECTOR_PERF |
307 | bool | |
308 | depends on HAVE_PERF_EVENTS_NMI | |
309 | help | |
310 | The arch chooses to use the generic perf-NMI-based hardlockup | |
311 | detector. Must define HAVE_PERF_EVENTS_NMI. | |
312 | ||
313 | config HAVE_NMI_WATCHDOG | |
314 | depends on HAVE_NMI | |
315 | bool | |
316 | help | |
317 | The arch provides a low level NMI watchdog. It provides | |
318 | asm/nmi.h, and defines its own arch_touch_nmi_watchdog(). | |
319 | ||
320 | config HAVE_HARDLOCKUP_DETECTOR_ARCH | |
321 | bool | |
322 | select HAVE_NMI_WATCHDOG | |
323 | help | |
324 | The arch chooses to provide its own hardlockup detector, which is | |
325 | a superset of the HAVE_NMI_WATCHDOG. It also conforms to config | |
326 | interfaces and parameters provided by hardlockup detector subsystem. | |
327 | ||
c5e63197 JO |
328 | config HAVE_PERF_REGS |
329 | bool | |
330 | help | |
331 | Support selective register dumps for perf events. This includes | |
332 | bit-mapping of each registers and a unique architecture id. | |
333 | ||
c5ebcedb JO |
334 | config HAVE_PERF_USER_STACK_DUMP |
335 | bool | |
336 | help | |
337 | Support user stack dumps for perf event samples. This needs | |
338 | access to the user stack pointer which is not unified across | |
339 | architectures. | |
340 | ||
bf5438fc JB |
341 | config HAVE_ARCH_JUMP_LABEL |
342 | bool | |
343 | ||
26723911 PZ |
344 | config HAVE_RCU_TABLE_FREE |
345 | bool | |
346 | ||
df013ffb HY |
347 | config ARCH_HAVE_NMI_SAFE_CMPXCHG |
348 | bool | |
349 | ||
43570fd2 HC |
350 | config HAVE_ALIGNED_STRUCT_PAGE |
351 | bool | |
352 | help | |
353 | This makes sure that struct pages are double word aligned and that | |
354 | e.g. the SLUB allocator can perform double word atomic operations | |
355 | on a struct page for better performance. However selecting this | |
356 | might increase the size of a struct page by a word. | |
357 | ||
4156153c HC |
358 | config HAVE_CMPXCHG_LOCAL |
359 | bool | |
360 | ||
2565409f HC |
361 | config HAVE_CMPXCHG_DOUBLE |
362 | bool | |
363 | ||
77e58496 PM |
364 | config ARCH_WEAK_RELEASE_ACQUIRE |
365 | bool | |
366 | ||
c1d7e01d WD |
367 | config ARCH_WANT_IPC_PARSE_VERSION |
368 | bool | |
369 | ||
370 | config ARCH_WANT_COMPAT_IPC_PARSE_VERSION | |
371 | bool | |
372 | ||
48b25c43 | 373 | config ARCH_WANT_OLD_COMPAT_IPC |
c1d7e01d | 374 | select ARCH_WANT_COMPAT_IPC_PARSE_VERSION |
48b25c43 CM |
375 | bool |
376 | ||
e2cfabdf WD |
377 | config HAVE_ARCH_SECCOMP_FILTER |
378 | bool | |
379 | help | |
fb0fadf9 | 380 | An arch should select this symbol if it provides all of these things: |
bb6ea430 WD |
381 | - syscall_get_arch() |
382 | - syscall_get_arguments() | |
383 | - syscall_rollback() | |
384 | - syscall_set_return_value() | |
fb0fadf9 WD |
385 | - SIGSYS siginfo_t support |
386 | - secure_computing is called from a ptrace_event()-safe context | |
387 | - secure_computing return value is checked and a return value of -1 | |
388 | results in the system call being skipped immediately. | |
48dc92b9 | 389 | - seccomp syscall wired up |
e2cfabdf WD |
390 | |
391 | config SECCOMP_FILTER | |
392 | def_bool y | |
393 | depends on HAVE_ARCH_SECCOMP_FILTER && SECCOMP && NET | |
394 | help | |
395 | Enable tasks to build secure computing environments defined | |
396 | in terms of Berkeley Packet Filter programs which implement | |
397 | task-defined system call filtering polices. | |
398 | ||
399 | See Documentation/prctl/seccomp_filter.txt for details. | |
400 | ||
6b90bd4b ER |
401 | config HAVE_GCC_PLUGINS |
402 | bool | |
403 | help | |
404 | An arch should select this symbol if it supports building with | |
405 | GCC plugins. | |
406 | ||
407 | menuconfig GCC_PLUGINS | |
408 | bool "GCC plugins" | |
409 | depends on HAVE_GCC_PLUGINS | |
a519167e | 410 | depends on !COMPILE_TEST |
6b90bd4b ER |
411 | help |
412 | GCC plugins are loadable modules that provide extra features to the | |
413 | compiler. They are useful for runtime instrumentation and static analysis. | |
414 | ||
415 | See Documentation/gcc-plugins.txt for details. | |
416 | ||
0dae776c | 417 | config GCC_PLUGIN_CYC_COMPLEXITY |
215e2aa6 | 418 | bool "Compute the cyclomatic complexity of a function" if EXPERT |
0dae776c | 419 | depends on GCC_PLUGINS |
215e2aa6 | 420 | depends on !COMPILE_TEST |
0dae776c ER |
421 | help |
422 | The complexity M of a function's control flow graph is defined as: | |
423 | M = E - N + 2P | |
424 | where | |
425 | ||
426 | E = the number of edges | |
427 | N = the number of nodes | |
428 | P = the number of connected components (exit nodes). | |
429 | ||
215e2aa6 KC |
430 | Enabling this plugin reports the complexity to stderr during the |
431 | build. It mainly serves as a simple example of how to create a | |
432 | gcc plugin for the kernel. | |
433 | ||
543c37cb ER |
434 | config GCC_PLUGIN_SANCOV |
435 | bool | |
436 | depends on GCC_PLUGINS | |
437 | help | |
438 | This plugin inserts a __sanitizer_cov_trace_pc() call at the start of | |
439 | basic blocks. It supports all gcc versions with plugin support (from | |
440 | gcc-4.5 on). It is based on the commit "Add fuzzing coverage support" | |
441 | by Dmitry Vyukov <dvyukov@google.com>. | |
442 | ||
38addce8 ER |
443 | config GCC_PLUGIN_LATENT_ENTROPY |
444 | bool "Generate some entropy during boot and runtime" | |
445 | depends on GCC_PLUGINS | |
446 | help | |
447 | By saying Y here the kernel will instrument some kernel code to | |
448 | extract some entropy from both original and artificially created | |
449 | program state. This will help especially embedded systems where | |
450 | there is little 'natural' source of entropy normally. The cost | |
451 | is some slowdown of the boot process (about 0.5%) and fork and | |
452 | irq processing. | |
453 | ||
454 | Note that entropy extracted this way is not cryptographically | |
455 | secure! | |
456 | ||
457 | This plugin was ported from grsecurity/PaX. More information at: | |
458 | * https://grsecurity.net/ | |
459 | * https://pax.grsecurity.net/ | |
460 | ||
c61f13ea KC |
461 | config GCC_PLUGIN_STRUCTLEAK |
462 | bool "Force initialization of variables containing userspace addresses" | |
463 | depends on GCC_PLUGINS | |
c9cf87ea DV |
464 | # Currently STRUCTLEAK inserts initialization out of live scope of |
465 | # variables from KASAN point of view. This leads to KASAN false | |
466 | # positive reports. Prohibit this combination for now. | |
467 | depends on !KASAN_EXTRA | |
c61f13ea | 468 | help |
f136e090 | 469 | This plugin zero-initializes any structures containing a |
c61f13ea KC |
470 | __user attribute. This can prevent some classes of information |
471 | exposures. | |
472 | ||
473 | This plugin was ported from grsecurity/PaX. More information at: | |
474 | * https://grsecurity.net/ | |
475 | * https://pax.grsecurity.net/ | |
476 | ||
f7dd2507 AB |
477 | config GCC_PLUGIN_STRUCTLEAK_BYREF_ALL |
478 | bool "Force initialize all struct type variables passed by reference" | |
479 | depends on GCC_PLUGIN_STRUCTLEAK | |
480 | help | |
481 | Zero initialize any struct type local variable that may be passed by | |
482 | reference without having been initialized. | |
483 | ||
c61f13ea KC |
484 | config GCC_PLUGIN_STRUCTLEAK_VERBOSE |
485 | bool "Report forcefully initialized variables" | |
486 | depends on GCC_PLUGIN_STRUCTLEAK | |
487 | depends on !COMPILE_TEST | |
488 | help | |
489 | This option will cause a warning to be printed each time the | |
490 | structleak plugin finds a variable it thinks needs to be | |
491 | initialized. Since not all existing initializers are detected | |
492 | by the plugin, this can produce false positive warnings. | |
493 | ||
313dd1b6 KC |
494 | config GCC_PLUGIN_RANDSTRUCT |
495 | bool "Randomize layout of sensitive kernel structures" | |
496 | depends on GCC_PLUGINS | |
497 | select MODVERSIONS if MODULES | |
498 | help | |
9225331b KC |
499 | If you say Y here, the layouts of structures that are entirely |
500 | function pointers (and have not been manually annotated with | |
501 | __no_randomize_layout), or structures that have been explicitly | |
502 | marked with __randomize_layout, will be randomized at compile-time. | |
503 | This can introduce the requirement of an additional information | |
504 | exposure vulnerability for exploits targeting these structure | |
505 | types. | |
313dd1b6 KC |
506 | |
507 | Enabling this feature will introduce some performance impact, | |
508 | slightly increase memory usage, and prevent the use of forensic | |
509 | tools like Volatility against the system (unless the kernel | |
510 | source tree isn't cleaned after kernel installation). | |
511 | ||
512 | The seed used for compilation is located at | |
513 | scripts/gcc-plgins/randomize_layout_seed.h. It remains after | |
514 | a make clean to allow for external modules to be compiled with | |
515 | the existing seed and will be removed by a make mrproper or | |
516 | make distclean. | |
517 | ||
518 | Note that the implementation requires gcc 4.7 or newer. | |
519 | ||
520 | This plugin was ported from grsecurity/PaX. More information at: | |
521 | * https://grsecurity.net/ | |
522 | * https://pax.grsecurity.net/ | |
523 | ||
524 | config GCC_PLUGIN_RANDSTRUCT_PERFORMANCE | |
525 | bool "Use cacheline-aware structure randomization" | |
526 | depends on GCC_PLUGIN_RANDSTRUCT | |
527 | depends on !COMPILE_TEST | |
528 | help | |
529 | If you say Y here, the RANDSTRUCT randomization will make a | |
530 | best effort at restricting randomization to cacheline-sized | |
531 | groups of elements. It will further not randomize bitfields | |
532 | in structures. This reduces the performance hit of RANDSTRUCT | |
533 | at the cost of weakened randomization. | |
534 | ||
19952a92 KC |
535 | config HAVE_CC_STACKPROTECTOR |
536 | bool | |
537 | help | |
538 | An arch should select this symbol if: | |
539 | - its compiler supports the -fstack-protector option | |
540 | - it has implemented a stack canary (e.g. __stack_chk_guard) | |
541 | ||
8779657d KC |
542 | choice |
543 | prompt "Stack Protector buffer overflow detection" | |
19952a92 | 544 | depends on HAVE_CC_STACKPROTECTOR |
44c6dc94 | 545 | default CC_STACKPROTECTOR_AUTO |
19952a92 | 546 | help |
8779657d | 547 | This option turns on the "stack-protector" GCC feature. This |
19952a92 KC |
548 | feature puts, at the beginning of functions, a canary value on |
549 | the stack just before the return address, and validates | |
550 | the value just before actually returning. Stack based buffer | |
551 | overflows (that need to overwrite this return address) now also | |
552 | overwrite the canary, which gets detected and the attack is then | |
553 | neutralized via a kernel panic. | |
554 | ||
8779657d KC |
555 | config CC_STACKPROTECTOR_NONE |
556 | bool "None" | |
557 | help | |
558 | Disable "stack-protector" GCC feature. | |
559 | ||
560 | config CC_STACKPROTECTOR_REGULAR | |
561 | bool "Regular" | |
8779657d KC |
562 | help |
563 | Functions will have the stack-protector canary logic added if they | |
564 | have an 8-byte or larger character array on the stack. | |
565 | ||
19952a92 | 566 | This feature requires gcc version 4.2 or above, or a distribution |
8779657d KC |
567 | gcc with the feature backported ("-fstack-protector"). |
568 | ||
569 | On an x86 "defconfig" build, this feature adds canary checks to | |
570 | about 3% of all kernel functions, which increases kernel code size | |
571 | by about 0.3%. | |
572 | ||
573 | config CC_STACKPROTECTOR_STRONG | |
574 | bool "Strong" | |
8779657d KC |
575 | help |
576 | Functions will have the stack-protector canary logic added in any | |
577 | of the following conditions: | |
578 | ||
579 | - local variable's address used as part of the right hand side of an | |
580 | assignment or function argument | |
581 | - local variable is an array (or union containing an array), | |
582 | regardless of array type or length | |
583 | - uses register local variables | |
584 | ||
585 | This feature requires gcc version 4.9 or above, or a distribution | |
586 | gcc with the feature backported ("-fstack-protector-strong"). | |
587 | ||
588 | On an x86 "defconfig" build, this feature adds canary checks to | |
589 | about 20% of all kernel functions, which increases the kernel code | |
590 | size by about 2%. | |
591 | ||
44c6dc94 KC |
592 | config CC_STACKPROTECTOR_AUTO |
593 | bool "Automatic" | |
594 | help | |
595 | If the compiler supports it, the best available stack-protector | |
596 | option will be chosen. | |
597 | ||
8779657d | 598 | endchoice |
19952a92 | 599 | |
b67067f1 NP |
600 | config LD_DEAD_CODE_DATA_ELIMINATION |
601 | bool | |
602 | help | |
603 | Select this if the architecture wants to do dead code and | |
604 | data elimination with the linker by compiling with | |
605 | -ffunction-sections -fdata-sections and linking with | |
606 | --gc-sections. | |
607 | ||
608 | This requires that the arch annotates or otherwise protects | |
609 | its external entry points from being discarded. Linker scripts | |
610 | must also merge .text.*, .data.*, and .bss.* correctly into | |
0f4c4af0 NP |
611 | output sections. Care must be taken not to pull in unrelated |
612 | sections (e.g., '.text.init'). Typically '.' in section names | |
613 | is used to distinguish them from label names / C identifiers. | |
b67067f1 | 614 | |
0f60a8ef KC |
615 | config HAVE_ARCH_WITHIN_STACK_FRAMES |
616 | bool | |
617 | help | |
618 | An architecture should select this if it can walk the kernel stack | |
619 | frames to determine if an object is part of either the arguments | |
620 | or local variables (i.e. that it excludes saved return addresses, | |
621 | and similar) by implementing an inline arch_within_stack_frames(), | |
622 | which is used by CONFIG_HARDENED_USERCOPY. | |
623 | ||
91d1aa43 | 624 | config HAVE_CONTEXT_TRACKING |
2b1d5024 FW |
625 | bool |
626 | help | |
91d1aa43 FW |
627 | Provide kernel/user boundaries probes necessary for subsystems |
628 | that need it, such as userspace RCU extended quiescent state. | |
629 | Syscalls need to be wrapped inside user_exit()-user_enter() through | |
630 | the slow path using TIF_NOHZ flag. Exceptions handlers must be | |
631 | wrapped as well. Irqs are already protected inside | |
632 | rcu_irq_enter/rcu_irq_exit() but preemption or signal handling on | |
633 | irq exit still need to be protected. | |
2b1d5024 | 634 | |
b952741c FW |
635 | config HAVE_VIRT_CPU_ACCOUNTING |
636 | bool | |
637 | ||
40565b5a SG |
638 | config ARCH_HAS_SCALED_CPUTIME |
639 | bool | |
640 | ||
554b0004 KH |
641 | config HAVE_VIRT_CPU_ACCOUNTING_GEN |
642 | bool | |
643 | default y if 64BIT | |
644 | help | |
645 | With VIRT_CPU_ACCOUNTING_GEN, cputime_t becomes 64-bit. | |
646 | Before enabling this option, arch code must be audited | |
647 | to ensure there are no races in concurrent read/write of | |
648 | cputime_t. For example, reading/writing 64-bit cputime_t on | |
649 | some 32-bit arches may require multiple accesses, so proper | |
650 | locking is needed to protect against concurrent accesses. | |
651 | ||
652 | ||
fdf9c356 FW |
653 | config HAVE_IRQ_TIME_ACCOUNTING |
654 | bool | |
655 | help | |
656 | Archs need to ensure they use a high enough resolution clock to | |
657 | support irq time accounting and then call enable_sched_clock_irqtime(). | |
658 | ||
15626062 GS |
659 | config HAVE_ARCH_TRANSPARENT_HUGEPAGE |
660 | bool | |
661 | ||
a00cc7d9 MW |
662 | config HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD |
663 | bool | |
664 | ||
0ddab1d2 TK |
665 | config HAVE_ARCH_HUGE_VMAP |
666 | bool | |
667 | ||
0f8975ec PE |
668 | config HAVE_ARCH_SOFT_DIRTY |
669 | bool | |
670 | ||
786d35d4 DH |
671 | config HAVE_MOD_ARCH_SPECIFIC |
672 | bool | |
673 | help | |
674 | The arch uses struct mod_arch_specific to store data. Many arches | |
675 | just need a simple module loader without arch specific data - those | |
676 | should not enable this. | |
677 | ||
678 | config MODULES_USE_ELF_RELA | |
679 | bool | |
680 | help | |
681 | Modules only use ELF RELA relocations. Modules with ELF REL | |
682 | relocations will give an error. | |
683 | ||
684 | config MODULES_USE_ELF_REL | |
685 | bool | |
686 | help | |
687 | Modules only use ELF REL relocations. Modules with ELF RELA | |
688 | relocations will give an error. | |
689 | ||
b92021b0 RR |
690 | config HAVE_UNDERSCORE_SYMBOL_PREFIX |
691 | bool | |
692 | help | |
693 | Some architectures generate an _ in front of C symbols; things like | |
694 | module loading and assembly files need to know about this. | |
695 | ||
cc1f0274 FW |
696 | config HAVE_IRQ_EXIT_ON_IRQ_STACK |
697 | bool | |
698 | help | |
699 | Architecture doesn't only execute the irq handler on the irq stack | |
700 | but also irq_exit(). This way we can process softirqs on this irq | |
701 | stack instead of switching to a new one when we call __do_softirq() | |
702 | in the end of an hardirq. | |
703 | This spares a stack switch and improves cache usage on softirq | |
704 | processing. | |
705 | ||
235a8f02 KS |
706 | config PGTABLE_LEVELS |
707 | int | |
708 | default 2 | |
709 | ||
2b68f6ca KC |
710 | config ARCH_HAS_ELF_RANDOMIZE |
711 | bool | |
712 | help | |
713 | An architecture supports choosing randomized locations for | |
714 | stack, mmap, brk, and ET_DYN. Defined functions: | |
715 | - arch_mmap_rnd() | |
204db6ed | 716 | - arch_randomize_brk() |
2b68f6ca | 717 | |
d07e2259 DC |
718 | config HAVE_ARCH_MMAP_RND_BITS |
719 | bool | |
720 | help | |
721 | An arch should select this symbol if it supports setting a variable | |
722 | number of bits for use in establishing the base address for mmap | |
723 | allocations, has MMU enabled and provides values for both: | |
724 | - ARCH_MMAP_RND_BITS_MIN | |
725 | - ARCH_MMAP_RND_BITS_MAX | |
726 | ||
5f56a5df JS |
727 | config HAVE_EXIT_THREAD |
728 | bool | |
729 | help | |
730 | An architecture implements exit_thread. | |
731 | ||
d07e2259 DC |
732 | config ARCH_MMAP_RND_BITS_MIN |
733 | int | |
734 | ||
735 | config ARCH_MMAP_RND_BITS_MAX | |
736 | int | |
737 | ||
738 | config ARCH_MMAP_RND_BITS_DEFAULT | |
739 | int | |
740 | ||
741 | config ARCH_MMAP_RND_BITS | |
742 | int "Number of bits to use for ASLR of mmap base address" if EXPERT | |
743 | range ARCH_MMAP_RND_BITS_MIN ARCH_MMAP_RND_BITS_MAX | |
744 | default ARCH_MMAP_RND_BITS_DEFAULT if ARCH_MMAP_RND_BITS_DEFAULT | |
745 | default ARCH_MMAP_RND_BITS_MIN | |
746 | depends on HAVE_ARCH_MMAP_RND_BITS | |
747 | help | |
748 | This value can be used to select the number of bits to use to | |
749 | determine the random offset to the base address of vma regions | |
750 | resulting from mmap allocations. This value will be bounded | |
751 | by the architecture's minimum and maximum supported values. | |
752 | ||
753 | This value can be changed after boot using the | |
754 | /proc/sys/vm/mmap_rnd_bits tunable | |
755 | ||
756 | config HAVE_ARCH_MMAP_RND_COMPAT_BITS | |
757 | bool | |
758 | help | |
759 | An arch should select this symbol if it supports running applications | |
760 | in compatibility mode, supports setting a variable number of bits for | |
761 | use in establishing the base address for mmap allocations, has MMU | |
762 | enabled and provides values for both: | |
763 | - ARCH_MMAP_RND_COMPAT_BITS_MIN | |
764 | - ARCH_MMAP_RND_COMPAT_BITS_MAX | |
765 | ||
766 | config ARCH_MMAP_RND_COMPAT_BITS_MIN | |
767 | int | |
768 | ||
769 | config ARCH_MMAP_RND_COMPAT_BITS_MAX | |
770 | int | |
771 | ||
772 | config ARCH_MMAP_RND_COMPAT_BITS_DEFAULT | |
773 | int | |
774 | ||
775 | config ARCH_MMAP_RND_COMPAT_BITS | |
776 | int "Number of bits to use for ASLR of mmap base address for compatible applications" if EXPERT | |
777 | range ARCH_MMAP_RND_COMPAT_BITS_MIN ARCH_MMAP_RND_COMPAT_BITS_MAX | |
778 | default ARCH_MMAP_RND_COMPAT_BITS_DEFAULT if ARCH_MMAP_RND_COMPAT_BITS_DEFAULT | |
779 | default ARCH_MMAP_RND_COMPAT_BITS_MIN | |
780 | depends on HAVE_ARCH_MMAP_RND_COMPAT_BITS | |
781 | help | |
782 | This value can be used to select the number of bits to use to | |
783 | determine the random offset to the base address of vma regions | |
784 | resulting from mmap allocations for compatible applications This | |
785 | value will be bounded by the architecture's minimum and maximum | |
786 | supported values. | |
787 | ||
788 | This value can be changed after boot using the | |
789 | /proc/sys/vm/mmap_rnd_compat_bits tunable | |
790 | ||
1b028f78 DS |
791 | config HAVE_ARCH_COMPAT_MMAP_BASES |
792 | bool | |
793 | help | |
794 | This allows 64bit applications to invoke 32-bit mmap() syscall | |
795 | and vice-versa 32-bit applications to call 64-bit mmap(). | |
796 | Required for applications doing different bitness syscalls. | |
797 | ||
3033f14a JT |
798 | config HAVE_COPY_THREAD_TLS |
799 | bool | |
800 | help | |
801 | Architecture provides copy_thread_tls to accept tls argument via | |
802 | normal C parameter passing, rather than extracting the syscall | |
803 | argument from pt_regs. | |
804 | ||
b9ab5ebb JP |
805 | config HAVE_STACK_VALIDATION |
806 | bool | |
807 | help | |
808 | Architecture supports the 'objtool check' host tool command, which | |
809 | performs compile-time stack metadata validation. | |
810 | ||
af085d90 JP |
811 | config HAVE_RELIABLE_STACKTRACE |
812 | bool | |
813 | help | |
814 | Architecture has a save_stack_trace_tsk_reliable() function which | |
815 | only returns a stack trace if it can guarantee the trace is reliable. | |
816 | ||
468a9428 GS |
817 | config HAVE_ARCH_HASH |
818 | bool | |
819 | default n | |
820 | help | |
821 | If this is set, the architecture provides an <asm/hash.h> | |
822 | file which provides platform-specific implementations of some | |
823 | functions in <linux/hash.h> or fs/namei.c. | |
824 | ||
3a495511 WBG |
825 | config ISA_BUS_API |
826 | def_bool ISA | |
827 | ||
d2125043 AV |
828 | # |
829 | # ABI hall of shame | |
830 | # | |
831 | config CLONE_BACKWARDS | |
832 | bool | |
833 | help | |
834 | Architecture has tls passed as the 4th argument of clone(2), | |
835 | not the 5th one. | |
836 | ||
837 | config CLONE_BACKWARDS2 | |
838 | bool | |
839 | help | |
840 | Architecture has the first two arguments of clone(2) swapped. | |
841 | ||
dfa9771a MS |
842 | config CLONE_BACKWARDS3 |
843 | bool | |
844 | help | |
845 | Architecture has tls passed as the 3rd argument of clone(2), | |
846 | not the 5th one. | |
847 | ||
eaca6eae AV |
848 | config ODD_RT_SIGACTION |
849 | bool | |
850 | help | |
851 | Architecture has unusual rt_sigaction(2) arguments | |
852 | ||
0a0e8cdf AV |
853 | config OLD_SIGSUSPEND |
854 | bool | |
855 | help | |
856 | Architecture has old sigsuspend(2) syscall, of one-argument variety | |
857 | ||
858 | config OLD_SIGSUSPEND3 | |
859 | bool | |
860 | help | |
861 | Even weirder antique ABI - three-argument sigsuspend(2) | |
862 | ||
495dfbf7 AV |
863 | config OLD_SIGACTION |
864 | bool | |
865 | help | |
866 | Architecture has old sigaction(2) syscall. Nope, not the same | |
867 | as OLD_SIGSUSPEND | OLD_SIGSUSPEND3 - alpha has sigsuspend(2), | |
868 | but fairly different variant of sigaction(2), thanks to OSF/1 | |
869 | compatibility... | |
870 | ||
871 | config COMPAT_OLD_SIGACTION | |
872 | bool | |
873 | ||
d4703dda DD |
874 | config 64BIT_TIME |
875 | def_bool ARCH_HAS_64BIT_TIME | |
876 | help | |
877 | This should be selected by all architectures that need to support | |
878 | new system calls with a 64-bit time_t. This is relevant on all 32-bit | |
879 | architectures, and 64-bit architectures as part of compat syscall | |
880 | handling. | |
881 | ||
17435e5f DD |
882 | config COMPAT_32BIT_TIME |
883 | def_bool (!64BIT && 64BIT_TIME) || COMPAT | |
884 | help | |
885 | This enables 32 bit time_t support in addition to 64 bit time_t support. | |
886 | This is relevant on all 32-bit architectures, and 64-bit architectures | |
887 | as part of compat syscall handling. | |
888 | ||
0d4a619b CH |
889 | config ARCH_NO_COHERENT_DMA_MMAP |
890 | bool | |
891 | ||
fff7fb0b ZZ |
892 | config CPU_NO_EFFICIENT_FFS |
893 | def_bool n | |
894 | ||
ba14a194 AL |
895 | config HAVE_ARCH_VMAP_STACK |
896 | def_bool n | |
897 | help | |
898 | An arch should select this symbol if it can support kernel stacks | |
899 | in vmalloc space. This means: | |
900 | ||
901 | - vmalloc space must be large enough to hold many kernel stacks. | |
902 | This may rule out many 32-bit architectures. | |
903 | ||
904 | - Stacks in vmalloc space need to work reliably. For example, if | |
905 | vmap page tables are created on demand, either this mechanism | |
906 | needs to work while the stack points to a virtual address with | |
907 | unpopulated page tables or arch code (switch_to() and switch_mm(), | |
908 | most likely) needs to ensure that the stack's page table entries | |
909 | are populated before running on a possibly unpopulated stack. | |
910 | ||
911 | - If the stack overflows into a guard page, something reasonable | |
912 | should happen. The definition of "reasonable" is flexible, but | |
913 | instantly rebooting without logging anything would be unfriendly. | |
914 | ||
915 | config VMAP_STACK | |
916 | default y | |
917 | bool "Use a virtually-mapped stack" | |
918 | depends on HAVE_ARCH_VMAP_STACK && !KASAN | |
919 | ---help--- | |
920 | Enable this if you want the use virtually-mapped kernel stacks | |
921 | with guard pages. This causes kernel stack overflows to be | |
922 | caught immediately rather than causing difficult-to-diagnose | |
923 | corruption. | |
924 | ||
925 | This is presently incompatible with KASAN because KASAN expects | |
926 | the stack to map directly to the KASAN shadow map using a formula | |
927 | that is incorrect if the stack is in vmalloc space. | |
928 | ||
ad21fc4f LA |
929 | config ARCH_OPTIONAL_KERNEL_RWX |
930 | def_bool n | |
931 | ||
932 | config ARCH_OPTIONAL_KERNEL_RWX_DEFAULT | |
933 | def_bool n | |
934 | ||
935 | config ARCH_HAS_STRICT_KERNEL_RWX | |
936 | def_bool n | |
937 | ||
0f5bf6d0 | 938 | config STRICT_KERNEL_RWX |
ad21fc4f LA |
939 | bool "Make kernel text and rodata read-only" if ARCH_OPTIONAL_KERNEL_RWX |
940 | depends on ARCH_HAS_STRICT_KERNEL_RWX | |
941 | default !ARCH_OPTIONAL_KERNEL_RWX || ARCH_OPTIONAL_KERNEL_RWX_DEFAULT | |
942 | help | |
943 | If this is set, kernel text and rodata memory will be made read-only, | |
944 | and non-text memory will be made non-executable. This provides | |
945 | protection against certain security exploits (e.g. executing the heap | |
946 | or modifying text) | |
947 | ||
948 | These features are considered standard security practice these days. | |
949 | You should say Y here in almost all cases. | |
950 | ||
951 | config ARCH_HAS_STRICT_MODULE_RWX | |
952 | def_bool n | |
953 | ||
0f5bf6d0 | 954 | config STRICT_MODULE_RWX |
ad21fc4f LA |
955 | bool "Set loadable kernel module data as NX and text as RO" if ARCH_OPTIONAL_KERNEL_RWX |
956 | depends on ARCH_HAS_STRICT_MODULE_RWX && MODULES | |
957 | default !ARCH_OPTIONAL_KERNEL_RWX || ARCH_OPTIONAL_KERNEL_RWX_DEFAULT | |
958 | help | |
959 | If this is set, module text and rodata memory will be made read-only, | |
960 | and non-text memory will be made non-executable. This provides | |
961 | protection against certain security exploits (e.g. writing to text) | |
962 | ||
ea8c64ac CH |
963 | # select if the architecture provides an asm/dma-direct.h header |
964 | config ARCH_HAS_PHYS_TO_DMA | |
965 | bool | |
966 | ||
7a46ec0e KC |
967 | config ARCH_HAS_REFCOUNT |
968 | bool | |
969 | help | |
970 | An architecture selects this when it has implemented refcount_t | |
971 | using open coded assembly primitives that provide an optimized | |
972 | refcount_t implementation, possibly at the expense of some full | |
973 | refcount state checks of CONFIG_REFCOUNT_FULL=y. | |
974 | ||
975 | The refcount overflow check behavior, however, must be retained. | |
976 | Catching overflows is the primary security concern for protecting | |
977 | against bugs in reference counts. | |
978 | ||
fd25d19f KC |
979 | config REFCOUNT_FULL |
980 | bool "Perform full reference count validation at the expense of speed" | |
981 | help | |
982 | Enabling this switches the refcounting infrastructure from a fast | |
983 | unchecked atomic_t implementation to a fully state checked | |
984 | implementation, which can be (slightly) slower but provides protections | |
985 | against various use-after-free conditions that can be used in | |
986 | security flaw exploits. | |
987 | ||
2521f2c2 | 988 | source "kernel/gcov/Kconfig" |