1# SPDX-License-Identifier: GPL-2.0 2# 3# General architecture dependent options 4# 5 6# 7# Note: arch/$(SRCARCH)/Kconfig needs to be included first so that it can 8# override the default values in this file. 9# 10source "arch/$(SRCARCH)/Kconfig" 11 12menu "General architecture-dependent options" 13 14config CRASH_CORE 15 bool 16 17config KEXEC_CORE 18 select CRASH_CORE 19 bool 20 21config KEXEC_ELF 22 bool 23 24config HAVE_IMA_KEXEC 25 bool 26 27config HOTPLUG_SMT 28 bool 29 30config GENERIC_ENTRY 31 bool 32 33config KPROBES 34 bool "Kprobes" 35 depends on MODULES 36 depends on HAVE_KPROBES 37 select KALLSYMS 38 help 39 Kprobes allows you to trap at almost any kernel address and 40 execute a callback function. register_kprobe() establishes 41 a probepoint and specifies the callback. Kprobes is useful 42 for kernel debugging, non-intrusive instrumentation and testing. 43 If in doubt, say "N". 44 45config JUMP_LABEL 46 bool "Optimize very unlikely/likely branches" 47 depends on HAVE_ARCH_JUMP_LABEL 48 depends on CC_HAS_ASM_GOTO 49 help 50 This option enables a transparent branch optimization that 51 makes certain almost-always-true or almost-always-false branch 52 conditions even cheaper to execute within the kernel. 53 54 Certain performance-sensitive kernel code, such as trace points, 55 scheduler functionality, networking code and KVM have such 56 branches and include support for this optimization technique. 57 58 If it is detected that the compiler has support for "asm goto", 59 the kernel will compile such branches with just a nop 60 instruction. When the condition flag is toggled to true, the 61 nop will be converted to a jump instruction to execute the 62 conditional block of instructions. 63 64 This technique lowers overhead and stress on the branch prediction 65 of the processor and generally makes the kernel faster. The update 66 of the condition is slower, but those are always very rare. 67 68 ( On 32-bit x86, the necessary options added to the compiler 69 flags may increase the size of the kernel slightly. ) 70 71config STATIC_KEYS_SELFTEST 72 bool "Static key selftest" 73 depends on JUMP_LABEL 74 help 75 Boot time self-test of the branch patching code. 76 77config STATIC_CALL_SELFTEST 78 bool "Static call selftest" 79 depends on HAVE_STATIC_CALL 80 help 81 Boot time self-test of the call patching code. 82 83config OPTPROBES 84 def_bool y 85 depends on KPROBES && HAVE_OPTPROBES 86 select TASKS_RCU if PREEMPTION 87 88config KPROBES_ON_FTRACE 89 def_bool y 90 depends on KPROBES && HAVE_KPROBES_ON_FTRACE 91 depends on DYNAMIC_FTRACE_WITH_REGS 92 help 93 If function tracer is enabled and the arch supports full 94 passing of pt_regs to function tracing, then kprobes can 95 optimize on top of function tracing. 96 97config UPROBES 98 def_bool n 99 depends on ARCH_SUPPORTS_UPROBES 100 help 101 Uprobes is the user-space counterpart to kprobes: they 102 enable instrumentation applications (such as 'perf probe') 103 to establish unintrusive probes in user-space binaries and 104 libraries, by executing handler functions when the probes 105 are hit by user-space applications. 106 107 ( These probes come in the form of single-byte breakpoints, 108 managed by the kernel and kept transparent to the probed 109 application. ) 110 111config HAVE_64BIT_ALIGNED_ACCESS 112 def_bool 64BIT && !HAVE_EFFICIENT_UNALIGNED_ACCESS 113 help 114 Some architectures require 64 bit accesses to be 64 bit 115 aligned, which also requires structs containing 64 bit values 116 to be 64 bit aligned too. This includes some 32 bit 117 architectures which can do 64 bit accesses, as well as 64 bit 118 architectures without unaligned access. 119 120 This symbol should be selected by an architecture if 64 bit 121 accesses are required to be 64 bit aligned in this way even 122 though it is not a 64 bit architecture. 123 124 See Documentation/core-api/unaligned-memory-access.rst for 125 more information on the topic of unaligned memory accesses. 126 127config HAVE_EFFICIENT_UNALIGNED_ACCESS 128 bool 129 help 130 Some architectures are unable to perform unaligned accesses 131 without the use of get_unaligned/put_unaligned. Others are 132 unable to perform such accesses efficiently (e.g. trap on 133 unaligned access and require fixing it up in the exception 134 handler.) 135 136 This symbol should be selected by an architecture if it can 137 perform unaligned accesses efficiently to allow different 138 code paths to be selected for these cases. Some network 139 drivers, for example, could opt to not fix up alignment 140 problems with received packets if doing so would not help 141 much. 142 143 See Documentation/core-api/unaligned-memory-access.rst for more 144 information on the topic of unaligned memory accesses. 145 146config ARCH_USE_BUILTIN_BSWAP 147 bool 148 help 149 Modern versions of GCC (since 4.4) have builtin functions 150 for handling byte-swapping. Using these, instead of the old 151 inline assembler that the architecture code provides in the 152 __arch_bswapXX() macros, allows the compiler to see what's 153 happening and offers more opportunity for optimisation. In 154 particular, the compiler will be able to combine the byteswap 155 with a nearby load or store and use load-and-swap or 156 store-and-swap instructions if the architecture has them. It 157 should almost *never* result in code which is worse than the 158 hand-coded assembler in <asm/swab.h>. But just in case it 159 does, the use of the builtins is optional. 160 161 Any architecture with load-and-swap or store-and-swap 162 instructions should set this. And it shouldn't hurt to set it 163 on architectures that don't have such instructions. 164 165config KRETPROBES 166 def_bool y 167 depends on KPROBES && HAVE_KRETPROBES 168 169config USER_RETURN_NOTIFIER 170 bool 171 depends on HAVE_USER_RETURN_NOTIFIER 172 help 173 Provide a kernel-internal notification when a cpu is about to 174 switch to user mode. 175 176config HAVE_IOREMAP_PROT 177 bool 178 179config HAVE_KPROBES 180 bool 181 182config HAVE_KRETPROBES 183 bool 184 185config HAVE_OPTPROBES 186 bool 187 188config HAVE_KPROBES_ON_FTRACE 189 bool 190 191config ARCH_CORRECT_STACKTRACE_ON_KRETPROBE 192 bool 193 help 194 Since kretprobes modifies return address on the stack, the 195 stacktrace may see the kretprobe trampoline address instead 196 of correct one. If the architecture stacktrace code and 197 unwinder can adjust such entries, select this configuration. 198 199config HAVE_FUNCTION_ERROR_INJECTION 200 bool 201 202config HAVE_NMI 203 bool 204 205config TRACE_IRQFLAGS_SUPPORT 206 bool 207 208# 209# An arch should select this if it provides all these things: 210# 211# task_pt_regs() in asm/processor.h or asm/ptrace.h 212# arch_has_single_step() if there is hardware single-step support 213# arch_has_block_step() if there is hardware block-step support 214# asm/syscall.h supplying asm-generic/syscall.h interface 215# linux/regset.h user_regset interfaces 216# CORE_DUMP_USE_REGSET #define'd in linux/elf.h 217# TIF_SYSCALL_TRACE calls tracehook_report_syscall_{entry,exit} 218# TIF_NOTIFY_RESUME calls tracehook_notify_resume() 219# signal delivery calls tracehook_signal_handler() 220# 221config HAVE_ARCH_TRACEHOOK 222 bool 223 224config HAVE_DMA_CONTIGUOUS 225 bool 226 227config GENERIC_SMP_IDLE_THREAD 228 bool 229 230config GENERIC_IDLE_POLL_SETUP 231 bool 232 233config ARCH_HAS_FORTIFY_SOURCE 234 bool 235 help 236 An architecture should select this when it can successfully 237 build and run with CONFIG_FORTIFY_SOURCE. 238 239# 240# Select if the arch provides a historic keepinit alias for the retain_initrd 241# command line option 242# 243config ARCH_HAS_KEEPINITRD 244 bool 245 246# Select if arch has all set_memory_ro/rw/x/nx() functions in asm/cacheflush.h 247config ARCH_HAS_SET_MEMORY 248 bool 249 250# Select if arch has all set_direct_map_invalid/default() functions 251config ARCH_HAS_SET_DIRECT_MAP 252 bool 253 254# 255# Select if the architecture provides the arch_dma_set_uncached symbol to 256# either provide an uncached segment alias for a DMA allocation, or 257# to remap the page tables in place. 258# 259config ARCH_HAS_DMA_SET_UNCACHED 260 bool 261 262# 263# Select if the architectures provides the arch_dma_clear_uncached symbol 264# to undo an in-place page table remap for uncached access. 265# 266config ARCH_HAS_DMA_CLEAR_UNCACHED 267 bool 268 269# Select if arch init_task must go in the __init_task_data section 270config ARCH_TASK_STRUCT_ON_STACK 271 bool 272 273# Select if arch has its private alloc_task_struct() function 274config ARCH_TASK_STRUCT_ALLOCATOR 275 bool 276 277config HAVE_ARCH_THREAD_STRUCT_WHITELIST 278 bool 279 depends on !ARCH_TASK_STRUCT_ALLOCATOR 280 help 281 An architecture should select this to provide hardened usercopy 282 knowledge about what region of the thread_struct should be 283 whitelisted for copying to userspace. Normally this is only the 284 FPU registers. Specifically, arch_thread_struct_whitelist() 285 should be implemented. Without this, the entire thread_struct 286 field in task_struct will be left whitelisted. 287 288# Select if arch has its private alloc_thread_stack() function 289config ARCH_THREAD_STACK_ALLOCATOR 290 bool 291 292# Select if arch wants to size task_struct dynamically via arch_task_struct_size: 293config ARCH_WANTS_DYNAMIC_TASK_STRUCT 294 bool 295 296config ARCH_WANTS_NO_INSTR 297 bool 298 help 299 An architecture should select this if the noinstr macro is being used on 300 functions to denote that the toolchain should avoid instrumenting such 301 functions and is required for correctness. 302 303config ARCH_32BIT_OFF_T 304 bool 305 depends on !64BIT 306 help 307 All new 32-bit architectures should have 64-bit off_t type on 308 userspace side which corresponds to the loff_t kernel type. This 309 is the requirement for modern ABIs. Some existing architectures 310 still support 32-bit off_t. This option is enabled for all such 311 architectures explicitly. 312 313# Selected by 64 bit architectures which have a 32 bit f_tinode in struct ustat 314config ARCH_32BIT_USTAT_F_TINODE 315 bool 316 317config HAVE_ASM_MODVERSIONS 318 bool 319 help 320 This symbol should be selected by an architecture if it provides 321 <asm/asm-prototypes.h> to support the module versioning for symbols 322 exported from assembly code. 323 324config HAVE_REGS_AND_STACK_ACCESS_API 325 bool 326 help 327 This symbol should be selected by an architecture if it supports 328 the API needed to access registers and stack entries from pt_regs, 329 declared in asm/ptrace.h 330 For example the kprobes-based event tracer needs this API. 331 332config HAVE_RSEQ 333 bool 334 depends on HAVE_REGS_AND_STACK_ACCESS_API 335 help 336 This symbol should be selected by an architecture if it 337 supports an implementation of restartable sequences. 338 339config HAVE_FUNCTION_ARG_ACCESS_API 340 bool 341 help 342 This symbol should be selected by an architecture if it supports 343 the API needed to access function arguments from pt_regs, 344 declared in asm/ptrace.h 345 346config HAVE_HW_BREAKPOINT 347 bool 348 depends on PERF_EVENTS 349 350config HAVE_MIXED_BREAKPOINTS_REGS 351 bool 352 depends on HAVE_HW_BREAKPOINT 353 help 354 Depending on the arch implementation of hardware breakpoints, 355 some of them have separate registers for data and instruction 356 breakpoints addresses, others have mixed registers to store 357 them but define the access type in a control register. 358 Select this option if your arch implements breakpoints under the 359 latter fashion. 360 361config HAVE_USER_RETURN_NOTIFIER 362 bool 363 364config HAVE_PERF_EVENTS_NMI 365 bool 366 help 367 System hardware can generate an NMI using the perf event 368 subsystem. Also has support for calculating CPU cycle events 369 to determine how many clock cycles in a given period. 370 371config HAVE_HARDLOCKUP_DETECTOR_PERF 372 bool 373 depends on HAVE_PERF_EVENTS_NMI 374 help 375 The arch chooses to use the generic perf-NMI-based hardlockup 376 detector. Must define HAVE_PERF_EVENTS_NMI. 377 378config HAVE_NMI_WATCHDOG 379 depends on HAVE_NMI 380 bool 381 help 382 The arch provides a low level NMI watchdog. It provides 383 asm/nmi.h, and defines its own arch_touch_nmi_watchdog(). 384 385config HAVE_HARDLOCKUP_DETECTOR_ARCH 386 bool 387 select HAVE_NMI_WATCHDOG 388 help 389 The arch chooses to provide its own hardlockup detector, which is 390 a superset of the HAVE_NMI_WATCHDOG. It also conforms to config 391 interfaces and parameters provided by hardlockup detector subsystem. 392 393config HAVE_PERF_REGS 394 bool 395 help 396 Support selective register dumps for perf events. This includes 397 bit-mapping of each registers and a unique architecture id. 398 399config HAVE_PERF_USER_STACK_DUMP 400 bool 401 help 402 Support user stack dumps for perf event samples. This needs 403 access to the user stack pointer which is not unified across 404 architectures. 405 406config HAVE_ARCH_JUMP_LABEL 407 bool 408 409config HAVE_ARCH_JUMP_LABEL_RELATIVE 410 bool 411 412config MMU_GATHER_TABLE_FREE 413 bool 414 415config MMU_GATHER_RCU_TABLE_FREE 416 bool 417 select MMU_GATHER_TABLE_FREE 418 419config MMU_GATHER_PAGE_SIZE 420 bool 421 422config MMU_GATHER_NO_RANGE 423 bool 424 425config MMU_GATHER_NO_GATHER 426 bool 427 depends on MMU_GATHER_TABLE_FREE 428 429config ARCH_WANT_IRQS_OFF_ACTIVATE_MM 430 bool 431 help 432 Temporary select until all architectures can be converted to have 433 irqs disabled over activate_mm. Architectures that do IPI based TLB 434 shootdowns should enable this. 435 436config ARCH_HAVE_NMI_SAFE_CMPXCHG 437 bool 438 439config HAVE_ALIGNED_STRUCT_PAGE 440 bool 441 help 442 This makes sure that struct pages are double word aligned and that 443 e.g. the SLUB allocator can perform double word atomic operations 444 on a struct page for better performance. However selecting this 445 might increase the size of a struct page by a word. 446 447config HAVE_CMPXCHG_LOCAL 448 bool 449 450config HAVE_CMPXCHG_DOUBLE 451 bool 452 453config ARCH_WEAK_RELEASE_ACQUIRE 454 bool 455 456config ARCH_WANT_IPC_PARSE_VERSION 457 bool 458 459config ARCH_WANT_COMPAT_IPC_PARSE_VERSION 460 bool 461 462config ARCH_WANT_OLD_COMPAT_IPC 463 select ARCH_WANT_COMPAT_IPC_PARSE_VERSION 464 bool 465 466config HAVE_ARCH_SECCOMP 467 bool 468 help 469 An arch should select this symbol to support seccomp mode 1 (the fixed 470 syscall policy), and must provide an overrides for __NR_seccomp_sigreturn, 471 and compat syscalls if the asm-generic/seccomp.h defaults need adjustment: 472 - __NR_seccomp_read_32 473 - __NR_seccomp_write_32 474 - __NR_seccomp_exit_32 475 - __NR_seccomp_sigreturn_32 476 477config HAVE_ARCH_SECCOMP_FILTER 478 bool 479 select HAVE_ARCH_SECCOMP 480 help 481 An arch should select this symbol if it provides all of these things: 482 - all the requirements for HAVE_ARCH_SECCOMP 483 - syscall_get_arch() 484 - syscall_get_arguments() 485 - syscall_rollback() 486 - syscall_set_return_value() 487 - SIGSYS siginfo_t support 488 - secure_computing is called from a ptrace_event()-safe context 489 - secure_computing return value is checked and a return value of -1 490 results in the system call being skipped immediately. 491 - seccomp syscall wired up 492 - if !HAVE_SPARSE_SYSCALL_NR, have SECCOMP_ARCH_NATIVE, 493 SECCOMP_ARCH_NATIVE_NR, SECCOMP_ARCH_NATIVE_NAME defined. If 494 COMPAT is supported, have the SECCOMP_ARCH_COMPAT* defines too. 495 496config SECCOMP 497 prompt "Enable seccomp to safely execute untrusted bytecode" 498 def_bool y 499 depends on HAVE_ARCH_SECCOMP 500 help 501 This kernel feature is useful for number crunching applications 502 that may need to handle untrusted bytecode during their 503 execution. By using pipes or other transports made available 504 to the process as file descriptors supporting the read/write 505 syscalls, it's possible to isolate those applications in their 506 own address space using seccomp. Once seccomp is enabled via 507 prctl(PR_SET_SECCOMP) or the seccomp() syscall, it cannot be 508 disabled and the task is only allowed to execute a few safe 509 syscalls defined by each seccomp mode. 510 511 If unsure, say Y. 512 513config SECCOMP_FILTER 514 def_bool y 515 depends on HAVE_ARCH_SECCOMP_FILTER && SECCOMP && NET 516 help 517 Enable tasks to build secure computing environments defined 518 in terms of Berkeley Packet Filter programs which implement 519 task-defined system call filtering polices. 520 521 See Documentation/userspace-api/seccomp_filter.rst for details. 522 523config SECCOMP_CACHE_DEBUG 524 bool "Show seccomp filter cache status in /proc/pid/seccomp_cache" 525 depends on SECCOMP_FILTER && !HAVE_SPARSE_SYSCALL_NR 526 depends on PROC_FS 527 help 528 This enables the /proc/pid/seccomp_cache interface to monitor 529 seccomp cache data. The file format is subject to change. Reading 530 the file requires CAP_SYS_ADMIN. 531 532 This option is for debugging only. Enabling presents the risk that 533 an adversary may be able to infer the seccomp filter logic. 534 535 If unsure, say N. 536 537config HAVE_ARCH_STACKLEAK 538 bool 539 help 540 An architecture should select this if it has the code which 541 fills the used part of the kernel stack with the STACKLEAK_POISON 542 value before returning from system calls. 543 544config HAVE_STACKPROTECTOR 545 bool 546 help 547 An arch should select this symbol if: 548 - it has implemented a stack canary (e.g. __stack_chk_guard) 549 550config STACKPROTECTOR 551 bool "Stack Protector buffer overflow detection" 552 depends on HAVE_STACKPROTECTOR 553 depends on $(cc-option,-fstack-protector) 554 default y 555 help 556 This option turns on the "stack-protector" GCC feature. This 557 feature puts, at the beginning of functions, a canary value on 558 the stack just before the return address, and validates 559 the value just before actually returning. Stack based buffer 560 overflows (that need to overwrite this return address) now also 561 overwrite the canary, which gets detected and the attack is then 562 neutralized via a kernel panic. 563 564 Functions will have the stack-protector canary logic added if they 565 have an 8-byte or larger character array on the stack. 566 567 This feature requires gcc version 4.2 or above, or a distribution 568 gcc with the feature backported ("-fstack-protector"). 569 570 On an x86 "defconfig" build, this feature adds canary checks to 571 about 3% of all kernel functions, which increases kernel code size 572 by about 0.3%. 573 574config STACKPROTECTOR_STRONG 575 bool "Strong Stack Protector" 576 depends on STACKPROTECTOR 577 depends on $(cc-option,-fstack-protector-strong) 578 default y 579 help 580 Functions will have the stack-protector canary logic added in any 581 of the following conditions: 582 583 - local variable's address used as part of the right hand side of an 584 assignment or function argument 585 - local variable is an array (or union containing an array), 586 regardless of array type or length 587 - uses register local variables 588 589 This feature requires gcc version 4.9 or above, or a distribution 590 gcc with the feature backported ("-fstack-protector-strong"). 591 592 On an x86 "defconfig" build, this feature adds canary checks to 593 about 20% of all kernel functions, which increases the kernel code 594 size by about 2%. 595 596config ARCH_SUPPORTS_SHADOW_CALL_STACK 597 bool 598 help 599 An architecture should select this if it supports Clang's Shadow 600 Call Stack and implements runtime support for shadow stack 601 switching. 602 603config SHADOW_CALL_STACK 604 bool "Clang Shadow Call Stack" 605 depends on CC_IS_CLANG && ARCH_SUPPORTS_SHADOW_CALL_STACK 606 depends on DYNAMIC_FTRACE_WITH_REGS || !FUNCTION_GRAPH_TRACER 607 help 608 This option enables Clang's Shadow Call Stack, which uses a 609 shadow stack to protect function return addresses from being 610 overwritten by an attacker. More information can be found in 611 Clang's documentation: 612 613 https://clang.llvm.org/docs/ShadowCallStack.html 614 615 Note that security guarantees in the kernel differ from the 616 ones documented for user space. The kernel must store addresses 617 of shadow stacks in memory, which means an attacker capable of 618 reading and writing arbitrary memory may be able to locate them 619 and hijack control flow by modifying the stacks. 620 621config LTO 622 bool 623 help 624 Selected if the kernel will be built using the compiler's LTO feature. 625 626config LTO_CLANG 627 bool 628 select LTO 629 help 630 Selected if the kernel will be built using Clang's LTO feature. 631 632config ARCH_SUPPORTS_LTO_CLANG 633 bool 634 help 635 An architecture should select this option if it supports: 636 - compiling with Clang, 637 - compiling inline assembly with Clang's integrated assembler, 638 - and linking with LLD. 639 640config ARCH_SUPPORTS_LTO_CLANG_THIN 641 bool 642 help 643 An architecture should select this option if it can support Clang's 644 ThinLTO mode. 645 646config HAS_LTO_CLANG 647 def_bool y 648 depends on CC_IS_CLANG && LD_IS_LLD && AS_IS_LLVM 649 depends on $(success,$(NM) --help | head -n 1 | grep -qi llvm) 650 depends on $(success,$(AR) --help | head -n 1 | grep -qi llvm) 651 depends on ARCH_SUPPORTS_LTO_CLANG 652 depends on !FTRACE_MCOUNT_USE_RECORDMCOUNT 653 depends on !KASAN || KASAN_HW_TAGS 654 depends on !GCOV_KERNEL 655 help 656 The compiler and Kconfig options support building with Clang's 657 LTO. 658 659choice 660 prompt "Link Time Optimization (LTO)" 661 default LTO_NONE 662 help 663 This option enables Link Time Optimization (LTO), which allows the 664 compiler to optimize binaries globally. 665 666 If unsure, select LTO_NONE. Note that LTO is very resource-intensive 667 so it's disabled by default. 668 669config LTO_NONE 670 bool "None" 671 help 672 Build the kernel normally, without Link Time Optimization (LTO). 673 674config LTO_CLANG_FULL 675 bool "Clang Full LTO (EXPERIMENTAL)" 676 depends on HAS_LTO_CLANG 677 depends on !COMPILE_TEST 678 select LTO_CLANG 679 help 680 This option enables Clang's full Link Time Optimization (LTO), which 681 allows the compiler to optimize the kernel globally. If you enable 682 this option, the compiler generates LLVM bitcode instead of ELF 683 object files, and the actual compilation from bitcode happens at 684 the LTO link step, which may take several minutes depending on the 685 kernel configuration. More information can be found from LLVM's 686 documentation: 687 688 https://llvm.org/docs/LinkTimeOptimization.html 689 690 During link time, this option can use a large amount of RAM, and 691 may take much longer than the ThinLTO option. 692 693config LTO_CLANG_THIN 694 bool "Clang ThinLTO (EXPERIMENTAL)" 695 depends on HAS_LTO_CLANG && ARCH_SUPPORTS_LTO_CLANG_THIN 696 select LTO_CLANG 697 help 698 This option enables Clang's ThinLTO, which allows for parallel 699 optimization and faster incremental compiles compared to the 700 CONFIG_LTO_CLANG_FULL option. More information can be found 701 from Clang's documentation: 702 703 https://clang.llvm.org/docs/ThinLTO.html 704 705 If unsure, say Y. 706endchoice 707 708config ARCH_SUPPORTS_CFI_CLANG 709 bool 710 help 711 An architecture should select this option if it can support Clang's 712 Control-Flow Integrity (CFI) checking. 713 714config CFI_CLANG 715 bool "Use Clang's Control Flow Integrity (CFI)" 716 depends on LTO_CLANG && ARCH_SUPPORTS_CFI_CLANG 717 # Clang >= 12: 718 # - https://bugs.llvm.org/show_bug.cgi?id=46258 719 # - https://bugs.llvm.org/show_bug.cgi?id=47479 720 depends on CLANG_VERSION >= 120000 721 select KALLSYMS 722 help 723 This option enables Clang’s forward-edge Control Flow Integrity 724 (CFI) checking, where the compiler injects a runtime check to each 725 indirect function call to ensure the target is a valid function with 726 the correct static type. This restricts possible call targets and 727 makes it more difficult for an attacker to exploit bugs that allow 728 the modification of stored function pointers. More information can be 729 found from Clang's documentation: 730 731 https://clang.llvm.org/docs/ControlFlowIntegrity.html 732 733config CFI_CLANG_SHADOW 734 bool "Use CFI shadow to speed up cross-module checks" 735 default y 736 depends on CFI_CLANG && MODULES 737 help 738 If you select this option, the kernel builds a fast look-up table of 739 CFI check functions in loaded modules to reduce performance overhead. 740 741 If unsure, say Y. 742 743config CFI_PERMISSIVE 744 bool "Use CFI in permissive mode" 745 depends on CFI_CLANG 746 help 747 When selected, Control Flow Integrity (CFI) violations result in a 748 warning instead of a kernel panic. This option should only be used 749 for finding indirect call type mismatches during development. 750 751 If unsure, say N. 752 753config HAVE_ARCH_WITHIN_STACK_FRAMES 754 bool 755 help 756 An architecture should select this if it can walk the kernel stack 757 frames to determine if an object is part of either the arguments 758 or local variables (i.e. that it excludes saved return addresses, 759 and similar) by implementing an inline arch_within_stack_frames(), 760 which is used by CONFIG_HARDENED_USERCOPY. 761 762config HAVE_CONTEXT_TRACKING 763 bool 764 help 765 Provide kernel/user boundaries probes necessary for subsystems 766 that need it, such as userspace RCU extended quiescent state. 767 Syscalls need to be wrapped inside user_exit()-user_enter(), either 768 optimized behind static key or through the slow path using TIF_NOHZ 769 flag. Exceptions handlers must be wrapped as well. Irqs are already 770 protected inside rcu_irq_enter/rcu_irq_exit() but preemption or signal 771 handling on irq exit still need to be protected. 772 773config HAVE_CONTEXT_TRACKING_OFFSTACK 774 bool 775 help 776 Architecture neither relies on exception_enter()/exception_exit() 777 nor on schedule_user(). Also preempt_schedule_notrace() and 778 preempt_schedule_irq() can't be called in a preemptible section 779 while context tracking is CONTEXT_USER. This feature reflects a sane 780 entry implementation where the following requirements are met on 781 critical entry code, ie: before user_exit() or after user_enter(): 782 783 - Critical entry code isn't preemptible (or better yet: 784 not interruptible). 785 - No use of RCU read side critical sections, unless rcu_nmi_enter() 786 got called. 787 - No use of instrumentation, unless instrumentation_begin() got 788 called. 789 790config HAVE_TIF_NOHZ 791 bool 792 help 793 Arch relies on TIF_NOHZ and syscall slow path to implement context 794 tracking calls to user_enter()/user_exit(). 795 796config HAVE_VIRT_CPU_ACCOUNTING 797 bool 798 799config HAVE_VIRT_CPU_ACCOUNTING_IDLE 800 bool 801 help 802 Architecture has its own way to account idle CPU time and therefore 803 doesn't implement vtime_account_idle(). 804 805config ARCH_HAS_SCALED_CPUTIME 806 bool 807 808config HAVE_VIRT_CPU_ACCOUNTING_GEN 809 bool 810 default y if 64BIT 811 help 812 With VIRT_CPU_ACCOUNTING_GEN, cputime_t becomes 64-bit. 813 Before enabling this option, arch code must be audited 814 to ensure there are no races in concurrent read/write of 815 cputime_t. For example, reading/writing 64-bit cputime_t on 816 some 32-bit arches may require multiple accesses, so proper 817 locking is needed to protect against concurrent accesses. 818 819config HAVE_IRQ_TIME_ACCOUNTING 820 bool 821 help 822 Archs need to ensure they use a high enough resolution clock to 823 support irq time accounting and then call enable_sched_clock_irqtime(). 824 825config HAVE_MOVE_PUD 826 bool 827 help 828 Architectures that select this are able to move page tables at the 829 PUD level. If there are only 3 page table levels, the move effectively 830 happens at the PGD level. 831 832config HAVE_MOVE_PMD 833 bool 834 help 835 Archs that select this are able to move page tables at the PMD level. 836 837config HAVE_ARCH_TRANSPARENT_HUGEPAGE 838 bool 839 840config HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD 841 bool 842 843config HAVE_ARCH_HUGE_VMAP 844 bool 845 846# 847# Archs that select this would be capable of PMD-sized vmaps (i.e., 848# arch_vmap_pmd_supported() returns true), and they must make no assumptions 849# that vmalloc memory is mapped with PAGE_SIZE ptes. The VM_NO_HUGE_VMAP flag 850# can be used to prohibit arch-specific allocations from using hugepages to 851# help with this (e.g., modules may require it). 852# 853config HAVE_ARCH_HUGE_VMALLOC 854 depends on HAVE_ARCH_HUGE_VMAP 855 bool 856 857config ARCH_WANT_HUGE_PMD_SHARE 858 bool 859 860config HAVE_ARCH_SOFT_DIRTY 861 bool 862 863config HAVE_MOD_ARCH_SPECIFIC 864 bool 865 help 866 The arch uses struct mod_arch_specific to store data. Many arches 867 just need a simple module loader without arch specific data - those 868 should not enable this. 869 870config MODULES_USE_ELF_RELA 871 bool 872 help 873 Modules only use ELF RELA relocations. Modules with ELF REL 874 relocations will give an error. 875 876config MODULES_USE_ELF_REL 877 bool 878 help 879 Modules only use ELF REL relocations. Modules with ELF RELA 880 relocations will give an error. 881 882config HAVE_IRQ_EXIT_ON_IRQ_STACK 883 bool 884 help 885 Architecture doesn't only execute the irq handler on the irq stack 886 but also irq_exit(). This way we can process softirqs on this irq 887 stack instead of switching to a new one when we call __do_softirq() 888 in the end of an hardirq. 889 This spares a stack switch and improves cache usage on softirq 890 processing. 891 892config HAVE_SOFTIRQ_ON_OWN_STACK 893 bool 894 help 895 Architecture provides a function to run __do_softirq() on a 896 separate stack. 897 898config ALTERNATE_USER_ADDRESS_SPACE 899 bool 900 help 901 Architectures set this when the CPU uses separate address 902 spaces for kernel and user space pointers. In this case, the 903 access_ok() check on a __user pointer is skipped. 904 905config PGTABLE_LEVELS 906 int 907 default 2 908 909config ARCH_HAS_ELF_RANDOMIZE 910 bool 911 help 912 An architecture supports choosing randomized locations for 913 stack, mmap, brk, and ET_DYN. Defined functions: 914 - arch_mmap_rnd() 915 - arch_randomize_brk() 916 917config HAVE_ARCH_MMAP_RND_BITS 918 bool 919 help 920 An arch should select this symbol if it supports setting a variable 921 number of bits for use in establishing the base address for mmap 922 allocations, has MMU enabled and provides values for both: 923 - ARCH_MMAP_RND_BITS_MIN 924 - ARCH_MMAP_RND_BITS_MAX 925 926config HAVE_EXIT_THREAD 927 bool 928 help 929 An architecture implements exit_thread. 930 931config ARCH_MMAP_RND_BITS_MIN 932 int 933 934config ARCH_MMAP_RND_BITS_MAX 935 int 936 937config ARCH_MMAP_RND_BITS_DEFAULT 938 int 939 940config ARCH_MMAP_RND_BITS 941 int "Number of bits to use for ASLR of mmap base address" if EXPERT 942 range ARCH_MMAP_RND_BITS_MIN ARCH_MMAP_RND_BITS_MAX 943 default ARCH_MMAP_RND_BITS_DEFAULT if ARCH_MMAP_RND_BITS_DEFAULT 944 default ARCH_MMAP_RND_BITS_MIN 945 depends on HAVE_ARCH_MMAP_RND_BITS 946 help 947 This value can be used to select the number of bits to use to 948 determine the random offset to the base address of vma regions 949 resulting from mmap allocations. This value will be bounded 950 by the architecture's minimum and maximum supported values. 951 952 This value can be changed after boot using the 953 /proc/sys/vm/mmap_rnd_bits tunable 954 955config HAVE_ARCH_MMAP_RND_COMPAT_BITS 956 bool 957 help 958 An arch should select this symbol if it supports running applications 959 in compatibility mode, supports setting a variable number of bits for 960 use in establishing the base address for mmap allocations, has MMU 961 enabled and provides values for both: 962 - ARCH_MMAP_RND_COMPAT_BITS_MIN 963 - ARCH_MMAP_RND_COMPAT_BITS_MAX 964 965config ARCH_MMAP_RND_COMPAT_BITS_MIN 966 int 967 968config ARCH_MMAP_RND_COMPAT_BITS_MAX 969 int 970 971config ARCH_MMAP_RND_COMPAT_BITS_DEFAULT 972 int 973 974config ARCH_MMAP_RND_COMPAT_BITS 975 int "Number of bits to use for ASLR of mmap base address for compatible applications" if EXPERT 976 range ARCH_MMAP_RND_COMPAT_BITS_MIN ARCH_MMAP_RND_COMPAT_BITS_MAX 977 default ARCH_MMAP_RND_COMPAT_BITS_DEFAULT if ARCH_MMAP_RND_COMPAT_BITS_DEFAULT 978 default ARCH_MMAP_RND_COMPAT_BITS_MIN 979 depends on HAVE_ARCH_MMAP_RND_COMPAT_BITS 980 help 981 This value can be used to select the number of bits to use to 982 determine the random offset to the base address of vma regions 983 resulting from mmap allocations for compatible applications This 984 value will be bounded by the architecture's minimum and maximum 985 supported values. 986 987 This value can be changed after boot using the 988 /proc/sys/vm/mmap_rnd_compat_bits tunable 989 990config HAVE_ARCH_COMPAT_MMAP_BASES 991 bool 992 help 993 This allows 64bit applications to invoke 32-bit mmap() syscall 994 and vice-versa 32-bit applications to call 64-bit mmap(). 995 Required for applications doing different bitness syscalls. 996 997config PAGE_SIZE_LESS_THAN_64KB 998 def_bool y 999 depends on !ARM64_64K_PAGES 1000 depends on !IA64_PAGE_SIZE_64KB 1001 depends on !PAGE_SIZE_64KB 1002 depends on !PARISC_PAGE_SIZE_64KB 1003 depends on !PPC_64K_PAGES 1004 depends on PAGE_SIZE_LESS_THAN_256KB 1005 1006config PAGE_SIZE_LESS_THAN_256KB 1007 def_bool y 1008 depends on !PPC_256K_PAGES 1009 depends on !PAGE_SIZE_256KB 1010 1011# This allows to use a set of generic functions to determine mmap base 1012# address by giving priority to top-down scheme only if the process 1013# is not in legacy mode (compat task, unlimited stack size or 1014# sysctl_legacy_va_layout). 1015# Architecture that selects this option can provide its own version of: 1016# - STACK_RND_MASK 1017config ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT 1018 bool 1019 depends on MMU 1020 select ARCH_HAS_ELF_RANDOMIZE 1021 1022config HAVE_STACK_VALIDATION 1023 bool 1024 help 1025 Architecture supports the 'objtool check' host tool command, which 1026 performs compile-time stack metadata validation. 1027 1028config HAVE_RELIABLE_STACKTRACE 1029 bool 1030 help 1031 Architecture has either save_stack_trace_tsk_reliable() or 1032 arch_stack_walk_reliable() function which only returns a stack trace 1033 if it can guarantee the trace is reliable. 1034 1035config HAVE_ARCH_HASH 1036 bool 1037 default n 1038 help 1039 If this is set, the architecture provides an <asm/hash.h> 1040 file which provides platform-specific implementations of some 1041 functions in <linux/hash.h> or fs/namei.c. 1042 1043config HAVE_ARCH_NVRAM_OPS 1044 bool 1045 1046config ISA_BUS_API 1047 def_bool ISA 1048 1049# 1050# ABI hall of shame 1051# 1052config CLONE_BACKWARDS 1053 bool 1054 help 1055 Architecture has tls passed as the 4th argument of clone(2), 1056 not the 5th one. 1057 1058config CLONE_BACKWARDS2 1059 bool 1060 help 1061 Architecture has the first two arguments of clone(2) swapped. 1062 1063config CLONE_BACKWARDS3 1064 bool 1065 help 1066 Architecture has tls passed as the 3rd argument of clone(2), 1067 not the 5th one. 1068 1069config ODD_RT_SIGACTION 1070 bool 1071 help 1072 Architecture has unusual rt_sigaction(2) arguments 1073 1074config OLD_SIGSUSPEND 1075 bool 1076 help 1077 Architecture has old sigsuspend(2) syscall, of one-argument variety 1078 1079config OLD_SIGSUSPEND3 1080 bool 1081 help 1082 Even weirder antique ABI - three-argument sigsuspend(2) 1083 1084config OLD_SIGACTION 1085 bool 1086 help 1087 Architecture has old sigaction(2) syscall. Nope, not the same 1088 as OLD_SIGSUSPEND | OLD_SIGSUSPEND3 - alpha has sigsuspend(2), 1089 but fairly different variant of sigaction(2), thanks to OSF/1 1090 compatibility... 1091 1092config COMPAT_OLD_SIGACTION 1093 bool 1094 1095config COMPAT_32BIT_TIME 1096 bool "Provide system calls for 32-bit time_t" 1097 default !64BIT || COMPAT 1098 help 1099 This enables 32 bit time_t support in addition to 64 bit time_t support. 1100 This is relevant on all 32-bit architectures, and 64-bit architectures 1101 as part of compat syscall handling. 1102 1103config ARCH_NO_PREEMPT 1104 bool 1105 1106config ARCH_EPHEMERAL_INODES 1107 def_bool n 1108 help 1109 An arch should select this symbol if it doesn't keep track of inode 1110 instances on its own, but instead relies on something else (e.g. the 1111 host kernel for an UML kernel). 1112 1113config ARCH_SUPPORTS_RT 1114 bool 1115 1116config CPU_NO_EFFICIENT_FFS 1117 def_bool n 1118 1119config HAVE_ARCH_VMAP_STACK 1120 def_bool n 1121 help 1122 An arch should select this symbol if it can support kernel stacks 1123 in vmalloc space. This means: 1124 1125 - vmalloc space must be large enough to hold many kernel stacks. 1126 This may rule out many 32-bit architectures. 1127 1128 - Stacks in vmalloc space need to work reliably. For example, if 1129 vmap page tables are created on demand, either this mechanism 1130 needs to work while the stack points to a virtual address with 1131 unpopulated page tables or arch code (switch_to() and switch_mm(), 1132 most likely) needs to ensure that the stack's page table entries 1133 are populated before running on a possibly unpopulated stack. 1134 1135 - If the stack overflows into a guard page, something reasonable 1136 should happen. The definition of "reasonable" is flexible, but 1137 instantly rebooting without logging anything would be unfriendly. 1138 1139config VMAP_STACK 1140 default y 1141 bool "Use a virtually-mapped stack" 1142 depends on HAVE_ARCH_VMAP_STACK 1143 depends on !KASAN || KASAN_HW_TAGS || KASAN_VMALLOC 1144 help 1145 Enable this if you want the use virtually-mapped kernel stacks 1146 with guard pages. This causes kernel stack overflows to be 1147 caught immediately rather than causing difficult-to-diagnose 1148 corruption. 1149 1150 To use this with software KASAN modes, the architecture must support 1151 backing virtual mappings with real shadow memory, and KASAN_VMALLOC 1152 must be enabled. 1153 1154config HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET 1155 def_bool n 1156 help 1157 An arch should select this symbol if it can support kernel stack 1158 offset randomization with calls to add_random_kstack_offset() 1159 during syscall entry and choose_random_kstack_offset() during 1160 syscall exit. Careful removal of -fstack-protector-strong and 1161 -fstack-protector should also be applied to the entry code and 1162 closely examined, as the artificial stack bump looks like an array 1163 to the compiler, so it will attempt to add canary checks regardless 1164 of the static branch state. 1165 1166config RANDOMIZE_KSTACK_OFFSET_DEFAULT 1167 bool "Randomize kernel stack offset on syscall entry" 1168 depends on HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET 1169 help 1170 The kernel stack offset can be randomized (after pt_regs) by 1171 roughly 5 bits of entropy, frustrating memory corruption 1172 attacks that depend on stack address determinism or 1173 cross-syscall address exposures. This feature is controlled 1174 by kernel boot param "randomize_kstack_offset=on/off", and this 1175 config chooses the default boot state. 1176 1177config ARCH_OPTIONAL_KERNEL_RWX 1178 def_bool n 1179 1180config ARCH_OPTIONAL_KERNEL_RWX_DEFAULT 1181 def_bool n 1182 1183config ARCH_HAS_STRICT_KERNEL_RWX 1184 def_bool n 1185 1186config STRICT_KERNEL_RWX 1187 bool "Make kernel text and rodata read-only" if ARCH_OPTIONAL_KERNEL_RWX 1188 depends on ARCH_HAS_STRICT_KERNEL_RWX 1189 default !ARCH_OPTIONAL_KERNEL_RWX || ARCH_OPTIONAL_KERNEL_RWX_DEFAULT 1190 help 1191 If this is set, kernel text and rodata memory will be made read-only, 1192 and non-text memory will be made non-executable. This provides 1193 protection against certain security exploits (e.g. executing the heap 1194 or modifying text) 1195 1196 These features are considered standard security practice these days. 1197 You should say Y here in almost all cases. 1198 1199config ARCH_HAS_STRICT_MODULE_RWX 1200 def_bool n 1201 1202config STRICT_MODULE_RWX 1203 bool "Set loadable kernel module data as NX and text as RO" if ARCH_OPTIONAL_KERNEL_RWX 1204 depends on ARCH_HAS_STRICT_MODULE_RWX && MODULES 1205 default !ARCH_OPTIONAL_KERNEL_RWX || ARCH_OPTIONAL_KERNEL_RWX_DEFAULT 1206 help 1207 If this is set, module text and rodata memory will be made read-only, 1208 and non-text memory will be made non-executable. This provides 1209 protection against certain security exploits (e.g. writing to text) 1210 1211# select if the architecture provides an asm/dma-direct.h header 1212config ARCH_HAS_PHYS_TO_DMA 1213 bool 1214 1215config HAVE_ARCH_COMPILER_H 1216 bool 1217 help 1218 An architecture can select this if it provides an 1219 asm/compiler.h header that should be included after 1220 linux/compiler-*.h in order to override macro definitions that those 1221 headers generally provide. 1222 1223config HAVE_ARCH_PREL32_RELOCATIONS 1224 bool 1225 help 1226 May be selected by an architecture if it supports place-relative 1227 32-bit relocations, both in the toolchain and in the module loader, 1228 in which case relative references can be used in special sections 1229 for PCI fixup, initcalls etc which are only half the size on 64 bit 1230 architectures, and don't require runtime relocation on relocatable 1231 kernels. 1232 1233config ARCH_USE_MEMREMAP_PROT 1234 bool 1235 1236config LOCK_EVENT_COUNTS 1237 bool "Locking event counts collection" 1238 depends on DEBUG_FS 1239 help 1240 Enable light-weight counting of various locking related events 1241 in the system with minimal performance impact. This reduces 1242 the chance of application behavior change because of timing 1243 differences. The counts are reported via debugfs. 1244 1245# Select if the architecture has support for applying RELR relocations. 1246config ARCH_HAS_RELR 1247 bool 1248 1249config RELR 1250 bool "Use RELR relocation packing" 1251 depends on ARCH_HAS_RELR && TOOLS_SUPPORT_RELR 1252 default y 1253 help 1254 Store the kernel's dynamic relocations in the RELR relocation packing 1255 format. Requires a compatible linker (LLD supports this feature), as 1256 well as compatible NM and OBJCOPY utilities (llvm-nm and llvm-objcopy 1257 are compatible). 1258 1259config ARCH_HAS_MEM_ENCRYPT 1260 bool 1261 1262config ARCH_HAS_CC_PLATFORM 1263 bool 1264 1265config HAVE_SPARSE_SYSCALL_NR 1266 bool 1267 help 1268 An architecture should select this if its syscall numbering is sparse 1269 to save space. For example, MIPS architecture has a syscall array with 1270 entries at 4000, 5000 and 6000 locations. This option turns on syscall 1271 related optimizations for a given architecture. 1272 1273config ARCH_HAS_VDSO_DATA 1274 bool 1275 1276config HAVE_STATIC_CALL 1277 bool 1278 1279config HAVE_STATIC_CALL_INLINE 1280 bool 1281 depends on HAVE_STATIC_CALL 1282 1283config HAVE_PREEMPT_DYNAMIC 1284 bool 1285 depends on HAVE_STATIC_CALL 1286 depends on GENERIC_ENTRY 1287 help 1288 Select this if the architecture support boot time preempt setting 1289 on top of static calls. It is strongly advised to support inline 1290 static call to avoid any overhead. 1291 1292config ARCH_WANT_LD_ORPHAN_WARN 1293 bool 1294 help 1295 An arch should select this symbol once all linker sections are explicitly 1296 included, size-asserted, or discarded in the linker scripts. This is 1297 important because we never want expected sections to be placed heuristically 1298 by the linker, since the locations of such sections can change between linker 1299 versions. 1300 1301config HAVE_ARCH_PFN_VALID 1302 bool 1303 1304config ARCH_SUPPORTS_DEBUG_PAGEALLOC 1305 bool 1306 1307config ARCH_SUPPORTS_PAGE_TABLE_CHECK 1308 bool 1309 1310config ARCH_SPLIT_ARG64 1311 bool 1312 help 1313 If a 32-bit architecture requires 64-bit arguments to be split into 1314 pairs of 32-bit arguments, select this option. 1315 1316config ARCH_HAS_ELFCORE_COMPAT 1317 bool 1318 1319config ARCH_HAS_PARANOID_L1D_FLUSH 1320 bool 1321 1322config DYNAMIC_SIGFRAME 1323 bool 1324 1325# Select, if arch has a named attribute group bound to NUMA device nodes. 1326config HAVE_ARCH_NODE_DEV_GROUP 1327 bool 1328 1329source "kernel/gcov/Kconfig" 1330 1331source "scripts/gcc-plugins/Kconfig" 1332 1333endmenu 1334