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 select TASKS_RCU if PREEMPTION 39 help 40 Kprobes allows you to trap at almost any kernel address and 41 execute a callback function. register_kprobe() establishes 42 a probepoint and specifies the callback. Kprobes is useful 43 for kernel debugging, non-intrusive instrumentation and testing. 44 If in doubt, say "N". 45 46config JUMP_LABEL 47 bool "Optimize very unlikely/likely branches" 48 depends on HAVE_ARCH_JUMP_LABEL 49 depends on CC_HAS_ASM_GOTO 50 help 51 This option enables a transparent branch optimization that 52 makes certain almost-always-true or almost-always-false branch 53 conditions even cheaper to execute within the kernel. 54 55 Certain performance-sensitive kernel code, such as trace points, 56 scheduler functionality, networking code and KVM have such 57 branches and include support for this optimization technique. 58 59 If it is detected that the compiler has support for "asm goto", 60 the kernel will compile such branches with just a nop 61 instruction. When the condition flag is toggled to true, the 62 nop will be converted to a jump instruction to execute the 63 conditional block of instructions. 64 65 This technique lowers overhead and stress on the branch prediction 66 of the processor and generally makes the kernel faster. The update 67 of the condition is slower, but those are always very rare. 68 69 ( On 32-bit x86, the necessary options added to the compiler 70 flags may increase the size of the kernel slightly. ) 71 72config STATIC_KEYS_SELFTEST 73 bool "Static key selftest" 74 depends on JUMP_LABEL 75 help 76 Boot time self-test of the branch patching code. 77 78config STATIC_CALL_SELFTEST 79 bool "Static call selftest" 80 depends on HAVE_STATIC_CALL 81 help 82 Boot time self-test of the call patching code. 83 84config OPTPROBES 85 def_bool y 86 depends on KPROBES && HAVE_OPTPROBES 87 select TASKS_RCU if PREEMPTION 88 89config KPROBES_ON_FTRACE 90 def_bool y 91 depends on KPROBES && HAVE_KPROBES_ON_FTRACE 92 depends on DYNAMIC_FTRACE_WITH_REGS 93 help 94 If function tracer is enabled and the arch supports full 95 passing of pt_regs to function tracing, then kprobes can 96 optimize on top of function tracing. 97 98config UPROBES 99 def_bool n 100 depends on ARCH_SUPPORTS_UPROBES 101 help 102 Uprobes is the user-space counterpart to kprobes: they 103 enable instrumentation applications (such as 'perf probe') 104 to establish unintrusive probes in user-space binaries and 105 libraries, by executing handler functions when the probes 106 are hit by user-space applications. 107 108 ( These probes come in the form of single-byte breakpoints, 109 managed by the kernel and kept transparent to the probed 110 application. ) 111 112config HAVE_64BIT_ALIGNED_ACCESS 113 def_bool 64BIT && !HAVE_EFFICIENT_UNALIGNED_ACCESS 114 help 115 Some architectures require 64 bit accesses to be 64 bit 116 aligned, which also requires structs containing 64 bit values 117 to be 64 bit aligned too. This includes some 32 bit 118 architectures which can do 64 bit accesses, as well as 64 bit 119 architectures without unaligned access. 120 121 This symbol should be selected by an architecture if 64 bit 122 accesses are required to be 64 bit aligned in this way even 123 though it is not a 64 bit architecture. 124 125 See Documentation/core-api/unaligned-memory-access.rst for 126 more information on the topic of unaligned memory accesses. 127 128config HAVE_EFFICIENT_UNALIGNED_ACCESS 129 bool 130 help 131 Some architectures are unable to perform unaligned accesses 132 without the use of get_unaligned/put_unaligned. Others are 133 unable to perform such accesses efficiently (e.g. trap on 134 unaligned access and require fixing it up in the exception 135 handler.) 136 137 This symbol should be selected by an architecture if it can 138 perform unaligned accesses efficiently to allow different 139 code paths to be selected for these cases. Some network 140 drivers, for example, could opt to not fix up alignment 141 problems with received packets if doing so would not help 142 much. 143 144 See Documentation/core-api/unaligned-memory-access.rst for more 145 information on the topic of unaligned memory accesses. 146 147config ARCH_USE_BUILTIN_BSWAP 148 bool 149 help 150 Modern versions of GCC (since 4.4) have builtin functions 151 for handling byte-swapping. Using these, instead of the old 152 inline assembler that the architecture code provides in the 153 __arch_bswapXX() macros, allows the compiler to see what's 154 happening and offers more opportunity for optimisation. In 155 particular, the compiler will be able to combine the byteswap 156 with a nearby load or store and use load-and-swap or 157 store-and-swap instructions if the architecture has them. It 158 should almost *never* result in code which is worse than the 159 hand-coded assembler in <asm/swab.h>. But just in case it 160 does, the use of the builtins is optional. 161 162 Any architecture with load-and-swap or store-and-swap 163 instructions should set this. And it shouldn't hurt to set it 164 on architectures that don't have such instructions. 165 166config KRETPROBES 167 def_bool y 168 depends on KPROBES && (HAVE_KRETPROBES || HAVE_RETHOOK) 169 170config KRETPROBE_ON_RETHOOK 171 def_bool y 172 depends on HAVE_RETHOOK 173 depends on KRETPROBES 174 select RETHOOK 175 176config USER_RETURN_NOTIFIER 177 bool 178 depends on HAVE_USER_RETURN_NOTIFIER 179 help 180 Provide a kernel-internal notification when a cpu is about to 181 switch to user mode. 182 183config HAVE_IOREMAP_PROT 184 bool 185 186config HAVE_KPROBES 187 bool 188 189config HAVE_KRETPROBES 190 bool 191 192config HAVE_OPTPROBES 193 bool 194 195config HAVE_KPROBES_ON_FTRACE 196 bool 197 198config ARCH_CORRECT_STACKTRACE_ON_KRETPROBE 199 bool 200 help 201 Since kretprobes modifies return address on the stack, the 202 stacktrace may see the kretprobe trampoline address instead 203 of correct one. If the architecture stacktrace code and 204 unwinder can adjust such entries, select this configuration. 205 206config HAVE_FUNCTION_ERROR_INJECTION 207 bool 208 209config HAVE_NMI 210 bool 211 212config HAVE_FUNCTION_DESCRIPTORS 213 bool 214 215config TRACE_IRQFLAGS_SUPPORT 216 bool 217 218# 219# An arch should select this if it provides all these things: 220# 221# task_pt_regs() in asm/processor.h or asm/ptrace.h 222# arch_has_single_step() if there is hardware single-step support 223# arch_has_block_step() if there is hardware block-step support 224# asm/syscall.h supplying asm-generic/syscall.h interface 225# linux/regset.h user_regset interfaces 226# CORE_DUMP_USE_REGSET #define'd in linux/elf.h 227# TIF_SYSCALL_TRACE calls ptrace_report_syscall_{entry,exit} 228# TIF_NOTIFY_RESUME calls resume_user_mode_work() 229# 230config HAVE_ARCH_TRACEHOOK 231 bool 232 233config HAVE_DMA_CONTIGUOUS 234 bool 235 236config GENERIC_SMP_IDLE_THREAD 237 bool 238 239config GENERIC_IDLE_POLL_SETUP 240 bool 241 242config ARCH_HAS_FORTIFY_SOURCE 243 bool 244 help 245 An architecture should select this when it can successfully 246 build and run with CONFIG_FORTIFY_SOURCE. 247 248# 249# Select if the arch provides a historic keepinit alias for the retain_initrd 250# command line option 251# 252config ARCH_HAS_KEEPINITRD 253 bool 254 255# Select if arch has all set_memory_ro/rw/x/nx() functions in asm/cacheflush.h 256config ARCH_HAS_SET_MEMORY 257 bool 258 259# Select if arch has all set_direct_map_invalid/default() functions 260config ARCH_HAS_SET_DIRECT_MAP 261 bool 262 263# 264# Select if the architecture provides the arch_dma_set_uncached symbol to 265# either provide an uncached segment alias for a DMA allocation, or 266# to remap the page tables in place. 267# 268config ARCH_HAS_DMA_SET_UNCACHED 269 bool 270 271# 272# Select if the architectures provides the arch_dma_clear_uncached symbol 273# to undo an in-place page table remap for uncached access. 274# 275config ARCH_HAS_DMA_CLEAR_UNCACHED 276 bool 277 278# Select if arch init_task must go in the __init_task_data section 279config ARCH_TASK_STRUCT_ON_STACK 280 bool 281 282# Select if arch has its private alloc_task_struct() function 283config ARCH_TASK_STRUCT_ALLOCATOR 284 bool 285 286config HAVE_ARCH_THREAD_STRUCT_WHITELIST 287 bool 288 depends on !ARCH_TASK_STRUCT_ALLOCATOR 289 help 290 An architecture should select this to provide hardened usercopy 291 knowledge about what region of the thread_struct should be 292 whitelisted for copying to userspace. Normally this is only the 293 FPU registers. Specifically, arch_thread_struct_whitelist() 294 should be implemented. Without this, the entire thread_struct 295 field in task_struct will be left whitelisted. 296 297# Select if arch has its private alloc_thread_stack() function 298config ARCH_THREAD_STACK_ALLOCATOR 299 bool 300 301# Select if arch wants to size task_struct dynamically via arch_task_struct_size: 302config ARCH_WANTS_DYNAMIC_TASK_STRUCT 303 bool 304 305config ARCH_WANTS_NO_INSTR 306 bool 307 help 308 An architecture should select this if the noinstr macro is being used on 309 functions to denote that the toolchain should avoid instrumenting such 310 functions and is required for correctness. 311 312config ARCH_32BIT_OFF_T 313 bool 314 depends on !64BIT 315 help 316 All new 32-bit architectures should have 64-bit off_t type on 317 userspace side which corresponds to the loff_t kernel type. This 318 is the requirement for modern ABIs. Some existing architectures 319 still support 32-bit off_t. This option is enabled for all such 320 architectures explicitly. 321 322# Selected by 64 bit architectures which have a 32 bit f_tinode in struct ustat 323config ARCH_32BIT_USTAT_F_TINODE 324 bool 325 326config HAVE_ASM_MODVERSIONS 327 bool 328 help 329 This symbol should be selected by an architecture if it provides 330 <asm/asm-prototypes.h> to support the module versioning for symbols 331 exported from assembly code. 332 333config HAVE_REGS_AND_STACK_ACCESS_API 334 bool 335 help 336 This symbol should be selected by an architecture if it supports 337 the API needed to access registers and stack entries from pt_regs, 338 declared in asm/ptrace.h 339 For example the kprobes-based event tracer needs this API. 340 341config HAVE_RSEQ 342 bool 343 depends on HAVE_REGS_AND_STACK_ACCESS_API 344 help 345 This symbol should be selected by an architecture if it 346 supports an implementation of restartable sequences. 347 348config HAVE_FUNCTION_ARG_ACCESS_API 349 bool 350 help 351 This symbol should be selected by an architecture if it supports 352 the API needed to access function arguments from pt_regs, 353 declared in asm/ptrace.h 354 355config HAVE_HW_BREAKPOINT 356 bool 357 depends on PERF_EVENTS 358 359config HAVE_MIXED_BREAKPOINTS_REGS 360 bool 361 depends on HAVE_HW_BREAKPOINT 362 help 363 Depending on the arch implementation of hardware breakpoints, 364 some of them have separate registers for data and instruction 365 breakpoints addresses, others have mixed registers to store 366 them but define the access type in a control register. 367 Select this option if your arch implements breakpoints under the 368 latter fashion. 369 370config HAVE_USER_RETURN_NOTIFIER 371 bool 372 373config HAVE_PERF_EVENTS_NMI 374 bool 375 help 376 System hardware can generate an NMI using the perf event 377 subsystem. Also has support for calculating CPU cycle events 378 to determine how many clock cycles in a given period. 379 380config HAVE_HARDLOCKUP_DETECTOR_PERF 381 bool 382 depends on HAVE_PERF_EVENTS_NMI 383 help 384 The arch chooses to use the generic perf-NMI-based hardlockup 385 detector. Must define HAVE_PERF_EVENTS_NMI. 386 387config HAVE_NMI_WATCHDOG 388 depends on HAVE_NMI 389 bool 390 help 391 The arch provides a low level NMI watchdog. It provides 392 asm/nmi.h, and defines its own arch_touch_nmi_watchdog(). 393 394config HAVE_HARDLOCKUP_DETECTOR_ARCH 395 bool 396 select HAVE_NMI_WATCHDOG 397 help 398 The arch chooses to provide its own hardlockup detector, which is 399 a superset of the HAVE_NMI_WATCHDOG. It also conforms to config 400 interfaces and parameters provided by hardlockup detector subsystem. 401 402config HAVE_PERF_REGS 403 bool 404 help 405 Support selective register dumps for perf events. This includes 406 bit-mapping of each registers and a unique architecture id. 407 408config HAVE_PERF_USER_STACK_DUMP 409 bool 410 help 411 Support user stack dumps for perf event samples. This needs 412 access to the user stack pointer which is not unified across 413 architectures. 414 415config HAVE_ARCH_JUMP_LABEL 416 bool 417 418config HAVE_ARCH_JUMP_LABEL_RELATIVE 419 bool 420 421config MMU_GATHER_TABLE_FREE 422 bool 423 424config MMU_GATHER_RCU_TABLE_FREE 425 bool 426 select MMU_GATHER_TABLE_FREE 427 428config MMU_GATHER_PAGE_SIZE 429 bool 430 431config MMU_GATHER_NO_RANGE 432 bool 433 434config MMU_GATHER_NO_GATHER 435 bool 436 depends on MMU_GATHER_TABLE_FREE 437 438config ARCH_WANT_IRQS_OFF_ACTIVATE_MM 439 bool 440 help 441 Temporary select until all architectures can be converted to have 442 irqs disabled over activate_mm. Architectures that do IPI based TLB 443 shootdowns should enable this. 444 445config ARCH_HAVE_NMI_SAFE_CMPXCHG 446 bool 447 448config HAVE_ALIGNED_STRUCT_PAGE 449 bool 450 help 451 This makes sure that struct pages are double word aligned and that 452 e.g. the SLUB allocator can perform double word atomic operations 453 on a struct page for better performance. However selecting this 454 might increase the size of a struct page by a word. 455 456config HAVE_CMPXCHG_LOCAL 457 bool 458 459config HAVE_CMPXCHG_DOUBLE 460 bool 461 462config ARCH_WEAK_RELEASE_ACQUIRE 463 bool 464 465config ARCH_WANT_IPC_PARSE_VERSION 466 bool 467 468config ARCH_WANT_COMPAT_IPC_PARSE_VERSION 469 bool 470 471config ARCH_WANT_OLD_COMPAT_IPC 472 select ARCH_WANT_COMPAT_IPC_PARSE_VERSION 473 bool 474 475config HAVE_ARCH_SECCOMP 476 bool 477 help 478 An arch should select this symbol to support seccomp mode 1 (the fixed 479 syscall policy), and must provide an overrides for __NR_seccomp_sigreturn, 480 and compat syscalls if the asm-generic/seccomp.h defaults need adjustment: 481 - __NR_seccomp_read_32 482 - __NR_seccomp_write_32 483 - __NR_seccomp_exit_32 484 - __NR_seccomp_sigreturn_32 485 486config HAVE_ARCH_SECCOMP_FILTER 487 bool 488 select HAVE_ARCH_SECCOMP 489 help 490 An arch should select this symbol if it provides all of these things: 491 - all the requirements for HAVE_ARCH_SECCOMP 492 - syscall_get_arch() 493 - syscall_get_arguments() 494 - syscall_rollback() 495 - syscall_set_return_value() 496 - SIGSYS siginfo_t support 497 - secure_computing is called from a ptrace_event()-safe context 498 - secure_computing return value is checked and a return value of -1 499 results in the system call being skipped immediately. 500 - seccomp syscall wired up 501 - if !HAVE_SPARSE_SYSCALL_NR, have SECCOMP_ARCH_NATIVE, 502 SECCOMP_ARCH_NATIVE_NR, SECCOMP_ARCH_NATIVE_NAME defined. If 503 COMPAT is supported, have the SECCOMP_ARCH_COMPAT* defines too. 504 505config SECCOMP 506 prompt "Enable seccomp to safely execute untrusted bytecode" 507 def_bool y 508 depends on HAVE_ARCH_SECCOMP 509 help 510 This kernel feature is useful for number crunching applications 511 that may need to handle untrusted bytecode during their 512 execution. By using pipes or other transports made available 513 to the process as file descriptors supporting the read/write 514 syscalls, it's possible to isolate those applications in their 515 own address space using seccomp. Once seccomp is enabled via 516 prctl(PR_SET_SECCOMP) or the seccomp() syscall, it cannot be 517 disabled and the task is only allowed to execute a few safe 518 syscalls defined by each seccomp mode. 519 520 If unsure, say Y. 521 522config SECCOMP_FILTER 523 def_bool y 524 depends on HAVE_ARCH_SECCOMP_FILTER && SECCOMP && NET 525 help 526 Enable tasks to build secure computing environments defined 527 in terms of Berkeley Packet Filter programs which implement 528 task-defined system call filtering polices. 529 530 See Documentation/userspace-api/seccomp_filter.rst for details. 531 532config SECCOMP_CACHE_DEBUG 533 bool "Show seccomp filter cache status in /proc/pid/seccomp_cache" 534 depends on SECCOMP_FILTER && !HAVE_SPARSE_SYSCALL_NR 535 depends on PROC_FS 536 help 537 This enables the /proc/pid/seccomp_cache interface to monitor 538 seccomp cache data. The file format is subject to change. Reading 539 the file requires CAP_SYS_ADMIN. 540 541 This option is for debugging only. Enabling presents the risk that 542 an adversary may be able to infer the seccomp filter logic. 543 544 If unsure, say N. 545 546config HAVE_ARCH_STACKLEAK 547 bool 548 help 549 An architecture should select this if it has the code which 550 fills the used part of the kernel stack with the STACKLEAK_POISON 551 value before returning from system calls. 552 553config HAVE_STACKPROTECTOR 554 bool 555 help 556 An arch should select this symbol if: 557 - it has implemented a stack canary (e.g. __stack_chk_guard) 558 559config STACKPROTECTOR 560 bool "Stack Protector buffer overflow detection" 561 depends on HAVE_STACKPROTECTOR 562 depends on $(cc-option,-fstack-protector) 563 default y 564 help 565 This option turns on the "stack-protector" GCC feature. This 566 feature puts, at the beginning of functions, a canary value on 567 the stack just before the return address, and validates 568 the value just before actually returning. Stack based buffer 569 overflows (that need to overwrite this return address) now also 570 overwrite the canary, which gets detected and the attack is then 571 neutralized via a kernel panic. 572 573 Functions will have the stack-protector canary logic added if they 574 have an 8-byte or larger character array on the stack. 575 576 This feature requires gcc version 4.2 or above, or a distribution 577 gcc with the feature backported ("-fstack-protector"). 578 579 On an x86 "defconfig" build, this feature adds canary checks to 580 about 3% of all kernel functions, which increases kernel code size 581 by about 0.3%. 582 583config STACKPROTECTOR_STRONG 584 bool "Strong Stack Protector" 585 depends on STACKPROTECTOR 586 depends on $(cc-option,-fstack-protector-strong) 587 default y 588 help 589 Functions will have the stack-protector canary logic added in any 590 of the following conditions: 591 592 - local variable's address used as part of the right hand side of an 593 assignment or function argument 594 - local variable is an array (or union containing an array), 595 regardless of array type or length 596 - uses register local variables 597 598 This feature requires gcc version 4.9 or above, or a distribution 599 gcc with the feature backported ("-fstack-protector-strong"). 600 601 On an x86 "defconfig" build, this feature adds canary checks to 602 about 20% of all kernel functions, which increases the kernel code 603 size by about 2%. 604 605config ARCH_SUPPORTS_SHADOW_CALL_STACK 606 bool 607 help 608 An architecture should select this if it supports the compiler's 609 Shadow Call Stack and implements runtime support for shadow stack 610 switching. 611 612config SHADOW_CALL_STACK 613 bool "Shadow Call Stack" 614 depends on ARCH_SUPPORTS_SHADOW_CALL_STACK 615 depends on DYNAMIC_FTRACE_WITH_REGS || !FUNCTION_GRAPH_TRACER 616 help 617 This option enables the compiler's Shadow Call Stack, which 618 uses a shadow stack to protect function return addresses from 619 being overwritten by an attacker. More information can be found 620 in the compiler's documentation: 621 622 - Clang: https://clang.llvm.org/docs/ShadowCallStack.html 623 - GCC: https://gcc.gnu.org/onlinedocs/gcc/Instrumentation-Options.html#Instrumentation-Options 624 625 Note that security guarantees in the kernel differ from the 626 ones documented for user space. The kernel must store addresses 627 of shadow stacks in memory, which means an attacker capable of 628 reading and writing arbitrary memory may be able to locate them 629 and hijack control flow by modifying the stacks. 630 631config LTO 632 bool 633 help 634 Selected if the kernel will be built using the compiler's LTO feature. 635 636config LTO_CLANG 637 bool 638 select LTO 639 help 640 Selected if the kernel will be built using Clang's LTO feature. 641 642config ARCH_SUPPORTS_LTO_CLANG 643 bool 644 help 645 An architecture should select this option if it supports: 646 - compiling with Clang, 647 - compiling inline assembly with Clang's integrated assembler, 648 - and linking with LLD. 649 650config ARCH_SUPPORTS_LTO_CLANG_THIN 651 bool 652 help 653 An architecture should select this option if it can support Clang's 654 ThinLTO mode. 655 656config HAS_LTO_CLANG 657 def_bool y 658 depends on CC_IS_CLANG && LD_IS_LLD && AS_IS_LLVM 659 depends on $(success,$(NM) --help | head -n 1 | grep -qi llvm) 660 depends on $(success,$(AR) --help | head -n 1 | grep -qi llvm) 661 depends on ARCH_SUPPORTS_LTO_CLANG 662 depends on !FTRACE_MCOUNT_USE_RECORDMCOUNT 663 depends on !KASAN || KASAN_HW_TAGS 664 depends on !GCOV_KERNEL 665 help 666 The compiler and Kconfig options support building with Clang's 667 LTO. 668 669choice 670 prompt "Link Time Optimization (LTO)" 671 default LTO_NONE 672 help 673 This option enables Link Time Optimization (LTO), which allows the 674 compiler to optimize binaries globally. 675 676 If unsure, select LTO_NONE. Note that LTO is very resource-intensive 677 so it's disabled by default. 678 679config LTO_NONE 680 bool "None" 681 help 682 Build the kernel normally, without Link Time Optimization (LTO). 683 684config LTO_CLANG_FULL 685 bool "Clang Full LTO (EXPERIMENTAL)" 686 depends on HAS_LTO_CLANG 687 depends on !COMPILE_TEST 688 select LTO_CLANG 689 help 690 This option enables Clang's full Link Time Optimization (LTO), which 691 allows the compiler to optimize the kernel globally. If you enable 692 this option, the compiler generates LLVM bitcode instead of ELF 693 object files, and the actual compilation from bitcode happens at 694 the LTO link step, which may take several minutes depending on the 695 kernel configuration. More information can be found from LLVM's 696 documentation: 697 698 https://llvm.org/docs/LinkTimeOptimization.html 699 700 During link time, this option can use a large amount of RAM, and 701 may take much longer than the ThinLTO option. 702 703config LTO_CLANG_THIN 704 bool "Clang ThinLTO (EXPERIMENTAL)" 705 depends on HAS_LTO_CLANG && ARCH_SUPPORTS_LTO_CLANG_THIN 706 select LTO_CLANG 707 help 708 This option enables Clang's ThinLTO, which allows for parallel 709 optimization and faster incremental compiles compared to the 710 CONFIG_LTO_CLANG_FULL option. More information can be found 711 from Clang's documentation: 712 713 https://clang.llvm.org/docs/ThinLTO.html 714 715 If unsure, say Y. 716endchoice 717 718config ARCH_SUPPORTS_CFI_CLANG 719 bool 720 help 721 An architecture should select this option if it can support Clang's 722 Control-Flow Integrity (CFI) checking. 723 724config CFI_CLANG 725 bool "Use Clang's Control Flow Integrity (CFI)" 726 depends on LTO_CLANG && ARCH_SUPPORTS_CFI_CLANG 727 # Clang >= 12: 728 # - https://bugs.llvm.org/show_bug.cgi?id=46258 729 # - https://bugs.llvm.org/show_bug.cgi?id=47479 730 depends on CLANG_VERSION >= 120000 731 select KALLSYMS 732 help 733 This option enables Clang’s forward-edge Control Flow Integrity 734 (CFI) checking, where the compiler injects a runtime check to each 735 indirect function call to ensure the target is a valid function with 736 the correct static type. This restricts possible call targets and 737 makes it more difficult for an attacker to exploit bugs that allow 738 the modification of stored function pointers. More information can be 739 found from Clang's documentation: 740 741 https://clang.llvm.org/docs/ControlFlowIntegrity.html 742 743config CFI_CLANG_SHADOW 744 bool "Use CFI shadow to speed up cross-module checks" 745 default y 746 depends on CFI_CLANG && MODULES 747 help 748 If you select this option, the kernel builds a fast look-up table of 749 CFI check functions in loaded modules to reduce performance overhead. 750 751 If unsure, say Y. 752 753config CFI_PERMISSIVE 754 bool "Use CFI in permissive mode" 755 depends on CFI_CLANG 756 help 757 When selected, Control Flow Integrity (CFI) violations result in a 758 warning instead of a kernel panic. This option should only be used 759 for finding indirect call type mismatches during development. 760 761 If unsure, say N. 762 763config HAVE_ARCH_WITHIN_STACK_FRAMES 764 bool 765 help 766 An architecture should select this if it can walk the kernel stack 767 frames to determine if an object is part of either the arguments 768 or local variables (i.e. that it excludes saved return addresses, 769 and similar) by implementing an inline arch_within_stack_frames(), 770 which is used by CONFIG_HARDENED_USERCOPY. 771 772config HAVE_CONTEXT_TRACKING 773 bool 774 help 775 Provide kernel/user boundaries probes necessary for subsystems 776 that need it, such as userspace RCU extended quiescent state. 777 Syscalls need to be wrapped inside user_exit()-user_enter(), either 778 optimized behind static key or through the slow path using TIF_NOHZ 779 flag. Exceptions handlers must be wrapped as well. Irqs are already 780 protected inside rcu_irq_enter/rcu_irq_exit() but preemption or signal 781 handling on irq exit still need to be protected. 782 783config HAVE_CONTEXT_TRACKING_OFFSTACK 784 bool 785 help 786 Architecture neither relies on exception_enter()/exception_exit() 787 nor on schedule_user(). Also preempt_schedule_notrace() and 788 preempt_schedule_irq() can't be called in a preemptible section 789 while context tracking is CONTEXT_USER. This feature reflects a sane 790 entry implementation where the following requirements are met on 791 critical entry code, ie: before user_exit() or after user_enter(): 792 793 - Critical entry code isn't preemptible (or better yet: 794 not interruptible). 795 - No use of RCU read side critical sections, unless rcu_nmi_enter() 796 got called. 797 - No use of instrumentation, unless instrumentation_begin() got 798 called. 799 800config HAVE_TIF_NOHZ 801 bool 802 help 803 Arch relies on TIF_NOHZ and syscall slow path to implement context 804 tracking calls to user_enter()/user_exit(). 805 806config HAVE_VIRT_CPU_ACCOUNTING 807 bool 808 809config HAVE_VIRT_CPU_ACCOUNTING_IDLE 810 bool 811 help 812 Architecture has its own way to account idle CPU time and therefore 813 doesn't implement vtime_account_idle(). 814 815config ARCH_HAS_SCALED_CPUTIME 816 bool 817 818config HAVE_VIRT_CPU_ACCOUNTING_GEN 819 bool 820 default y if 64BIT 821 help 822 With VIRT_CPU_ACCOUNTING_GEN, cputime_t becomes 64-bit. 823 Before enabling this option, arch code must be audited 824 to ensure there are no races in concurrent read/write of 825 cputime_t. For example, reading/writing 64-bit cputime_t on 826 some 32-bit arches may require multiple accesses, so proper 827 locking is needed to protect against concurrent accesses. 828 829config HAVE_IRQ_TIME_ACCOUNTING 830 bool 831 help 832 Archs need to ensure they use a high enough resolution clock to 833 support irq time accounting and then call enable_sched_clock_irqtime(). 834 835config HAVE_MOVE_PUD 836 bool 837 help 838 Architectures that select this are able to move page tables at the 839 PUD level. If there are only 3 page table levels, the move effectively 840 happens at the PGD level. 841 842config HAVE_MOVE_PMD 843 bool 844 help 845 Archs that select this are able to move page tables at the PMD level. 846 847config HAVE_ARCH_TRANSPARENT_HUGEPAGE 848 bool 849 850config HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD 851 bool 852 853config HAVE_ARCH_HUGE_VMAP 854 bool 855 856# 857# Archs that select this would be capable of PMD-sized vmaps (i.e., 858# arch_vmap_pmd_supported() returns true). The VM_ALLOW_HUGE_VMAP flag 859# must be used to enable allocations to use hugepages. 860# 861config HAVE_ARCH_HUGE_VMALLOC 862 depends on HAVE_ARCH_HUGE_VMAP 863 bool 864 865config ARCH_WANT_HUGE_PMD_SHARE 866 bool 867 868config HAVE_ARCH_SOFT_DIRTY 869 bool 870 871config HAVE_MOD_ARCH_SPECIFIC 872 bool 873 help 874 The arch uses struct mod_arch_specific to store data. Many arches 875 just need a simple module loader without arch specific data - those 876 should not enable this. 877 878config MODULES_USE_ELF_RELA 879 bool 880 help 881 Modules only use ELF RELA relocations. Modules with ELF REL 882 relocations will give an error. 883 884config MODULES_USE_ELF_REL 885 bool 886 help 887 Modules only use ELF REL relocations. Modules with ELF RELA 888 relocations will give an error. 889 890config HAVE_IRQ_EXIT_ON_IRQ_STACK 891 bool 892 help 893 Architecture doesn't only execute the irq handler on the irq stack 894 but also irq_exit(). This way we can process softirqs on this irq 895 stack instead of switching to a new one when we call __do_softirq() 896 in the end of an hardirq. 897 This spares a stack switch and improves cache usage on softirq 898 processing. 899 900config HAVE_SOFTIRQ_ON_OWN_STACK 901 bool 902 help 903 Architecture provides a function to run __do_softirq() on a 904 separate stack. 905 906config ALTERNATE_USER_ADDRESS_SPACE 907 bool 908 help 909 Architectures set this when the CPU uses separate address 910 spaces for kernel and user space pointers. In this case, the 911 access_ok() check on a __user pointer is skipped. 912 913config PGTABLE_LEVELS 914 int 915 default 2 916 917config ARCH_HAS_ELF_RANDOMIZE 918 bool 919 help 920 An architecture supports choosing randomized locations for 921 stack, mmap, brk, and ET_DYN. Defined functions: 922 - arch_mmap_rnd() 923 - arch_randomize_brk() 924 925config HAVE_ARCH_MMAP_RND_BITS 926 bool 927 help 928 An arch should select this symbol if it supports setting a variable 929 number of bits for use in establishing the base address for mmap 930 allocations, has MMU enabled and provides values for both: 931 - ARCH_MMAP_RND_BITS_MIN 932 - ARCH_MMAP_RND_BITS_MAX 933 934config HAVE_EXIT_THREAD 935 bool 936 help 937 An architecture implements exit_thread. 938 939config ARCH_MMAP_RND_BITS_MIN 940 int 941 942config ARCH_MMAP_RND_BITS_MAX 943 int 944 945config ARCH_MMAP_RND_BITS_DEFAULT 946 int 947 948config ARCH_MMAP_RND_BITS 949 int "Number of bits to use for ASLR of mmap base address" if EXPERT 950 range ARCH_MMAP_RND_BITS_MIN ARCH_MMAP_RND_BITS_MAX 951 default ARCH_MMAP_RND_BITS_DEFAULT if ARCH_MMAP_RND_BITS_DEFAULT 952 default ARCH_MMAP_RND_BITS_MIN 953 depends on HAVE_ARCH_MMAP_RND_BITS 954 help 955 This value can be used to select the number of bits to use to 956 determine the random offset to the base address of vma regions 957 resulting from mmap allocations. This value will be bounded 958 by the architecture's minimum and maximum supported values. 959 960 This value can be changed after boot using the 961 /proc/sys/vm/mmap_rnd_bits tunable 962 963config HAVE_ARCH_MMAP_RND_COMPAT_BITS 964 bool 965 help 966 An arch should select this symbol if it supports running applications 967 in compatibility mode, supports setting a variable number of bits for 968 use in establishing the base address for mmap allocations, has MMU 969 enabled and provides values for both: 970 - ARCH_MMAP_RND_COMPAT_BITS_MIN 971 - ARCH_MMAP_RND_COMPAT_BITS_MAX 972 973config ARCH_MMAP_RND_COMPAT_BITS_MIN 974 int 975 976config ARCH_MMAP_RND_COMPAT_BITS_MAX 977 int 978 979config ARCH_MMAP_RND_COMPAT_BITS_DEFAULT 980 int 981 982config ARCH_MMAP_RND_COMPAT_BITS 983 int "Number of bits to use for ASLR of mmap base address for compatible applications" if EXPERT 984 range ARCH_MMAP_RND_COMPAT_BITS_MIN ARCH_MMAP_RND_COMPAT_BITS_MAX 985 default ARCH_MMAP_RND_COMPAT_BITS_DEFAULT if ARCH_MMAP_RND_COMPAT_BITS_DEFAULT 986 default ARCH_MMAP_RND_COMPAT_BITS_MIN 987 depends on HAVE_ARCH_MMAP_RND_COMPAT_BITS 988 help 989 This value can be used to select the number of bits to use to 990 determine the random offset to the base address of vma regions 991 resulting from mmap allocations for compatible applications This 992 value will be bounded by the architecture's minimum and maximum 993 supported values. 994 995 This value can be changed after boot using the 996 /proc/sys/vm/mmap_rnd_compat_bits tunable 997 998config HAVE_ARCH_COMPAT_MMAP_BASES 999 bool 1000 help 1001 This allows 64bit applications to invoke 32-bit mmap() syscall 1002 and vice-versa 32-bit applications to call 64-bit mmap(). 1003 Required for applications doing different bitness syscalls. 1004 1005config PAGE_SIZE_LESS_THAN_64KB 1006 def_bool y 1007 depends on !ARM64_64K_PAGES 1008 depends on !IA64_PAGE_SIZE_64KB 1009 depends on !PAGE_SIZE_64KB 1010 depends on !PARISC_PAGE_SIZE_64KB 1011 depends on !PPC_64K_PAGES 1012 depends on PAGE_SIZE_LESS_THAN_256KB 1013 1014config PAGE_SIZE_LESS_THAN_256KB 1015 def_bool y 1016 depends on !PPC_256K_PAGES 1017 depends on !PAGE_SIZE_256KB 1018 1019# This allows to use a set of generic functions to determine mmap base 1020# address by giving priority to top-down scheme only if the process 1021# is not in legacy mode (compat task, unlimited stack size or 1022# sysctl_legacy_va_layout). 1023# Architecture that selects this option can provide its own version of: 1024# - STACK_RND_MASK 1025config ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT 1026 bool 1027 depends on MMU 1028 select ARCH_HAS_ELF_RANDOMIZE 1029 1030config HAVE_STACK_VALIDATION 1031 bool 1032 help 1033 Architecture supports the 'objtool check' host tool command, which 1034 performs compile-time stack metadata validation. 1035 1036config HAVE_RELIABLE_STACKTRACE 1037 bool 1038 help 1039 Architecture has either save_stack_trace_tsk_reliable() or 1040 arch_stack_walk_reliable() function which only returns a stack trace 1041 if it can guarantee the trace is reliable. 1042 1043config HAVE_ARCH_HASH 1044 bool 1045 default n 1046 help 1047 If this is set, the architecture provides an <asm/hash.h> 1048 file which provides platform-specific implementations of some 1049 functions in <linux/hash.h> or fs/namei.c. 1050 1051config HAVE_ARCH_NVRAM_OPS 1052 bool 1053 1054config ISA_BUS_API 1055 def_bool ISA 1056 1057# 1058# ABI hall of shame 1059# 1060config CLONE_BACKWARDS 1061 bool 1062 help 1063 Architecture has tls passed as the 4th argument of clone(2), 1064 not the 5th one. 1065 1066config CLONE_BACKWARDS2 1067 bool 1068 help 1069 Architecture has the first two arguments of clone(2) swapped. 1070 1071config CLONE_BACKWARDS3 1072 bool 1073 help 1074 Architecture has tls passed as the 3rd argument of clone(2), 1075 not the 5th one. 1076 1077config ODD_RT_SIGACTION 1078 bool 1079 help 1080 Architecture has unusual rt_sigaction(2) arguments 1081 1082config OLD_SIGSUSPEND 1083 bool 1084 help 1085 Architecture has old sigsuspend(2) syscall, of one-argument variety 1086 1087config OLD_SIGSUSPEND3 1088 bool 1089 help 1090 Even weirder antique ABI - three-argument sigsuspend(2) 1091 1092config OLD_SIGACTION 1093 bool 1094 help 1095 Architecture has old sigaction(2) syscall. Nope, not the same 1096 as OLD_SIGSUSPEND | OLD_SIGSUSPEND3 - alpha has sigsuspend(2), 1097 but fairly different variant of sigaction(2), thanks to OSF/1 1098 compatibility... 1099 1100config COMPAT_OLD_SIGACTION 1101 bool 1102 1103config COMPAT_32BIT_TIME 1104 bool "Provide system calls for 32-bit time_t" 1105 default !64BIT || COMPAT 1106 help 1107 This enables 32 bit time_t support in addition to 64 bit time_t support. 1108 This is relevant on all 32-bit architectures, and 64-bit architectures 1109 as part of compat syscall handling. 1110 1111config ARCH_NO_PREEMPT 1112 bool 1113 1114config ARCH_EPHEMERAL_INODES 1115 def_bool n 1116 help 1117 An arch should select this symbol if it doesn't keep track of inode 1118 instances on its own, but instead relies on something else (e.g. the 1119 host kernel for an UML kernel). 1120 1121config ARCH_SUPPORTS_RT 1122 bool 1123 1124config CPU_NO_EFFICIENT_FFS 1125 def_bool n 1126 1127config HAVE_ARCH_VMAP_STACK 1128 def_bool n 1129 help 1130 An arch should select this symbol if it can support kernel stacks 1131 in vmalloc space. This means: 1132 1133 - vmalloc space must be large enough to hold many kernel stacks. 1134 This may rule out many 32-bit architectures. 1135 1136 - Stacks in vmalloc space need to work reliably. For example, if 1137 vmap page tables are created on demand, either this mechanism 1138 needs to work while the stack points to a virtual address with 1139 unpopulated page tables or arch code (switch_to() and switch_mm(), 1140 most likely) needs to ensure that the stack's page table entries 1141 are populated before running on a possibly unpopulated stack. 1142 1143 - If the stack overflows into a guard page, something reasonable 1144 should happen. The definition of "reasonable" is flexible, but 1145 instantly rebooting without logging anything would be unfriendly. 1146 1147config VMAP_STACK 1148 default y 1149 bool "Use a virtually-mapped stack" 1150 depends on HAVE_ARCH_VMAP_STACK 1151 depends on !KASAN || KASAN_HW_TAGS || KASAN_VMALLOC 1152 help 1153 Enable this if you want the use virtually-mapped kernel stacks 1154 with guard pages. This causes kernel stack overflows to be 1155 caught immediately rather than causing difficult-to-diagnose 1156 corruption. 1157 1158 To use this with software KASAN modes, the architecture must support 1159 backing virtual mappings with real shadow memory, and KASAN_VMALLOC 1160 must be enabled. 1161 1162config HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET 1163 def_bool n 1164 help 1165 An arch should select this symbol if it can support kernel stack 1166 offset randomization with calls to add_random_kstack_offset() 1167 during syscall entry and choose_random_kstack_offset() during 1168 syscall exit. Careful removal of -fstack-protector-strong and 1169 -fstack-protector should also be applied to the entry code and 1170 closely examined, as the artificial stack bump looks like an array 1171 to the compiler, so it will attempt to add canary checks regardless 1172 of the static branch state. 1173 1174config RANDOMIZE_KSTACK_OFFSET 1175 bool "Support for randomizing kernel stack offset on syscall entry" if EXPERT 1176 default y 1177 depends on HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET 1178 depends on INIT_STACK_NONE || !CC_IS_CLANG || CLANG_VERSION >= 140000 1179 help 1180 The kernel stack offset can be randomized (after pt_regs) by 1181 roughly 5 bits of entropy, frustrating memory corruption 1182 attacks that depend on stack address determinism or 1183 cross-syscall address exposures. 1184 1185 The feature is controlled via the "randomize_kstack_offset=on/off" 1186 kernel boot param, and if turned off has zero overhead due to its use 1187 of static branches (see JUMP_LABEL). 1188 1189 If unsure, say Y. 1190 1191config RANDOMIZE_KSTACK_OFFSET_DEFAULT 1192 bool "Default state of kernel stack offset randomization" 1193 depends on RANDOMIZE_KSTACK_OFFSET 1194 help 1195 Kernel stack offset randomization is controlled by kernel boot param 1196 "randomize_kstack_offset=on/off", and this config chooses the default 1197 boot state. 1198 1199config ARCH_OPTIONAL_KERNEL_RWX 1200 def_bool n 1201 1202config ARCH_OPTIONAL_KERNEL_RWX_DEFAULT 1203 def_bool n 1204 1205config ARCH_HAS_STRICT_KERNEL_RWX 1206 def_bool n 1207 1208config STRICT_KERNEL_RWX 1209 bool "Make kernel text and rodata read-only" if ARCH_OPTIONAL_KERNEL_RWX 1210 depends on ARCH_HAS_STRICT_KERNEL_RWX 1211 default !ARCH_OPTIONAL_KERNEL_RWX || ARCH_OPTIONAL_KERNEL_RWX_DEFAULT 1212 help 1213 If this is set, kernel text and rodata memory will be made read-only, 1214 and non-text memory will be made non-executable. This provides 1215 protection against certain security exploits (e.g. executing the heap 1216 or modifying text) 1217 1218 These features are considered standard security practice these days. 1219 You should say Y here in almost all cases. 1220 1221config ARCH_HAS_STRICT_MODULE_RWX 1222 def_bool n 1223 1224config STRICT_MODULE_RWX 1225 bool "Set loadable kernel module data as NX and text as RO" if ARCH_OPTIONAL_KERNEL_RWX 1226 depends on ARCH_HAS_STRICT_MODULE_RWX && MODULES 1227 default !ARCH_OPTIONAL_KERNEL_RWX || ARCH_OPTIONAL_KERNEL_RWX_DEFAULT 1228 help 1229 If this is set, module text and rodata memory will be made read-only, 1230 and non-text memory will be made non-executable. This provides 1231 protection against certain security exploits (e.g. writing to text) 1232 1233# select if the architecture provides an asm/dma-direct.h header 1234config ARCH_HAS_PHYS_TO_DMA 1235 bool 1236 1237config HAVE_ARCH_COMPILER_H 1238 bool 1239 help 1240 An architecture can select this if it provides an 1241 asm/compiler.h header that should be included after 1242 linux/compiler-*.h in order to override macro definitions that those 1243 headers generally provide. 1244 1245config HAVE_ARCH_PREL32_RELOCATIONS 1246 bool 1247 help 1248 May be selected by an architecture if it supports place-relative 1249 32-bit relocations, both in the toolchain and in the module loader, 1250 in which case relative references can be used in special sections 1251 for PCI fixup, initcalls etc which are only half the size on 64 bit 1252 architectures, and don't require runtime relocation on relocatable 1253 kernels. 1254 1255config ARCH_USE_MEMREMAP_PROT 1256 bool 1257 1258config LOCK_EVENT_COUNTS 1259 bool "Locking event counts collection" 1260 depends on DEBUG_FS 1261 help 1262 Enable light-weight counting of various locking related events 1263 in the system with minimal performance impact. This reduces 1264 the chance of application behavior change because of timing 1265 differences. The counts are reported via debugfs. 1266 1267# Select if the architecture has support for applying RELR relocations. 1268config ARCH_HAS_RELR 1269 bool 1270 1271config RELR 1272 bool "Use RELR relocation packing" 1273 depends on ARCH_HAS_RELR && TOOLS_SUPPORT_RELR 1274 default y 1275 help 1276 Store the kernel's dynamic relocations in the RELR relocation packing 1277 format. Requires a compatible linker (LLD supports this feature), as 1278 well as compatible NM and OBJCOPY utilities (llvm-nm and llvm-objcopy 1279 are compatible). 1280 1281config ARCH_HAS_MEM_ENCRYPT 1282 bool 1283 1284config ARCH_HAS_CC_PLATFORM 1285 bool 1286 1287config HAVE_SPARSE_SYSCALL_NR 1288 bool 1289 help 1290 An architecture should select this if its syscall numbering is sparse 1291 to save space. For example, MIPS architecture has a syscall array with 1292 entries at 4000, 5000 and 6000 locations. This option turns on syscall 1293 related optimizations for a given architecture. 1294 1295config ARCH_HAS_VDSO_DATA 1296 bool 1297 1298config HAVE_STATIC_CALL 1299 bool 1300 1301config HAVE_STATIC_CALL_INLINE 1302 bool 1303 depends on HAVE_STATIC_CALL 1304 1305config HAVE_PREEMPT_DYNAMIC 1306 bool 1307 1308config HAVE_PREEMPT_DYNAMIC_CALL 1309 bool 1310 depends on HAVE_STATIC_CALL 1311 select HAVE_PREEMPT_DYNAMIC 1312 help 1313 An architecture should select this if it can handle the preemption 1314 model being selected at boot time using static calls. 1315 1316 Where an architecture selects HAVE_STATIC_CALL_INLINE, any call to a 1317 preemption function will be patched directly. 1318 1319 Where an architecture does not select HAVE_STATIC_CALL_INLINE, any 1320 call to a preemption function will go through a trampoline, and the 1321 trampoline will be patched. 1322 1323 It is strongly advised to support inline static call to avoid any 1324 overhead. 1325 1326config HAVE_PREEMPT_DYNAMIC_KEY 1327 bool 1328 depends on HAVE_ARCH_JUMP_LABEL && CC_HAS_ASM_GOTO 1329 select HAVE_PREEMPT_DYNAMIC 1330 help 1331 An architecture should select this if it can handle the preemption 1332 model being selected at boot time using static keys. 1333 1334 Each preemption function will be given an early return based on a 1335 static key. This should have slightly lower overhead than non-inline 1336 static calls, as this effectively inlines each trampoline into the 1337 start of its callee. This may avoid redundant work, and may 1338 integrate better with CFI schemes. 1339 1340 This will have greater overhead than using inline static calls as 1341 the call to the preemption function cannot be entirely elided. 1342 1343config ARCH_WANT_LD_ORPHAN_WARN 1344 bool 1345 help 1346 An arch should select this symbol once all linker sections are explicitly 1347 included, size-asserted, or discarded in the linker scripts. This is 1348 important because we never want expected sections to be placed heuristically 1349 by the linker, since the locations of such sections can change between linker 1350 versions. 1351 1352config HAVE_ARCH_PFN_VALID 1353 bool 1354 1355config ARCH_SUPPORTS_DEBUG_PAGEALLOC 1356 bool 1357 1358config ARCH_SUPPORTS_PAGE_TABLE_CHECK 1359 bool 1360 1361config ARCH_SPLIT_ARG64 1362 bool 1363 help 1364 If a 32-bit architecture requires 64-bit arguments to be split into 1365 pairs of 32-bit arguments, select this option. 1366 1367config ARCH_HAS_ELFCORE_COMPAT 1368 bool 1369 1370config ARCH_HAS_PARANOID_L1D_FLUSH 1371 bool 1372 1373config DYNAMIC_SIGFRAME 1374 bool 1375 1376# Select, if arch has a named attribute group bound to NUMA device nodes. 1377config HAVE_ARCH_NODE_DEV_GROUP 1378 bool 1379 1380source "kernel/gcov/Kconfig" 1381 1382source "scripts/gcc-plugins/Kconfig" 1383 1384endmenu 1385