1# SPDX-License-Identifier: GPL-2.0-only 2menu "Kernel hacking" 3 4menu "printk and dmesg options" 5 6config PRINTK_TIME 7 bool "Show timing information on printks" 8 depends on PRINTK 9 help 10 Selecting this option causes time stamps of the printk() 11 messages to be added to the output of the syslog() system 12 call and at the console. 13 14 The timestamp is always recorded internally, and exported 15 to /dev/kmsg. This flag just specifies if the timestamp should 16 be included, not that the timestamp is recorded. 17 18 The behavior is also controlled by the kernel command line 19 parameter printk.time=1. See Documentation/admin-guide/kernel-parameters.rst 20 21config PRINTK_CALLER 22 bool "Show caller information on printks" 23 depends on PRINTK 24 help 25 Selecting this option causes printk() to add a caller "thread id" (if 26 in task context) or a caller "processor id" (if not in task context) 27 to every message. 28 29 This option is intended for environments where multiple threads 30 concurrently call printk() for many times, for it is difficult to 31 interpret without knowing where these lines (or sometimes individual 32 line which was divided into multiple lines due to race) came from. 33 34 Since toggling after boot makes the code racy, currently there is 35 no option to enable/disable at the kernel command line parameter or 36 sysfs interface. 37 38config STACKTRACE_BUILD_ID 39 bool "Show build ID information in stacktraces" 40 depends on PRINTK 41 help 42 Selecting this option adds build ID information for symbols in 43 stacktraces printed with the printk format '%p[SR]b'. 44 45 This option is intended for distros where debuginfo is not easily 46 accessible but can be downloaded given the build ID of the vmlinux or 47 kernel module where the function is located. 48 49config CONSOLE_LOGLEVEL_DEFAULT 50 int "Default console loglevel (1-15)" 51 range 1 15 52 default "7" 53 help 54 Default loglevel to determine what will be printed on the console. 55 56 Setting a default here is equivalent to passing in loglevel=<x> in 57 the kernel bootargs. loglevel=<x> continues to override whatever 58 value is specified here as well. 59 60 Note: This does not affect the log level of un-prefixed printk() 61 usage in the kernel. That is controlled by the MESSAGE_LOGLEVEL_DEFAULT 62 option. 63 64config CONSOLE_LOGLEVEL_QUIET 65 int "quiet console loglevel (1-15)" 66 range 1 15 67 default "4" 68 help 69 loglevel to use when "quiet" is passed on the kernel commandline. 70 71 When "quiet" is passed on the kernel commandline this loglevel 72 will be used as the loglevel. IOW passing "quiet" will be the 73 equivalent of passing "loglevel=<CONSOLE_LOGLEVEL_QUIET>" 74 75config MESSAGE_LOGLEVEL_DEFAULT 76 int "Default message log level (1-7)" 77 range 1 7 78 default "4" 79 help 80 Default log level for printk statements with no specified priority. 81 82 This was hard-coded to KERN_WARNING since at least 2.6.10 but folks 83 that are auditing their logs closely may want to set it to a lower 84 priority. 85 86 Note: This does not affect what message level gets printed on the console 87 by default. To change that, use loglevel=<x> in the kernel bootargs, 88 or pick a different CONSOLE_LOGLEVEL_DEFAULT configuration value. 89 90config BOOT_PRINTK_DELAY 91 bool "Delay each boot printk message by N milliseconds" 92 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY 93 help 94 This build option allows you to read kernel boot messages 95 by inserting a short delay after each one. The delay is 96 specified in milliseconds on the kernel command line, 97 using "boot_delay=N". 98 99 It is likely that you would also need to use "lpj=M" to preset 100 the "loops per jiffy" value. 101 See a previous boot log for the "lpj" value to use for your 102 system, and then set "lpj=M" before setting "boot_delay=N". 103 NOTE: Using this option may adversely affect SMP systems. 104 I.e., processors other than the first one may not boot up. 105 BOOT_PRINTK_DELAY also may cause LOCKUP_DETECTOR to detect 106 what it believes to be lockup conditions. 107 108config DYNAMIC_DEBUG 109 bool "Enable dynamic printk() support" 110 default n 111 depends on PRINTK 112 depends on (DEBUG_FS || PROC_FS) 113 select DYNAMIC_DEBUG_CORE 114 help 115 116 Compiles debug level messages into the kernel, which would not 117 otherwise be available at runtime. These messages can then be 118 enabled/disabled based on various levels of scope - per source file, 119 function, module, format string, and line number. This mechanism 120 implicitly compiles in all pr_debug() and dev_dbg() calls, which 121 enlarges the kernel text size by about 2%. 122 123 If a source file is compiled with DEBUG flag set, any 124 pr_debug() calls in it are enabled by default, but can be 125 disabled at runtime as below. Note that DEBUG flag is 126 turned on by many CONFIG_*DEBUG* options. 127 128 Usage: 129 130 Dynamic debugging is controlled via the 'dynamic_debug/control' file, 131 which is contained in the 'debugfs' filesystem or procfs. 132 Thus, the debugfs or procfs filesystem must first be mounted before 133 making use of this feature. 134 We refer the control file as: <debugfs>/dynamic_debug/control. This 135 file contains a list of the debug statements that can be enabled. The 136 format for each line of the file is: 137 138 filename:lineno [module]function flags format 139 140 filename : source file of the debug statement 141 lineno : line number of the debug statement 142 module : module that contains the debug statement 143 function : function that contains the debug statement 144 flags : '=p' means the line is turned 'on' for printing 145 format : the format used for the debug statement 146 147 From a live system: 148 149 nullarbor:~ # cat <debugfs>/dynamic_debug/control 150 # filename:lineno [module]function flags format 151 fs/aio.c:222 [aio]__put_ioctx =_ "__put_ioctx:\040freeing\040%p\012" 152 fs/aio.c:248 [aio]ioctx_alloc =_ "ENOMEM:\040nr_events\040too\040high\012" 153 fs/aio.c:1770 [aio]sys_io_cancel =_ "calling\040cancel\012" 154 155 Example usage: 156 157 // enable the message at line 1603 of file svcsock.c 158 nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' > 159 <debugfs>/dynamic_debug/control 160 161 // enable all the messages in file svcsock.c 162 nullarbor:~ # echo -n 'file svcsock.c +p' > 163 <debugfs>/dynamic_debug/control 164 165 // enable all the messages in the NFS server module 166 nullarbor:~ # echo -n 'module nfsd +p' > 167 <debugfs>/dynamic_debug/control 168 169 // enable all 12 messages in the function svc_process() 170 nullarbor:~ # echo -n 'func svc_process +p' > 171 <debugfs>/dynamic_debug/control 172 173 // disable all 12 messages in the function svc_process() 174 nullarbor:~ # echo -n 'func svc_process -p' > 175 <debugfs>/dynamic_debug/control 176 177 See Documentation/admin-guide/dynamic-debug-howto.rst for additional 178 information. 179 180config DYNAMIC_DEBUG_CORE 181 bool "Enable core function of dynamic debug support" 182 depends on PRINTK 183 depends on (DEBUG_FS || PROC_FS) 184 help 185 Enable core functional support of dynamic debug. It is useful 186 when you want to tie dynamic debug to your kernel modules with 187 DYNAMIC_DEBUG_MODULE defined for each of them, especially for 188 the case of embedded system where the kernel image size is 189 sensitive for people. 190 191config SYMBOLIC_ERRNAME 192 bool "Support symbolic error names in printf" 193 default y if PRINTK 194 help 195 If you say Y here, the kernel's printf implementation will 196 be able to print symbolic error names such as ENOSPC instead 197 of the number 28. It makes the kernel image slightly larger 198 (about 3KB), but can make the kernel logs easier to read. 199 200config DEBUG_BUGVERBOSE 201 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT 202 depends on BUG && (GENERIC_BUG || HAVE_DEBUG_BUGVERBOSE) 203 default y 204 help 205 Say Y here to make BUG() panics output the file name and line number 206 of the BUG call as well as the EIP and oops trace. This aids 207 debugging but costs about 70-100K of memory. 208 209config DEBUG_BUGVERBOSE_DETAILED 210 bool "Verbose WARN_ON_ONCE() reporting (adds 100K)" if DEBUG_BUGVERBOSE 211 help 212 Say Y here to make WARN_ON_ONCE() output the condition string of the 213 warning, in addition to the file name and line number. 214 This helps debugging, but costs about 100K of memory. 215 216 Say N if unsure. 217 218 219endmenu # "printk and dmesg options" 220 221config DEBUG_KERNEL 222 bool "Kernel debugging" 223 help 224 Say Y here if you are developing drivers or trying to debug and 225 identify kernel problems. 226 227config DEBUG_MISC 228 bool "Miscellaneous debug code" 229 default DEBUG_KERNEL 230 depends on DEBUG_KERNEL 231 help 232 Say Y here if you need to enable miscellaneous debug code that should 233 be under a more specific debug option but isn't. 234 235menu "Compile-time checks and compiler options" 236 237config DEBUG_INFO 238 bool 239 help 240 A kernel debug info option other than "None" has been selected 241 in the "Debug information" choice below, indicating that debug 242 information will be generated for build targets. 243 244# Clang generates .uleb128 with label differences for DWARF v5, a feature that 245# older binutils ports do not support when utilizing RISC-V style linker 246# relaxation: https://sourceware.org/bugzilla/show_bug.cgi?id=27215 247config AS_HAS_NON_CONST_ULEB128 248 def_bool $(as-instr,.uleb128 .Lexpr_end4 - .Lexpr_start3\n.Lexpr_start3:\n.Lexpr_end4:) 249 250choice 251 prompt "Debug information" 252 depends on DEBUG_KERNEL 253 help 254 Selecting something other than "None" results in a kernel image 255 that will include debugging info resulting in a larger kernel image. 256 This adds debug symbols to the kernel and modules (gcc -g), and 257 is needed if you intend to use kernel crashdump or binary object 258 tools like crash, kgdb, LKCD, gdb, etc on the kernel. 259 260 Choose which version of DWARF debug info to emit. If unsure, 261 select "Toolchain default". 262 263config DEBUG_INFO_NONE 264 bool "Disable debug information" 265 help 266 Do not build the kernel with debugging information, which will 267 result in a faster and smaller build. 268 269config DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT 270 bool "Rely on the toolchain's implicit default DWARF version" 271 select DEBUG_INFO 272 depends on !CC_IS_CLANG || AS_IS_LLVM || (AS_IS_GNU && AS_VERSION >= 23502 && AS_HAS_NON_CONST_ULEB128) 273 help 274 The implicit default version of DWARF debug info produced by a 275 toolchain changes over time. 276 277 This can break consumers of the debug info that haven't upgraded to 278 support newer revisions, and prevent testing newer versions, but 279 those should be less common scenarios. 280 281config DEBUG_INFO_DWARF4 282 bool "Generate DWARF Version 4 debuginfo" 283 select DEBUG_INFO 284 depends on !CC_IS_CLANG || AS_IS_LLVM || (AS_IS_GNU && AS_VERSION >= 23502) 285 help 286 Generate DWARF v4 debug info. This requires gcc 4.5+, binutils 2.35.2 287 if using clang without clang's integrated assembler, and gdb 7.0+. 288 289 If you have consumers of DWARF debug info that are not ready for 290 newer revisions of DWARF, you may wish to choose this or have your 291 config select this. 292 293config DEBUG_INFO_DWARF5 294 bool "Generate DWARF Version 5 debuginfo" 295 select DEBUG_INFO 296 depends on !ARCH_HAS_BROKEN_DWARF5 297 depends on !CC_IS_CLANG || AS_IS_LLVM || (AS_IS_GNU && AS_VERSION >= 23502 && AS_HAS_NON_CONST_ULEB128) 298 help 299 Generate DWARF v5 debug info. Requires binutils 2.35.2, gcc 5.0+ (gcc 300 5.0+ accepts the -gdwarf-5 flag but only had partial support for some 301 draft features until 7.0), and gdb 8.0+. 302 303 Changes to the structure of debug info in Version 5 allow for around 304 15-18% savings in resulting image and debug info section sizes as 305 compared to DWARF Version 4. DWARF Version 5 standardizes previous 306 extensions such as accelerators for symbol indexing and the format 307 for fission (.dwo/.dwp) files. Users may not want to select this 308 config if they rely on tooling that has not yet been updated to 309 support DWARF Version 5. 310 311endchoice # "Debug information" 312 313if DEBUG_INFO 314 315config DEBUG_INFO_REDUCED 316 bool "Reduce debugging information" 317 help 318 If you say Y here gcc is instructed to generate less debugging 319 information for structure types. This means that tools that 320 need full debugging information (like kgdb or systemtap) won't 321 be happy. But if you merely need debugging information to 322 resolve line numbers there is no loss. Advantage is that 323 build directory object sizes shrink dramatically over a full 324 DEBUG_INFO build and compile times are reduced too. 325 Only works with newer gcc versions. 326 327choice 328 prompt "Compressed Debug information" 329 help 330 Compress the resulting debug info. Results in smaller debug info sections, 331 but requires that consumers are able to decompress the results. 332 333 If unsure, choose DEBUG_INFO_COMPRESSED_NONE. 334 335config DEBUG_INFO_COMPRESSED_NONE 336 bool "Don't compress debug information" 337 help 338 Don't compress debug info sections. 339 340config DEBUG_INFO_COMPRESSED_ZLIB 341 bool "Compress debugging information with zlib" 342 depends on $(cc-option,-gz=zlib) 343 depends on $(ld-option,--compress-debug-sections=zlib) 344 help 345 Compress the debug information using zlib. 346 347 Users of dpkg-deb via debian/rules may find an increase in 348 size of their debug .deb packages with this config set, due to the 349 debug info being compressed with zlib, then the object files being 350 recompressed with a different compression scheme. But this is still 351 preferable to setting KDEB_COMPRESS or DPKG_DEB_COMPRESSOR_TYPE to 352 "none" which would be even larger. 353 354config DEBUG_INFO_COMPRESSED_ZSTD 355 bool "Compress debugging information with zstd" 356 depends on $(cc-option,-gz=zstd) 357 depends on $(ld-option,--compress-debug-sections=zstd) 358 help 359 Compress the debug information using zstd. This may provide better 360 compression than zlib, for about the same time costs, but requires newer 361 toolchain support. Requires GCC 13.0+ or Clang 16.0+, binutils 2.40+, and 362 zstd. 363 364endchoice # "Compressed Debug information" 365 366config DEBUG_INFO_SPLIT 367 bool "Produce split debuginfo in .dwo files" 368 depends on $(cc-option,-gsplit-dwarf) 369 # RISC-V linker relaxation + -gsplit-dwarf has issues with LLVM and GCC 370 # prior to 12.x: 371 # https://github.com/llvm/llvm-project/issues/56642 372 # https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99090 373 depends on !RISCV || GCC_VERSION >= 120000 374 help 375 Generate debug info into separate .dwo files. This significantly 376 reduces the build directory size for builds with DEBUG_INFO, 377 because it stores the information only once on disk in .dwo 378 files instead of multiple times in object files and executables. 379 In addition the debug information is also compressed. 380 381 Requires recent gcc (4.7+) and recent gdb/binutils. 382 Any tool that packages or reads debug information would need 383 to know about the .dwo files and include them. 384 Incompatible with older versions of ccache. 385 386config DEBUG_INFO_BTF 387 bool "Generate BTF type information" 388 depends on !DEBUG_INFO_SPLIT && !DEBUG_INFO_REDUCED 389 depends on !GCC_PLUGIN_RANDSTRUCT || COMPILE_TEST 390 depends on BPF_SYSCALL 391 depends on PAHOLE_VERSION >= 116 392 depends on DEBUG_INFO_DWARF4 || PAHOLE_VERSION >= 121 393 # pahole uses elfutils, which does not have support for Hexagon relocations 394 depends on !HEXAGON 395 help 396 Generate deduplicated BTF type information from DWARF debug info. 397 Turning this on requires pahole v1.16 or later (v1.21 or later to 398 support DWARF 5), which will convert DWARF type info into equivalent 399 deduplicated BTF type info. 400 401config PAHOLE_HAS_SPLIT_BTF 402 def_bool PAHOLE_VERSION >= 119 403 404config PAHOLE_HAS_BTF_TAG 405 def_bool PAHOLE_VERSION >= 123 406 depends on CC_IS_CLANG 407 help 408 Decide whether pahole emits btf_tag attributes (btf_type_tag and 409 btf_decl_tag) or not. Currently only clang compiler implements 410 these attributes, so make the config depend on CC_IS_CLANG. 411 412config PAHOLE_HAS_LANG_EXCLUDE 413 def_bool PAHOLE_VERSION >= 124 414 help 415 Support for the --lang_exclude flag which makes pahole exclude 416 compilation units from the supplied language. Used in Kbuild to 417 omit Rust CUs which are not supported in version 1.24 of pahole, 418 otherwise it would emit malformed kernel and module binaries when 419 using DEBUG_INFO_BTF_MODULES. 420 421config DEBUG_INFO_BTF_MODULES 422 bool "Generate BTF type information for kernel modules" 423 default y 424 depends on DEBUG_INFO_BTF && MODULES && PAHOLE_HAS_SPLIT_BTF 425 help 426 Generate compact split BTF type information for kernel modules. 427 428config MODULE_ALLOW_BTF_MISMATCH 429 bool "Allow loading modules with non-matching BTF type info" 430 depends on DEBUG_INFO_BTF_MODULES 431 help 432 For modules whose split BTF does not match vmlinux, load without 433 BTF rather than refusing to load. The default behavior with 434 module BTF enabled is to reject modules with such mismatches; 435 this option will still load module BTF where possible but ignore 436 it when a mismatch is found. 437 438config GDB_SCRIPTS 439 bool "Provide GDB scripts for kernel debugging" 440 help 441 This creates the required links to GDB helper scripts in the 442 build directory. If you load vmlinux into gdb, the helper 443 scripts will be automatically imported by gdb as well, and 444 additional functions are available to analyze a Linux kernel 445 instance. See Documentation/process/debugging/gdb-kernel-debugging.rst 446 for further details. 447 448endif # DEBUG_INFO 449 450config FRAME_WARN 451 int "Warn for stack frames larger than" 452 range 0 8192 453 default 0 if KMSAN 454 default 2048 if GCC_PLUGIN_LATENT_ENTROPY 455 default 2048 if PARISC 456 default 1536 if (!64BIT && XTENSA) 457 default 1280 if !64BIT 458 default 2048 if 64BIT 459 help 460 Tell the compiler to warn at build time for stack frames larger than this. 461 Setting this too low will cause a lot of warnings. 462 Setting it to 0 disables the warning. 463 464config STRIP_ASM_SYMS 465 bool "Strip assembler-generated symbols during link" 466 default n 467 help 468 Strip internal assembler-generated symbols during a link (symbols 469 that look like '.Lxxx') so they don't pollute the output of 470 get_wchan() and suchlike. 471 472config READABLE_ASM 473 bool "Generate readable assembler code" 474 depends on DEBUG_KERNEL 475 depends on CC_IS_GCC 476 help 477 Disable some compiler optimizations that tend to generate human unreadable 478 assembler output. This may make the kernel slightly slower, but it helps 479 to keep kernel developers who have to stare a lot at assembler listings 480 sane. 481 482config HEADERS_INSTALL 483 bool "Install uapi headers to usr/include" 484 help 485 This option will install uapi headers (headers exported to user-space) 486 into the usr/include directory for use during the kernel build. 487 This is unneeded for building the kernel itself, but needed for some 488 user-space program samples. It is also needed by some features such 489 as uapi header sanity checks. 490 491config DEBUG_SECTION_MISMATCH 492 bool "Enable full Section mismatch analysis" 493 depends on CC_IS_GCC 494 help 495 The section mismatch analysis checks if there are illegal references 496 from one section to another. During linktime or runtime, some 497 sections are dropped; any use of code/data previously in these 498 sections would most likely result in an oops. 499 500 In the code, functions and variables are annotated with __init, 501 __initdata, and so on (see the full list in include/linux/init.h). 502 This directs the toolchain to place code/data in specific sections. 503 504 The section mismatch analysis is always performed after a full 505 kernel build, and enabling this option causes the option 506 -fno-inline-functions-called-once to be added to gcc commands. 507 508 However, when inlining a function annotated with __init in 509 a non-init function, we would lose the section information and thus 510 the analysis would not catch the illegal reference. This option 511 tells gcc to inline less (but it does result in a larger kernel). 512 513config SECTION_MISMATCH_WARN_ONLY 514 bool "Make section mismatch errors non-fatal" 515 default y 516 help 517 If you say N here, the build process will fail if there are any 518 section mismatch, instead of just throwing warnings. 519 520 If unsure, say Y. 521 522config DEBUG_FORCE_FUNCTION_ALIGN_64B 523 bool "Force all function address 64B aligned" 524 depends on EXPERT && (X86_64 || ARM64 || PPC32 || PPC64 || ARC || RISCV || S390) 525 select FUNCTION_ALIGNMENT_64B 526 help 527 There are cases that a commit from one domain changes the function 528 address alignment of other domains, and cause magic performance 529 bump (regression or improvement). Enable this option will help to 530 verify if the bump is caused by function alignment changes, while 531 it will slightly increase the kernel size and affect icache usage. 532 533 It is mainly for debug and performance tuning use. 534 535# 536# Select this config option from the architecture Kconfig, if it 537# is preferred to always offer frame pointers as a config 538# option on the architecture (regardless of KERNEL_DEBUG): 539# 540config ARCH_WANT_FRAME_POINTERS 541 bool 542 543config FRAME_POINTER 544 bool "Compile the kernel with frame pointers" 545 depends on DEBUG_KERNEL && (M68K || UML || SUPERH) || ARCH_WANT_FRAME_POINTERS 546 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS 547 help 548 If you say Y here the resulting kernel image will be slightly 549 larger and slower, but it gives very useful debugging information 550 in case of kernel bugs. (precise oopses/stacktraces/warnings) 551 552config OBJTOOL 553 bool 554 555config OBJTOOL_WERROR 556 bool "Upgrade objtool warnings to errors" 557 depends on OBJTOOL && !COMPILE_TEST 558 help 559 Fail the build on objtool warnings. 560 561 Objtool warnings can indicate kernel instability, including boot 562 failures. This option is highly recommended. 563 564 If unsure, say Y. 565 566config STACK_VALIDATION 567 bool "Compile-time stack metadata validation" 568 depends on HAVE_STACK_VALIDATION && UNWINDER_FRAME_POINTER 569 select OBJTOOL 570 default n 571 help 572 Validate frame pointer rules at compile-time. This helps ensure that 573 runtime stack traces are more reliable. 574 575 For more information, see 576 tools/objtool/Documentation/objtool.txt. 577 578config NOINSTR_VALIDATION 579 bool 580 depends on HAVE_NOINSTR_VALIDATION && DEBUG_ENTRY 581 select OBJTOOL 582 default y 583 584config VMLINUX_MAP 585 bool "Generate vmlinux.map file when linking" 586 depends on EXPERT 587 help 588 Selecting this option will pass "-Map=vmlinux.map" to ld 589 when linking vmlinux. That file can be useful for verifying 590 and debugging magic section games, and for seeing which 591 pieces of code get eliminated with 592 CONFIG_LD_DEAD_CODE_DATA_ELIMINATION. 593 594config BUILTIN_MODULE_RANGES 595 bool "Generate address range information for builtin modules" 596 depends on !LTO 597 depends on VMLINUX_MAP 598 help 599 When modules are built into the kernel, there will be no module name 600 associated with its symbols in /proc/kallsyms. Tracers may want to 601 identify symbols by module name and symbol name regardless of whether 602 the module is configured as loadable or not. 603 604 This option generates modules.builtin.ranges in the build tree with 605 offset ranges (per ELF section) for the module(s) they belong to. 606 It also records an anchor symbol to determine the load address of the 607 section. 608 609config DEBUG_FORCE_WEAK_PER_CPU 610 bool "Force weak per-cpu definitions" 611 depends on DEBUG_KERNEL 612 help 613 s390 and alpha require percpu variables in modules to be 614 defined weak to work around addressing range issue which 615 puts the following two restrictions on percpu variable 616 definitions. 617 618 1. percpu symbols must be unique whether static or not 619 2. percpu variables can't be defined inside a function 620 621 To ensure that generic code follows the above rules, this 622 option forces all percpu variables to be defined as weak. 623 624config WARN_CONTEXT_ANALYSIS 625 bool "Compiler context-analysis warnings" 626 depends on CC_IS_CLANG && CLANG_VERSION >= 220000 627 # Branch profiling re-defines "if", which messes with the compiler's 628 # ability to analyze __cond_acquires(..), resulting in false positives. 629 depends on !TRACE_BRANCH_PROFILING 630 default y 631 help 632 Context Analysis is a language extension, which enables statically 633 checking that required contexts are active (or inactive) by acquiring 634 and releasing user-definable "context locks". 635 636 Clang's name of the feature is "Thread Safety Analysis". Requires 637 Clang 22 or later. 638 639 Produces warnings by default. Select CONFIG_WERROR if you wish to 640 turn these warnings into errors. 641 642 For more details, see Documentation/dev-tools/context-analysis.rst. 643 644config WARN_CONTEXT_ANALYSIS_ALL 645 bool "Enable context analysis for all source files" 646 depends on WARN_CONTEXT_ANALYSIS 647 depends on EXPERT && !COMPILE_TEST 648 help 649 Enable tree-wide context analysis. This is likely to produce a 650 large number of false positives - enable at your own risk. 651 652 If unsure, say N. 653 654endmenu # "Compiler options" 655 656menu "Generic Kernel Debugging Instruments" 657 658config MAGIC_SYSRQ 659 bool "Magic SysRq key" 660 depends on !UML 661 help 662 If you say Y here, you will have some control over the system even 663 if the system crashes for example during kernel debugging (e.g., you 664 will be able to flush the buffer cache to disk, reboot the system 665 immediately or dump some status information). This is accomplished 666 by pressing various keys while holding SysRq (Alt+PrintScreen). It 667 also works on a serial console (on PC hardware at least), if you 668 send a BREAK and then within 5 seconds a command keypress. The 669 keys are documented in <file:Documentation/admin-guide/sysrq.rst>. 670 Don't say Y unless you really know what this hack does. 671 672config MAGIC_SYSRQ_DEFAULT_ENABLE 673 hex "Enable magic SysRq key functions by default" 674 depends on MAGIC_SYSRQ 675 default 0x1 676 help 677 Specifies which SysRq key functions are enabled by default. 678 This may be set to 1 or 0 to enable or disable them all, or 679 to a bitmask as described in Documentation/admin-guide/sysrq.rst. 680 681config MAGIC_SYSRQ_SERIAL 682 bool "Enable magic SysRq key over serial" 683 depends on MAGIC_SYSRQ 684 default y 685 help 686 Many embedded boards have a disconnected TTL level serial which can 687 generate some garbage that can lead to spurious false sysrq detects. 688 This option allows you to decide whether you want to enable the 689 magic SysRq key. 690 691config MAGIC_SYSRQ_SERIAL_SEQUENCE 692 string "Char sequence that enables magic SysRq over serial" 693 depends on MAGIC_SYSRQ_SERIAL 694 default "" 695 help 696 Specifies a sequence of characters that can follow BREAK to enable 697 SysRq on a serial console. 698 699 If unsure, leave an empty string and the option will not be enabled. 700 701config DEBUG_FS 702 bool "Debug Filesystem" 703 help 704 debugfs is a virtual file system that kernel developers use to put 705 debugging files into. Enable this option to be able to read and 706 write to these files. 707 708 For detailed documentation on the debugfs API, see 709 Documentation/filesystems/. 710 711 If unsure, say N. 712 713choice 714 prompt "Debugfs default access" 715 depends on DEBUG_FS 716 default DEBUG_FS_ALLOW_ALL 717 help 718 This selects the default access restrictions for debugfs. 719 It can be overridden with kernel command line option 720 debugfs=[on,off]. The restrictions apply for API access 721 and filesystem registration. 722 723config DEBUG_FS_ALLOW_ALL 724 bool "Access normal" 725 help 726 No restrictions apply. Both API and filesystem registration 727 is on. This is the normal default operation. 728 729config DEBUG_FS_ALLOW_NONE 730 bool "No access" 731 help 732 Access is off. Clients get -PERM when trying to create nodes in 733 debugfs tree and debugfs is not registered as a filesystem. 734 Client can then back-off or continue without debugfs access. 735 736endchoice 737 738source "lib/Kconfig.kgdb" 739source "lib/Kconfig.ubsan" 740source "lib/Kconfig.kcsan" 741 742endmenu 743 744menu "Networking Debugging" 745 746source "net/Kconfig.debug" 747 748endmenu # "Networking Debugging" 749 750menu "Memory Debugging" 751 752source "mm/Kconfig.debug" 753 754config DEBUG_OBJECTS 755 bool "Debug object operations" 756 depends on DEBUG_KERNEL 757 help 758 If you say Y here, additional code will be inserted into the 759 kernel to track the life time of various objects and validate 760 the operations on those objects. 761 762config DEBUG_OBJECTS_SELFTEST 763 bool "Debug objects selftest" 764 depends on DEBUG_OBJECTS 765 help 766 This enables the selftest of the object debug code. 767 768config DEBUG_OBJECTS_FREE 769 bool "Debug objects in freed memory" 770 depends on DEBUG_OBJECTS 771 help 772 This enables checks whether a k/v free operation frees an area 773 which contains an object which has not been deactivated 774 properly. This can make kmalloc/kfree-intensive workloads 775 much slower. 776 777config DEBUG_OBJECTS_TIMERS 778 bool "Debug timer objects" 779 depends on DEBUG_OBJECTS 780 help 781 If you say Y here, additional code will be inserted into the 782 timer routines to track the life time of timer objects and 783 validate the timer operations. 784 785config DEBUG_OBJECTS_WORK 786 bool "Debug work objects" 787 depends on DEBUG_OBJECTS 788 help 789 If you say Y here, additional code will be inserted into the 790 work queue routines to track the life time of work objects and 791 validate the work operations. 792 793config DEBUG_OBJECTS_RCU_HEAD 794 bool "Debug RCU callbacks objects" 795 depends on DEBUG_OBJECTS 796 help 797 Enable this to turn on debugging of RCU list heads (call_rcu() usage). 798 799config DEBUG_OBJECTS_PERCPU_COUNTER 800 bool "Debug percpu counter objects" 801 depends on DEBUG_OBJECTS 802 help 803 If you say Y here, additional code will be inserted into the 804 percpu counter routines to track the life time of percpu counter 805 objects and validate the percpu counter operations. 806 807config DEBUG_OBJECTS_ENABLE_DEFAULT 808 int "debug_objects bootup default value (0-1)" 809 range 0 1 810 default "1" 811 depends on DEBUG_OBJECTS 812 help 813 Debug objects boot parameter default value 814 815config SHRINKER_DEBUG 816 bool "Enable shrinker debugging support" 817 depends on DEBUG_FS 818 help 819 Say Y to enable the shrinker debugfs interface which provides 820 visibility into the kernel memory shrinkers subsystem. 821 Disable it to avoid an extra memory footprint. 822 823config DEBUG_STACK_USAGE 824 bool "Stack utilization instrumentation" 825 depends on DEBUG_KERNEL 826 help 827 Enables the display of the minimum amount of free stack which each 828 task has ever had available in the sysrq-T and sysrq-P debug output. 829 Also emits a message to dmesg when a process exits if that process 830 used more stack space than previously exiting processes. 831 832 This option will slow down process creation somewhat. 833 834config SCHED_STACK_END_CHECK 835 bool "Detect stack corruption on calls to schedule()" 836 depends on DEBUG_KERNEL 837 default n 838 help 839 This option checks for a stack overrun on calls to schedule(). 840 If the stack end location is found to be over written always panic as 841 the content of the corrupted region can no longer be trusted. 842 This is to ensure no erroneous behaviour occurs which could result in 843 data corruption or a sporadic crash at a later stage once the region 844 is examined. The runtime overhead introduced is minimal. 845 846config ARCH_HAS_DEBUG_VM_PGTABLE 847 bool 848 help 849 An architecture should select this when it can successfully 850 build and run DEBUG_VM_PGTABLE. 851 852config DEBUG_VFS 853 bool "Debug VFS" 854 depends on DEBUG_KERNEL 855 help 856 Enable this to turn on extended checks in the VFS layer that may impact 857 performance. 858 859 If unsure, say N. 860 861config DEBUG_VM_IRQSOFF 862 def_bool DEBUG_VM && !PREEMPT_RT 863 864config DEBUG_VM 865 bool "Debug VM" 866 depends on DEBUG_KERNEL 867 help 868 Enable this to turn on extended checks in the virtual-memory system 869 that may impact performance. 870 871 If unsure, say N. 872 873config DEBUG_VM_SHOOT_LAZIES 874 bool "Debug MMU_LAZY_TLB_SHOOTDOWN implementation" 875 depends on DEBUG_VM 876 depends on MMU_LAZY_TLB_SHOOTDOWN 877 help 878 Enable additional IPIs that ensure lazy tlb mm references are removed 879 before the mm is freed. 880 881 If unsure, say N. 882 883config DEBUG_VM_MAPLE_TREE 884 bool "Debug VM maple trees" 885 depends on DEBUG_VM 886 select DEBUG_MAPLE_TREE 887 help 888 Enable VM maple tree debugging information and extra validations. 889 890 If unsure, say N. 891 892config DEBUG_VM_RB 893 bool "Debug VM red-black trees" 894 depends on DEBUG_VM 895 help 896 Enable VM red-black tree debugging information and extra validations. 897 898 If unsure, say N. 899 900config DEBUG_VM_PGFLAGS 901 bool "Debug page-flags operations" 902 depends on DEBUG_VM 903 help 904 Enables extra validation on page flags operations. 905 906 If unsure, say N. 907 908config DEBUG_VM_PGTABLE 909 bool "Debug arch page table for semantics compliance" 910 depends on MMU 911 depends on ARCH_HAS_DEBUG_VM_PGTABLE 912 default y if DEBUG_VM 913 help 914 This option provides a debug method which can be used to test 915 architecture page table helper functions on various platforms in 916 verifying if they comply with expected generic MM semantics. This 917 will help architecture code in making sure that any changes or 918 new additions of these helpers still conform to expected 919 semantics of the generic MM. Platforms will have to opt in for 920 this through ARCH_HAS_DEBUG_VM_PGTABLE. 921 922 If unsure, say N. 923 924config ARCH_HAS_DEBUG_VIRTUAL 925 bool 926 927config DEBUG_VIRTUAL 928 bool "Debug VM translations" 929 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL 930 help 931 Enable some costly sanity checks in virtual to page code. This can 932 catch mistakes with virt_to_page() and friends. 933 934 If unsure, say N. 935 936config DEBUG_NOMMU_REGIONS 937 bool "Debug the global anon/private NOMMU mapping region tree" 938 depends on DEBUG_KERNEL && !MMU 939 help 940 This option causes the global tree of anonymous and private mapping 941 regions to be regularly checked for invalid topology. 942 943config DEBUG_MEMORY_INIT 944 bool "Debug memory initialisation" if EXPERT 945 default !EXPERT 946 help 947 Enable this for additional checks during memory initialisation. 948 The sanity checks verify aspects of the VM such as the memory model 949 and other information provided by the architecture. Verbose 950 information will be printed at KERN_DEBUG loglevel depending 951 on the mminit_loglevel= command-line option. 952 953 If unsure, say Y 954 955config MEMORY_NOTIFIER_ERROR_INJECT 956 tristate "Memory hotplug notifier error injection module" 957 depends on MEMORY_HOTPLUG && NOTIFIER_ERROR_INJECTION 958 help 959 This option provides the ability to inject artificial errors to 960 memory hotplug notifier chain callbacks. It is controlled through 961 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory 962 963 If the notifier call chain should be failed with some events 964 notified, write the error code to "actions/<notifier event>/error". 965 966 Example: Inject memory hotplug offline error (-12 == -ENOMEM) 967 968 # cd /sys/kernel/debug/notifier-error-inject/memory 969 # echo -12 > actions/MEM_GOING_OFFLINE/error 970 # echo offline > /sys/devices/system/memory/memoryXXX/state 971 bash: echo: write error: Cannot allocate memory 972 973 To compile this code as a module, choose M here: the module will 974 be called memory-notifier-error-inject. 975 976 If unsure, say N. 977 978config DEBUG_PER_CPU_MAPS 979 bool "Debug access to per_cpu maps" 980 depends on DEBUG_KERNEL 981 depends on SMP 982 help 983 Say Y to verify that the per_cpu map being accessed has 984 been set up. This adds a fair amount of code to kernel memory 985 and decreases performance. 986 987 Say N if unsure. 988 989config DEBUG_KMAP_LOCAL 990 bool "Debug kmap_local temporary mappings" 991 depends on DEBUG_KERNEL && KMAP_LOCAL 992 help 993 This option enables additional error checking for the kmap_local 994 infrastructure. Disable for production use. 995 996config ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP 997 bool 998 999config DEBUG_KMAP_LOCAL_FORCE_MAP 1000 bool "Enforce kmap_local temporary mappings" 1001 depends on DEBUG_KERNEL && ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP 1002 select KMAP_LOCAL 1003 select DEBUG_KMAP_LOCAL 1004 help 1005 This option enforces temporary mappings through the kmap_local 1006 mechanism for non-highmem pages and on non-highmem systems. 1007 Disable this for production systems! 1008 1009config DEBUG_HIGHMEM 1010 bool "Highmem debugging" 1011 depends on DEBUG_KERNEL && HIGHMEM 1012 select DEBUG_KMAP_LOCAL_FORCE_MAP if ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP 1013 select DEBUG_KMAP_LOCAL 1014 help 1015 This option enables additional error checking for high memory 1016 systems. Disable for production systems. 1017 1018config HAVE_DEBUG_STACKOVERFLOW 1019 bool 1020 1021config DEBUG_STACKOVERFLOW 1022 bool "Check for stack overflows" 1023 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW 1024 help 1025 Say Y here if you want to check for overflows of kernel, IRQ 1026 and exception stacks (if your architecture uses them). This 1027 option will show detailed messages if free stack space drops 1028 below a certain limit. 1029 1030 These kinds of bugs usually occur when call-chains in the 1031 kernel get too deep, especially when interrupts are 1032 involved. 1033 1034 Use this in cases where you see apparently random memory 1035 corruption, especially if it appears in 'struct thread_info' 1036 1037 If in doubt, say "N". 1038 1039config CODE_TAGGING 1040 bool 1041 select KALLSYMS 1042 1043config MEM_ALLOC_PROFILING 1044 bool "Enable memory allocation profiling" 1045 default n 1046 depends on MMU 1047 depends on PROC_FS 1048 depends on !DEBUG_FORCE_WEAK_PER_CPU 1049 select CODE_TAGGING 1050 select PAGE_EXTENSION 1051 select SLAB_OBJ_EXT 1052 help 1053 Track allocation source code and record total allocation size 1054 initiated at that code location. The mechanism can be used to track 1055 memory leaks with a low performance and memory impact. 1056 1057config MEM_ALLOC_PROFILING_ENABLED_BY_DEFAULT 1058 bool "Enable memory allocation profiling by default" 1059 default y 1060 depends on MEM_ALLOC_PROFILING 1061 1062config MEM_ALLOC_PROFILING_DEBUG 1063 bool "Memory allocation profiler debugging" 1064 default n 1065 depends on MEM_ALLOC_PROFILING 1066 select MEM_ALLOC_PROFILING_ENABLED_BY_DEFAULT 1067 help 1068 Adds warnings with helpful error messages for memory allocation 1069 profiling. 1070 1071source "lib/Kconfig.kasan" 1072source "lib/Kconfig.kfence" 1073source "lib/Kconfig.kmsan" 1074 1075endmenu # "Memory Debugging" 1076 1077config DEBUG_SHIRQ 1078 bool "Debug shared IRQ handlers" 1079 depends on DEBUG_KERNEL 1080 help 1081 Enable this to generate a spurious interrupt just before a shared 1082 interrupt handler is deregistered (generating one when registering 1083 is currently disabled). Drivers need to handle this correctly. Some 1084 don't and need to be caught. 1085 1086menu "Debug Oops, Lockups and Hangs" 1087 1088config PANIC_ON_OOPS 1089 bool "Panic on Oops" 1090 help 1091 Say Y here to enable the kernel to panic when it oopses. This 1092 has the same effect as setting oops=panic on the kernel command 1093 line. 1094 1095 This feature is useful to ensure that the kernel does not do 1096 anything erroneous after an oops which could result in data 1097 corruption or other issues. 1098 1099 Say N if unsure. 1100 1101config PANIC_TIMEOUT 1102 int "panic timeout" 1103 default 0 1104 help 1105 Set the timeout value (in seconds) until a reboot occurs when 1106 the kernel panics. If n = 0, then we wait forever. A timeout 1107 value n > 0 will wait n seconds before rebooting, while a timeout 1108 value n < 0 will reboot immediately. This setting can be overridden 1109 with the kernel command line option panic=, and from userspace via 1110 /proc/sys/kernel/panic. 1111 1112config LOCKUP_DETECTOR 1113 bool 1114 1115config SOFTLOCKUP_DETECTOR 1116 bool "Detect Soft Lockups" 1117 depends on DEBUG_KERNEL && !S390 1118 select LOCKUP_DETECTOR 1119 help 1120 Say Y here to enable the kernel to act as a watchdog to detect 1121 soft lockups. 1122 1123 Softlockups are bugs that cause the kernel to loop in kernel 1124 mode for more than 20 seconds, without giving other tasks a 1125 chance to run. The current stack trace is displayed upon 1126 detection and the system will stay locked up. 1127 1128config SOFTLOCKUP_DETECTOR_INTR_STORM 1129 bool "Detect Interrupt Storm in Soft Lockups" 1130 depends on SOFTLOCKUP_DETECTOR && IRQ_TIME_ACCOUNTING 1131 select GENERIC_IRQ_STAT_SNAPSHOT 1132 default y if NR_CPUS <= 128 1133 help 1134 Say Y here to enable the kernel to detect interrupt storm 1135 during "soft lockups". 1136 1137 "soft lockups" can be caused by a variety of reasons. If one is 1138 caused by an interrupt storm, then the storming interrupts will not 1139 be on the callstack. To detect this case, it is necessary to report 1140 the CPU stats and the interrupt counts during the "soft lockups". 1141 1142config BOOTPARAM_SOFTLOCKUP_PANIC 1143 bool "Panic (Reboot) On Soft Lockups" 1144 depends on SOFTLOCKUP_DETECTOR 1145 help 1146 Say Y here to enable the kernel to panic on "soft lockups", 1147 which are bugs that cause the kernel to loop in kernel 1148 mode for more than 20 seconds (configurable using the watchdog_thresh 1149 sysctl), without giving other tasks a chance to run. 1150 1151 The panic can be used in combination with panic_timeout, 1152 to cause the system to reboot automatically after a 1153 lockup has been detected. This feature is useful for 1154 high-availability systems that have uptime guarantees and 1155 where a lockup must be resolved ASAP. 1156 1157 Say N if unsure. 1158 1159config HAVE_HARDLOCKUP_DETECTOR_BUDDY 1160 bool 1161 depends on SMP 1162 default y 1163 1164# 1165# Global switch whether to build a hardlockup detector at all. It is available 1166# only when the architecture supports at least one implementation. There are 1167# two exceptions. The hardlockup detector is never enabled on: 1168# 1169# s390: it reported many false positives there 1170# 1171# sparc64: has a custom implementation which is not using the common 1172# hardlockup command line options and sysctl interface. 1173# 1174config HARDLOCKUP_DETECTOR 1175 bool "Detect Hard Lockups" 1176 depends on DEBUG_KERNEL && !S390 && !HARDLOCKUP_DETECTOR_SPARC64 1177 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_BUDDY || HAVE_HARDLOCKUP_DETECTOR_ARCH 1178 imply HARDLOCKUP_DETECTOR_PERF 1179 imply HARDLOCKUP_DETECTOR_BUDDY 1180 imply HARDLOCKUP_DETECTOR_ARCH 1181 select LOCKUP_DETECTOR 1182 1183 help 1184 Say Y here to enable the kernel to act as a watchdog to detect 1185 hard lockups. 1186 1187 Hardlockups are bugs that cause the CPU to loop in kernel mode 1188 for more than 10 seconds, without letting other interrupts have a 1189 chance to run. The current stack trace is displayed upon detection 1190 and the system will stay locked up. 1191 1192# 1193# Note that arch-specific variants are always preferred. 1194# 1195config HARDLOCKUP_DETECTOR_PREFER_BUDDY 1196 bool "Prefer the buddy CPU hardlockup detector" 1197 depends on HARDLOCKUP_DETECTOR 1198 depends on HAVE_HARDLOCKUP_DETECTOR_PERF && HAVE_HARDLOCKUP_DETECTOR_BUDDY 1199 depends on !HAVE_HARDLOCKUP_DETECTOR_ARCH 1200 help 1201 Say Y here to prefer the buddy hardlockup detector over the perf one. 1202 1203 With the buddy detector, each CPU uses its softlockup hrtimer 1204 to check that the next CPU is processing hrtimer interrupts by 1205 verifying that a counter is increasing. 1206 1207 This hardlockup detector is useful on systems that don't have 1208 an arch-specific hardlockup detector or if resources needed 1209 for the hardlockup detector are better used for other things. 1210 1211config HARDLOCKUP_DETECTOR_PERF 1212 bool 1213 depends on HARDLOCKUP_DETECTOR 1214 depends on HAVE_HARDLOCKUP_DETECTOR_PERF && !HARDLOCKUP_DETECTOR_PREFER_BUDDY 1215 depends on !HAVE_HARDLOCKUP_DETECTOR_ARCH 1216 select HARDLOCKUP_DETECTOR_COUNTS_HRTIMER 1217 1218config HARDLOCKUP_DETECTOR_BUDDY 1219 bool 1220 depends on HARDLOCKUP_DETECTOR 1221 depends on HAVE_HARDLOCKUP_DETECTOR_BUDDY 1222 depends on !HAVE_HARDLOCKUP_DETECTOR_PERF || HARDLOCKUP_DETECTOR_PREFER_BUDDY 1223 depends on !HAVE_HARDLOCKUP_DETECTOR_ARCH 1224 select HARDLOCKUP_DETECTOR_COUNTS_HRTIMER 1225 1226config HARDLOCKUP_DETECTOR_ARCH 1227 bool 1228 depends on HARDLOCKUP_DETECTOR 1229 depends on HAVE_HARDLOCKUP_DETECTOR_ARCH 1230 help 1231 The arch-specific implementation of the hardlockup detector will 1232 be used. 1233 1234# 1235# Both the "perf" and "buddy" hardlockup detectors count hrtimer 1236# interrupts. This config enables functions managing this common code. 1237# 1238config HARDLOCKUP_DETECTOR_COUNTS_HRTIMER 1239 bool 1240 select SOFTLOCKUP_DETECTOR 1241 1242# 1243# Enables a timestamp based low pass filter to compensate for perf based 1244# hard lockup detection which runs too fast due to turbo modes. 1245# 1246config HARDLOCKUP_CHECK_TIMESTAMP 1247 bool 1248 1249config BOOTPARAM_HARDLOCKUP_PANIC 1250 bool "Panic (Reboot) On Hard Lockups" 1251 depends on HARDLOCKUP_DETECTOR 1252 help 1253 Say Y here to enable the kernel to panic on "hard lockups", 1254 which are bugs that cause the kernel to loop in kernel 1255 mode with interrupts disabled for more than 10 seconds (configurable 1256 using the watchdog_thresh sysctl). 1257 1258 Say N if unsure. 1259 1260config DETECT_HUNG_TASK 1261 bool "Detect Hung Tasks" 1262 depends on DEBUG_KERNEL 1263 default SOFTLOCKUP_DETECTOR 1264 help 1265 Say Y here to enable the kernel to detect "hung tasks", 1266 which are bugs that cause the task to be stuck in 1267 uninterruptible "D" state indefinitely. 1268 1269 When a hung task is detected, the kernel will print the 1270 current stack trace (which you should report), but the 1271 task will stay in uninterruptible state. If lockdep is 1272 enabled then all held locks will also be reported. This 1273 feature has negligible overhead. 1274 1275config DEFAULT_HUNG_TASK_TIMEOUT 1276 int "Default timeout for hung task detection (in seconds)" 1277 depends on DETECT_HUNG_TASK 1278 default 120 1279 help 1280 This option controls the default timeout (in seconds) used 1281 to determine when a task has become non-responsive and should 1282 be considered hung. 1283 1284 It can be adjusted at runtime via the kernel.hung_task_timeout_secs 1285 sysctl or by writing a value to 1286 /proc/sys/kernel/hung_task_timeout_secs. 1287 1288 A timeout of 0 disables the check. The default is two minutes. 1289 Keeping the default should be fine in most cases. 1290 1291config BOOTPARAM_HUNG_TASK_PANIC 1292 int "Number of hung tasks to trigger kernel panic" 1293 depends on DETECT_HUNG_TASK 1294 default 0 1295 help 1296 When set to a non-zero value, a kernel panic will be triggered 1297 if the number of hung tasks found during a single scan reaches 1298 this value. 1299 1300 The panic can be used in combination with panic_timeout, 1301 to cause the system to reboot automatically after a 1302 hung task has been detected. This feature is useful for 1303 high-availability systems that have uptime guarantees and 1304 where a hung tasks must be resolved ASAP. 1305 1306 Say N if unsure. 1307 1308config DETECT_HUNG_TASK_BLOCKER 1309 bool "Dump Hung Tasks Blocker" 1310 depends on DETECT_HUNG_TASK 1311 depends on !PREEMPT_RT 1312 default y 1313 help 1314 Say Y here to show the blocker task's stacktrace who acquires 1315 the mutex lock which "hung tasks" are waiting. 1316 This will add overhead a bit but shows suspicious tasks and 1317 call trace if it comes from waiting a mutex. 1318 1319config WQ_WATCHDOG 1320 bool "Detect Workqueue Stalls" 1321 depends on DEBUG_KERNEL 1322 help 1323 Say Y here to enable stall detection on workqueues. If a 1324 worker pool doesn't make forward progress on a pending work 1325 item for over a given amount of time, 30s by default, a 1326 warning message is printed along with dump of workqueue 1327 state. This can be configured through kernel parameter 1328 "workqueue.watchdog_thresh" and its sysfs counterpart. 1329 1330config WQ_CPU_INTENSIVE_REPORT 1331 bool "Report per-cpu work items which hog CPU for too long" 1332 depends on DEBUG_KERNEL 1333 help 1334 Say Y here to enable reporting of concurrency-managed per-cpu work 1335 items that hog CPUs for longer than 1336 workqueue.cpu_intensive_thresh_us. Workqueue automatically 1337 detects and excludes them from concurrency management to prevent 1338 them from stalling other per-cpu work items. Occassional 1339 triggering may not necessarily indicate a problem. Repeated 1340 triggering likely indicates that the work item should be switched 1341 to use an unbound workqueue. 1342 1343config TEST_LOCKUP 1344 tristate "Test module to generate lockups" 1345 depends on m 1346 help 1347 This builds the "test_lockup" module that helps to make sure 1348 that watchdogs and lockup detectors are working properly. 1349 1350 Depending on module parameters it could emulate soft or hard 1351 lockup, "hung task", or locking arbitrary lock for a long time. 1352 Also it could generate series of lockups with cooling-down periods. 1353 1354 If unsure, say N. 1355 1356endmenu # "Debug lockups and hangs" 1357 1358menu "Scheduler Debugging" 1359 1360config SCHED_INFO 1361 bool 1362 default n 1363 1364config SCHEDSTATS 1365 bool "Collect scheduler statistics" 1366 depends on PROC_FS 1367 select SCHED_INFO 1368 help 1369 If you say Y here, additional code will be inserted into the 1370 scheduler and related routines to collect statistics about 1371 scheduler behavior and provide them in /proc/schedstat. These 1372 stats may be useful for both tuning and debugging the scheduler 1373 If you aren't debugging the scheduler or trying to tune a specific 1374 application, you can say N to avoid the very slight overhead 1375 this adds. 1376 1377endmenu 1378 1379config DEBUG_PREEMPT 1380 bool "Debug preemptible kernel" 1381 depends on DEBUG_KERNEL && PREEMPTION && TRACE_IRQFLAGS_SUPPORT 1382 help 1383 If you say Y here then the kernel will use a debug variant of the 1384 commonly used smp_processor_id() function and will print warnings 1385 if kernel code uses it in a preemption-unsafe way. Also, the kernel 1386 will detect preemption count underflows. 1387 1388 This option has potential to introduce high runtime overhead, 1389 depending on workload as it triggers debugging routines for each 1390 this_cpu operation. It should only be used for debugging purposes. 1391 1392menu "Lock Debugging (spinlocks, mutexes, etc...)" 1393 1394config LOCK_DEBUGGING_SUPPORT 1395 bool 1396 depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1397 default y 1398 1399config PROVE_LOCKING 1400 bool "Lock debugging: prove locking correctness" 1401 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1402 select LOCKDEP 1403 select DEBUG_SPINLOCK 1404 select DEBUG_MUTEXES if !PREEMPT_RT 1405 select DEBUG_RT_MUTEXES if RT_MUTEXES 1406 select DEBUG_RWSEMS if !PREEMPT_RT 1407 select DEBUG_WW_MUTEX_SLOWPATH 1408 select DEBUG_LOCK_ALLOC 1409 select PREEMPT_COUNT if !ARCH_NO_PREEMPT 1410 select TRACE_IRQFLAGS 1411 default n 1412 help 1413 This feature enables the kernel to prove that all locking 1414 that occurs in the kernel runtime is mathematically 1415 correct: that under no circumstance could an arbitrary (and 1416 not yet triggered) combination of observed locking 1417 sequences (on an arbitrary number of CPUs, running an 1418 arbitrary number of tasks and interrupt contexts) cause a 1419 deadlock. 1420 1421 In short, this feature enables the kernel to report locking 1422 related deadlocks before they actually occur. 1423 1424 The proof does not depend on how hard and complex a 1425 deadlock scenario would be to trigger: how many 1426 participant CPUs, tasks and irq-contexts would be needed 1427 for it to trigger. The proof also does not depend on 1428 timing: if a race and a resulting deadlock is possible 1429 theoretically (no matter how unlikely the race scenario 1430 is), it will be proven so and will immediately be 1431 reported by the kernel (once the event is observed that 1432 makes the deadlock theoretically possible). 1433 1434 If a deadlock is impossible (i.e. the locking rules, as 1435 observed by the kernel, are mathematically correct), the 1436 kernel reports nothing. 1437 1438 NOTE: this feature can also be enabled for rwlocks, mutexes 1439 and rwsems - in which case all dependencies between these 1440 different locking variants are observed and mapped too, and 1441 the proof of observed correctness is also maintained for an 1442 arbitrary combination of these separate locking variants. 1443 1444 For more details, see Documentation/locking/lockdep-design.rst. 1445 1446config PROVE_RAW_LOCK_NESTING 1447 bool "Enable raw_spinlock - spinlock nesting checks" if !ARCH_SUPPORTS_RT 1448 depends on PROVE_LOCKING 1449 default y if ARCH_SUPPORTS_RT 1450 help 1451 Enable the raw_spinlock vs. spinlock nesting checks which ensure 1452 that the lock nesting rules for PREEMPT_RT enabled kernels are 1453 not violated. 1454 1455config LOCK_STAT 1456 bool "Lock usage statistics" 1457 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1458 select LOCKDEP 1459 select DEBUG_SPINLOCK 1460 select DEBUG_MUTEXES if !PREEMPT_RT 1461 select DEBUG_RT_MUTEXES if RT_MUTEXES 1462 select DEBUG_LOCK_ALLOC 1463 default n 1464 help 1465 This feature enables tracking lock contention points 1466 1467 For more details, see Documentation/locking/lockstat.rst 1468 1469 This also enables lock events required by "perf lock", 1470 subcommand of perf. 1471 If you want to use "perf lock", you also need to turn on 1472 CONFIG_EVENT_TRACING. 1473 1474 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events. 1475 (CONFIG_LOCKDEP defines "acquire" and "release" events.) 1476 1477config DEBUG_RT_MUTEXES 1478 bool "RT Mutex debugging, deadlock detection" 1479 depends on DEBUG_KERNEL && RT_MUTEXES 1480 help 1481 This allows rt mutex semantics violations and rt mutex related 1482 deadlocks (lockups) to be detected and reported automatically. 1483 1484config DEBUG_SPINLOCK 1485 bool "Spinlock and rw-lock debugging: basic checks" 1486 depends on DEBUG_KERNEL 1487 select UNINLINE_SPIN_UNLOCK 1488 help 1489 Say Y here and build SMP to catch missing spinlock initialization 1490 and certain other kinds of spinlock errors commonly made. This is 1491 best used in conjunction with the NMI watchdog so that spinlock 1492 deadlocks are also debuggable. 1493 1494config DEBUG_MUTEXES 1495 bool "Mutex debugging: basic checks" 1496 depends on DEBUG_KERNEL && !PREEMPT_RT 1497 help 1498 This feature allows mutex semantics violations to be detected and 1499 reported. 1500 1501config DEBUG_WW_MUTEX_SLOWPATH 1502 bool "Wait/wound mutex debugging: Slowpath testing" 1503 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1504 select DEBUG_LOCK_ALLOC 1505 select DEBUG_SPINLOCK 1506 select DEBUG_MUTEXES if !PREEMPT_RT 1507 select DEBUG_RT_MUTEXES if PREEMPT_RT 1508 help 1509 This feature enables slowpath testing for w/w mutex users by 1510 injecting additional -EDEADLK wound/backoff cases. Together with 1511 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this 1512 will test all possible w/w mutex interface abuse with the 1513 exception of simply not acquiring all the required locks. 1514 Note that this feature can introduce significant overhead, so 1515 it really should not be enabled in a production or distro kernel, 1516 even a debug kernel. If you are a driver writer, enable it. If 1517 you are a distro, do not. 1518 1519config DEBUG_RWSEMS 1520 bool "RW Semaphore debugging: basic checks" 1521 depends on DEBUG_KERNEL && !PREEMPT_RT 1522 help 1523 This debugging feature allows mismatched rw semaphore locks 1524 and unlocks to be detected and reported. 1525 1526config DEBUG_LOCK_ALLOC 1527 bool "Lock debugging: detect incorrect freeing of live locks" 1528 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1529 select DEBUG_SPINLOCK 1530 select DEBUG_MUTEXES if !PREEMPT_RT 1531 select DEBUG_RT_MUTEXES if RT_MUTEXES 1532 select LOCKDEP 1533 help 1534 This feature will check whether any held lock (spinlock, rwlock, 1535 mutex or rwsem) is incorrectly freed by the kernel, via any of the 1536 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(), 1537 vfree(), etc.), whether a live lock is incorrectly reinitialized via 1538 spin_lock_init()/mutex_init()/etc., or whether there is any lock 1539 held during task exit. 1540 1541config LOCKDEP 1542 bool 1543 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1544 select STACKTRACE 1545 select KALLSYMS 1546 select KALLSYMS_ALL 1547 1548config LOCKDEP_SMALL 1549 bool 1550 1551config LOCKDEP_BITS 1552 int "Size for MAX_LOCKDEP_ENTRIES (as Nth power of 2)" 1553 depends on LOCKDEP && !LOCKDEP_SMALL 1554 range 10 24 1555 default 15 1556 help 1557 Try increasing this value if you hit "BUG: MAX_LOCKDEP_ENTRIES too low!" message. 1558 1559config LOCKDEP_CHAINS_BITS 1560 int "Size for MAX_LOCKDEP_CHAINS (as Nth power of 2)" 1561 depends on LOCKDEP && !LOCKDEP_SMALL 1562 range 10 21 1563 default 16 1564 help 1565 Try increasing this value if you hit "BUG: MAX_LOCKDEP_CHAINS too low!" message. 1566 1567config LOCKDEP_STACK_TRACE_BITS 1568 int "Size for MAX_STACK_TRACE_ENTRIES (as Nth power of 2)" 1569 depends on LOCKDEP && !LOCKDEP_SMALL 1570 range 10 26 1571 default 19 1572 help 1573 Try increasing this value if you hit "BUG: MAX_STACK_TRACE_ENTRIES too low!" message. 1574 1575config LOCKDEP_STACK_TRACE_HASH_BITS 1576 int "Size for STACK_TRACE_HASH_SIZE (as Nth power of 2)" 1577 depends on LOCKDEP && !LOCKDEP_SMALL 1578 range 10 26 1579 default 14 1580 help 1581 Try increasing this value if you need large STACK_TRACE_HASH_SIZE. 1582 1583config LOCKDEP_CIRCULAR_QUEUE_BITS 1584 int "Size for elements in circular_queue struct (as Nth power of 2)" 1585 depends on LOCKDEP 1586 range 10 26 1587 default 12 1588 help 1589 Try increasing this value if you hit "lockdep bfs error:-1" warning due to __cq_enqueue() failure. 1590 1591config DEBUG_LOCKDEP 1592 bool "Lock dependency engine debugging" 1593 depends on DEBUG_KERNEL && LOCKDEP 1594 select DEBUG_IRQFLAGS 1595 help 1596 If you say Y here, the lock dependency engine will do 1597 additional runtime checks to debug itself, at the price 1598 of more runtime overhead. 1599 1600config DEBUG_ATOMIC_SLEEP 1601 bool "Sleep inside atomic section checking" 1602 select PREEMPT_COUNT 1603 depends on DEBUG_KERNEL 1604 depends on !ARCH_NO_PREEMPT 1605 help 1606 If you say Y here, various routines which may sleep will become very 1607 noisy if they are called inside atomic sections: when a spinlock is 1608 held, inside an rcu read side critical section, inside preempt disabled 1609 sections, inside an interrupt, etc... 1610 1611config DEBUG_LOCKING_API_SELFTESTS 1612 bool "Locking API boot-time self-tests" 1613 depends on DEBUG_KERNEL 1614 help 1615 Say Y here if you want the kernel to run a short self-test during 1616 bootup. The self-test checks whether common types of locking bugs 1617 are detected by debugging mechanisms or not. (if you disable 1618 lock debugging then those bugs won't be detected of course.) 1619 The following locking APIs are covered: spinlocks, rwlocks, 1620 mutexes and rwsems. 1621 1622config LOCK_TORTURE_TEST 1623 tristate "torture tests for locking" 1624 depends on DEBUG_KERNEL 1625 select TORTURE_TEST 1626 help 1627 This option provides a kernel module that runs torture tests 1628 on kernel locking primitives. The kernel module may be built 1629 after the fact on the running kernel to be tested, if desired. 1630 1631 Say Y here if you want kernel locking-primitive torture tests 1632 to be built into the kernel. 1633 Say M if you want these torture tests to build as a module. 1634 Say N if you are unsure. 1635 1636config WW_MUTEX_SELFTEST 1637 tristate "Wait/wound mutex selftests" 1638 help 1639 This option provides a kernel module that runs tests on the 1640 on the struct ww_mutex locking API. 1641 1642 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction 1643 with this test harness. 1644 1645 Say M if you want these self tests to build as a module. 1646 Say N if you are unsure. 1647 1648config SCF_TORTURE_TEST 1649 tristate "torture tests for smp_call_function*()" 1650 depends on DEBUG_KERNEL 1651 select TORTURE_TEST 1652 help 1653 This option provides a kernel module that runs torture tests 1654 on the smp_call_function() family of primitives. The kernel 1655 module may be built after the fact on the running kernel to 1656 be tested, if desired. 1657 1658config CSD_LOCK_WAIT_DEBUG 1659 bool "Debugging for csd_lock_wait(), called from smp_call_function*()" 1660 depends on DEBUG_KERNEL 1661 depends on SMP 1662 depends on 64BIT 1663 default n 1664 help 1665 This option enables debug prints when CPUs are slow to respond 1666 to the smp_call_function*() IPI wrappers. These debug prints 1667 include the IPI handler function currently executing (if any) 1668 and relevant stack traces. 1669 1670config CSD_LOCK_WAIT_DEBUG_DEFAULT 1671 bool "Default csd_lock_wait() debugging on at boot time" 1672 depends on CSD_LOCK_WAIT_DEBUG 1673 depends on 64BIT 1674 default n 1675 help 1676 This option causes the csdlock_debug= kernel boot parameter to 1677 default to 1 (basic debugging) instead of 0 (no debugging). 1678 1679endmenu # lock debugging 1680 1681config TRACE_IRQFLAGS 1682 depends on TRACE_IRQFLAGS_SUPPORT 1683 bool 1684 help 1685 Enables hooks to interrupt enabling and disabling for 1686 either tracing or lock debugging. 1687 1688config TRACE_IRQFLAGS_NMI 1689 def_bool y 1690 depends on TRACE_IRQFLAGS 1691 depends on TRACE_IRQFLAGS_NMI_SUPPORT 1692 1693config NMI_CHECK_CPU 1694 bool "Debugging for CPUs failing to respond to backtrace requests" 1695 depends on DEBUG_KERNEL 1696 depends on X86 1697 default n 1698 help 1699 Enables debug prints when a CPU fails to respond to a given 1700 backtrace NMI. These prints provide some reasons why a CPU 1701 might legitimately be failing to respond, for example, if it 1702 is offline of if ignore_nmis is set. 1703 1704config DEBUG_IRQFLAGS 1705 bool "Debug IRQ flag manipulation" 1706 help 1707 Enables checks for potentially unsafe enabling or disabling of 1708 interrupts, such as calling raw_local_irq_restore() when interrupts 1709 are enabled. 1710 1711config STACKTRACE 1712 bool "Stack backtrace support" 1713 depends on STACKTRACE_SUPPORT 1714 help 1715 This option causes the kernel to create a /proc/pid/stack for 1716 every process, showing its current stack trace. 1717 It is also used by various kernel debugging features that require 1718 stack trace generation. 1719 1720config WARN_ALL_UNSEEDED_RANDOM 1721 bool "Warn for all uses of unseeded randomness" 1722 default n 1723 help 1724 Some parts of the kernel contain bugs relating to their use of 1725 cryptographically secure random numbers before it's actually possible 1726 to generate those numbers securely. This setting ensures that these 1727 flaws don't go unnoticed, by enabling a message, should this ever 1728 occur. This will allow people with obscure setups to know when things 1729 are going wrong, so that they might contact developers about fixing 1730 it. 1731 1732 Unfortunately, on some models of some architectures getting 1733 a fully seeded CRNG is extremely difficult, and so this can 1734 result in dmesg getting spammed for a surprisingly long 1735 time. This is really bad from a security perspective, and 1736 so architecture maintainers really need to do what they can 1737 to get the CRNG seeded sooner after the system is booted. 1738 However, since users cannot do anything actionable to 1739 address this, by default this option is disabled. 1740 1741 Say Y here if you want to receive warnings for all uses of 1742 unseeded randomness. This will be of use primarily for 1743 those developers interested in improving the security of 1744 Linux kernels running on their architecture (or 1745 subarchitecture). 1746 1747config DEBUG_KOBJECT 1748 bool "kobject debugging" 1749 depends on DEBUG_KERNEL 1750 help 1751 If you say Y here, some extra kobject debugging messages will be sent 1752 to the syslog. 1753 1754config DEBUG_KOBJECT_RELEASE 1755 bool "kobject release debugging" 1756 depends on DEBUG_OBJECTS_TIMERS 1757 help 1758 kobjects are reference counted objects. This means that their 1759 last reference count put is not predictable, and the kobject can 1760 live on past the point at which a driver decides to drop its 1761 initial reference to the kobject gained on allocation. An 1762 example of this would be a struct device which has just been 1763 unregistered. 1764 1765 However, some buggy drivers assume that after such an operation, 1766 the memory backing the kobject can be immediately freed. This 1767 goes completely against the principles of a refcounted object. 1768 1769 If you say Y here, the kernel will delay the release of kobjects 1770 on the last reference count to improve the visibility of this 1771 kind of kobject release bug. 1772 1773config HAVE_DEBUG_BUGVERBOSE 1774 bool 1775 1776menu "Debug kernel data structures" 1777 1778config DEBUG_LIST 1779 bool "Debug linked list manipulation" 1780 depends on DEBUG_KERNEL 1781 select LIST_HARDENED 1782 help 1783 Enable this to turn on extended checks in the linked-list walking 1784 routines. 1785 1786 This option trades better quality error reports for performance, and 1787 is more suitable for kernel debugging. If you care about performance, 1788 you should only enable CONFIG_LIST_HARDENED instead. 1789 1790 If unsure, say N. 1791 1792config DEBUG_PLIST 1793 bool "Debug priority linked list manipulation" 1794 depends on DEBUG_KERNEL 1795 help 1796 Enable this to turn on extended checks in the priority-ordered 1797 linked-list (plist) walking routines. This checks the entire 1798 list multiple times during each manipulation. 1799 1800 If unsure, say N. 1801 1802config DEBUG_SG 1803 bool "Debug SG table operations" 1804 depends on DEBUG_KERNEL 1805 help 1806 Enable this to turn on checks on scatter-gather tables. This can 1807 help find problems with drivers that do not properly initialize 1808 their sg tables. 1809 1810 If unsure, say N. 1811 1812config DEBUG_NOTIFIERS 1813 bool "Debug notifier call chains" 1814 depends on DEBUG_KERNEL 1815 help 1816 Enable this to turn on sanity checking for notifier call chains. 1817 This is most useful for kernel developers to make sure that 1818 modules properly unregister themselves from notifier chains. 1819 This is a relatively cheap check but if you care about maximum 1820 performance, say N. 1821 1822config DEBUG_CLOSURES 1823 bool "Debug closures (bcache async widgits)" 1824 depends on CLOSURES 1825 select DEBUG_FS 1826 help 1827 Keeps all active closures in a linked list and provides a debugfs 1828 interface to list them, which makes it possible to see asynchronous 1829 operations that get stuck. 1830 1831config DEBUG_MAPLE_TREE 1832 bool "Debug maple trees" 1833 depends on DEBUG_KERNEL 1834 help 1835 Enable maple tree debugging information and extra validations. 1836 1837 If unsure, say N. 1838 1839endmenu 1840 1841source "kernel/rcu/Kconfig.debug" 1842 1843config DEBUG_WQ_FORCE_RR_CPU 1844 bool "Force round-robin CPU selection for unbound work items" 1845 depends on DEBUG_KERNEL 1846 default n 1847 help 1848 Workqueue used to implicitly guarantee that work items queued 1849 without explicit CPU specified are put on the local CPU. This 1850 guarantee is no longer true and while local CPU is still 1851 preferred work items may be put on foreign CPUs. Kernel 1852 parameter "workqueue.debug_force_rr_cpu" is added to force 1853 round-robin CPU selection to flush out usages which depend on the 1854 now broken guarantee. This config option enables the debug 1855 feature by default. When enabled, memory and cache locality will 1856 be impacted. 1857 1858config CPU_HOTPLUG_STATE_CONTROL 1859 bool "Enable CPU hotplug state control" 1860 depends on DEBUG_KERNEL 1861 depends on HOTPLUG_CPU 1862 default n 1863 help 1864 Allows to write steps between "offline" and "online" to the CPUs 1865 sysfs target file so states can be stepped granular. This is a debug 1866 option for now as the hotplug machinery cannot be stopped and 1867 restarted at arbitrary points yet. 1868 1869 Say N if your are unsure. 1870 1871config LATENCYTOP 1872 bool "Latency measuring infrastructure" 1873 depends on DEBUG_KERNEL 1874 depends on STACKTRACE_SUPPORT 1875 depends on PROC_FS 1876 depends on FRAME_POINTER || MIPS || PPC || S390 || MICROBLAZE || ARM || ARC || X86 1877 select KALLSYMS 1878 select KALLSYMS_ALL 1879 select STACKTRACE 1880 select SCHEDSTATS 1881 help 1882 Enable this option if you want to use the LatencyTOP tool 1883 to find out which userspace is blocking on what kernel operations. 1884 1885config DEBUG_CGROUP_REF 1886 bool "Disable inlining of cgroup css reference count functions" 1887 depends on DEBUG_KERNEL 1888 depends on CGROUPS 1889 depends on KPROBES 1890 default n 1891 help 1892 Force cgroup css reference count functions to not be inlined so 1893 that they can be kprobed for debugging. 1894 1895source "kernel/trace/Kconfig" 1896 1897config PROVIDE_OHCI1394_DMA_INIT 1898 bool "Remote debugging over FireWire early on boot" 1899 depends on PCI && X86 1900 help 1901 If you want to debug problems which hang or crash the kernel early 1902 on boot and the crashing machine has a FireWire port, you can use 1903 this feature to remotely access the memory of the crashed machine 1904 over FireWire. This employs remote DMA as part of the OHCI1394 1905 specification which is now the standard for FireWire controllers. 1906 1907 With remote DMA, you can monitor the printk buffer remotely using 1908 firescope and access all memory below 4GB using fireproxy from gdb. 1909 Even controlling a kernel debugger is possible using remote DMA. 1910 1911 Usage: 1912 1913 If ohci1394_dma=early is used as boot parameter, it will initialize 1914 all OHCI1394 controllers which are found in the PCI config space. 1915 1916 As all changes to the FireWire bus such as enabling and disabling 1917 devices cause a bus reset and thereby disable remote DMA for all 1918 devices, be sure to have the cable plugged and FireWire enabled on 1919 the debugging host before booting the debug target for debugging. 1920 1921 This code (~1k) is freed after boot. By then, the firewire stack 1922 in charge of the OHCI-1394 controllers should be used instead. 1923 1924 See Documentation/core-api/debugging-via-ohci1394.rst for more information. 1925 1926source "samples/Kconfig" 1927 1928config ARCH_HAS_DEVMEM_IS_ALLOWED 1929 bool 1930 1931config STRICT_DEVMEM 1932 bool "Filter access to /dev/mem" 1933 depends on MMU && DEVMEM 1934 depends on ARCH_HAS_DEVMEM_IS_ALLOWED || GENERIC_LIB_DEVMEM_IS_ALLOWED 1935 default y if PPC || X86 || ARM64 || S390 1936 help 1937 If this option is disabled, you allow userspace (root) access to all 1938 of memory, including kernel and userspace memory. Accidental 1939 access to this is obviously disastrous, but specific access can 1940 be used by people debugging the kernel. Note that with PAT support 1941 enabled, even in this case there are restrictions on /dev/mem 1942 use due to the cache aliasing requirements. 1943 1944 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem 1945 file only allows userspace access to PCI space and the BIOS code and 1946 data regions. This is sufficient for dosemu and X and all common 1947 users of /dev/mem. 1948 1949 If in doubt, say Y. 1950 1951config IO_STRICT_DEVMEM 1952 bool "Filter I/O access to /dev/mem" 1953 depends on STRICT_DEVMEM 1954 help 1955 If this option is disabled, you allow userspace (root) access to all 1956 io-memory regardless of whether a driver is actively using that 1957 range. Accidental access to this is obviously disastrous, but 1958 specific access can be used by people debugging kernel drivers. 1959 1960 If this option is switched on, the /dev/mem file only allows 1961 userspace access to *idle* io-memory ranges (see /proc/iomem) This 1962 may break traditional users of /dev/mem (dosemu, legacy X, etc...) 1963 if the driver using a given range cannot be disabled. 1964 1965 If in doubt, say Y. 1966 1967menu "$(SRCARCH) Debugging" 1968 1969source "arch/$(SRCARCH)/Kconfig.debug" 1970 1971endmenu 1972 1973menu "Kernel Testing and Coverage" 1974 1975source "lib/kunit/Kconfig" 1976 1977config NOTIFIER_ERROR_INJECTION 1978 tristate "Notifier error injection" 1979 depends on DEBUG_KERNEL 1980 select DEBUG_FS 1981 help 1982 This option provides the ability to inject artificial errors to 1983 specified notifier chain callbacks. It is useful to test the error 1984 handling of notifier call chain failures. 1985 1986 Say N if unsure. 1987 1988config PM_NOTIFIER_ERROR_INJECT 1989 tristate "PM notifier error injection module" 1990 depends on PM && NOTIFIER_ERROR_INJECTION 1991 default m if PM_DEBUG 1992 help 1993 This option provides the ability to inject artificial errors to 1994 PM notifier chain callbacks. It is controlled through debugfs 1995 interface /sys/kernel/debug/notifier-error-inject/pm 1996 1997 If the notifier call chain should be failed with some events 1998 notified, write the error code to "actions/<notifier event>/error". 1999 2000 Example: Inject PM suspend error (-12 = -ENOMEM) 2001 2002 # cd /sys/kernel/debug/notifier-error-inject/pm/ 2003 # echo -12 > actions/PM_SUSPEND_PREPARE/error 2004 # echo mem > /sys/power/state 2005 bash: echo: write error: Cannot allocate memory 2006 2007 To compile this code as a module, choose M here: the module will 2008 be called pm-notifier-error-inject. 2009 2010 If unsure, say N. 2011 2012config OF_RECONFIG_NOTIFIER_ERROR_INJECT 2013 tristate "OF reconfig notifier error injection module" 2014 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION 2015 help 2016 This option provides the ability to inject artificial errors to 2017 OF reconfig notifier chain callbacks. It is controlled 2018 through debugfs interface under 2019 /sys/kernel/debug/notifier-error-inject/OF-reconfig/ 2020 2021 If the notifier call chain should be failed with some events 2022 notified, write the error code to "actions/<notifier event>/error". 2023 2024 To compile this code as a module, choose M here: the module will 2025 be called of-reconfig-notifier-error-inject. 2026 2027 If unsure, say N. 2028 2029config NETDEV_NOTIFIER_ERROR_INJECT 2030 tristate "Netdev notifier error injection module" 2031 depends on NET && NOTIFIER_ERROR_INJECTION 2032 help 2033 This option provides the ability to inject artificial errors to 2034 netdevice notifier chain callbacks. It is controlled through debugfs 2035 interface /sys/kernel/debug/notifier-error-inject/netdev 2036 2037 If the notifier call chain should be failed with some events 2038 notified, write the error code to "actions/<notifier event>/error". 2039 2040 Example: Inject netdevice mtu change error (-22 = -EINVAL) 2041 2042 # cd /sys/kernel/debug/notifier-error-inject/netdev 2043 # echo -22 > actions/NETDEV_CHANGEMTU/error 2044 # ip link set eth0 mtu 1024 2045 RTNETLINK answers: Invalid argument 2046 2047 To compile this code as a module, choose M here: the module will 2048 be called netdev-notifier-error-inject. 2049 2050 If unsure, say N. 2051 2052config FUNCTION_ERROR_INJECTION 2053 bool "Fault-injections of functions" 2054 depends on HAVE_FUNCTION_ERROR_INJECTION && KPROBES 2055 help 2056 Add fault injections into various functions that are annotated with 2057 ALLOW_ERROR_INJECTION() in the kernel. BPF may also modify the return 2058 value of these functions. This is useful to test error paths of code. 2059 2060 If unsure, say N 2061 2062config FAULT_INJECTION 2063 bool "Fault-injection framework" 2064 depends on DEBUG_KERNEL 2065 help 2066 Provide fault-injection framework. 2067 For more details, see Documentation/fault-injection/. 2068 2069config FAILSLAB 2070 bool "Fault-injection capability for kmalloc" 2071 depends on FAULT_INJECTION 2072 help 2073 Provide fault-injection capability for kmalloc. 2074 2075config FAIL_PAGE_ALLOC 2076 bool "Fault-injection capability for alloc_pages()" 2077 depends on FAULT_INJECTION 2078 help 2079 Provide fault-injection capability for alloc_pages(). 2080 2081config FAULT_INJECTION_USERCOPY 2082 bool "Fault injection capability for usercopy functions" 2083 depends on FAULT_INJECTION 2084 help 2085 Provides fault-injection capability to inject failures 2086 in usercopy functions (copy_from_user(), get_user(), ...). 2087 2088config FAIL_MAKE_REQUEST 2089 bool "Fault-injection capability for disk IO" 2090 depends on FAULT_INJECTION && BLOCK 2091 help 2092 Provide fault-injection capability for disk IO. 2093 2094config FAIL_IO_TIMEOUT 2095 bool "Fault-injection capability for faking disk interrupts" 2096 depends on FAULT_INJECTION && BLOCK 2097 help 2098 Provide fault-injection capability on end IO handling. This 2099 will make the block layer "forget" an interrupt as configured, 2100 thus exercising the error handling. 2101 2102 Only works with drivers that use the generic timeout handling, 2103 for others it won't do anything. 2104 2105config FAIL_FUTEX 2106 bool "Fault-injection capability for futexes" 2107 select DEBUG_FS 2108 depends on FAULT_INJECTION && FUTEX 2109 help 2110 Provide fault-injection capability for futexes. 2111 2112config FAULT_INJECTION_DEBUG_FS 2113 bool "Debugfs entries for fault-injection capabilities" 2114 depends on FAULT_INJECTION && SYSFS && DEBUG_FS 2115 help 2116 Enable configuration of fault-injection capabilities via debugfs. 2117 2118config FAIL_FUNCTION 2119 bool "Fault-injection capability for functions" 2120 depends on FAULT_INJECTION_DEBUG_FS && FUNCTION_ERROR_INJECTION 2121 help 2122 Provide function-based fault-injection capability. 2123 This will allow you to override a specific function with a return 2124 with given return value. As a result, function caller will see 2125 an error value and have to handle it. This is useful to test the 2126 error handling in various subsystems. 2127 2128config FAIL_MMC_REQUEST 2129 bool "Fault-injection capability for MMC IO" 2130 depends on FAULT_INJECTION_DEBUG_FS && MMC 2131 help 2132 Provide fault-injection capability for MMC IO. 2133 This will make the mmc core return data errors. This is 2134 useful to test the error handling in the mmc block device 2135 and to test how the mmc host driver handles retries from 2136 the block device. 2137 2138config FAIL_SUNRPC 2139 bool "Fault-injection capability for SunRPC" 2140 depends on FAULT_INJECTION_DEBUG_FS && SUNRPC_DEBUG 2141 help 2142 Provide fault-injection capability for SunRPC and 2143 its consumers. 2144 2145config FAIL_SKB_REALLOC 2146 bool "Fault-injection capability forcing skb to reallocate" 2147 depends on FAULT_INJECTION_DEBUG_FS 2148 help 2149 Provide fault-injection capability that forces the skb to be 2150 reallocated, catching possible invalid pointers to the skb. 2151 2152 For more information, check 2153 Documentation/fault-injection/fault-injection.rst 2154 2155config FAULT_INJECTION_CONFIGFS 2156 bool "Configfs interface for fault-injection capabilities" 2157 depends on FAULT_INJECTION 2158 select CONFIGFS_FS 2159 help 2160 This option allows configfs-based drivers to dynamically configure 2161 fault-injection via configfs. Each parameter for driver-specific 2162 fault-injection can be made visible as a configfs attribute in a 2163 configfs group. 2164 2165 2166config FAULT_INJECTION_STACKTRACE_FILTER 2167 bool "stacktrace filter for fault-injection capabilities" 2168 depends on FAULT_INJECTION 2169 depends on (FAULT_INJECTION_DEBUG_FS || FAULT_INJECTION_CONFIGFS) && STACKTRACE_SUPPORT 2170 select STACKTRACE 2171 depends on FRAME_POINTER || MIPS || PPC || S390 || MICROBLAZE || ARM || ARC || X86 2172 help 2173 Provide stacktrace filter for fault-injection capabilities 2174 2175config ARCH_HAS_KCOV 2176 bool 2177 help 2178 An architecture should select this when it can successfully 2179 build and run with CONFIG_KCOV. This typically requires 2180 disabling instrumentation for some early boot code. 2181 2182config KCOV 2183 bool "Code coverage for fuzzing" 2184 depends on ARCH_HAS_KCOV 2185 depends on !ARCH_WANTS_NO_INSTR || HAVE_NOINSTR_HACK || \ 2186 GCC_VERSION >= 120000 || CC_IS_CLANG 2187 select DEBUG_FS 2188 select OBJTOOL if HAVE_NOINSTR_HACK 2189 help 2190 KCOV exposes kernel code coverage information in a form suitable 2191 for coverage-guided fuzzing (randomized testing). 2192 2193 For more details, see Documentation/dev-tools/kcov.rst. 2194 2195config KCOV_ENABLE_COMPARISONS 2196 bool "Enable comparison operands collection by KCOV" 2197 depends on KCOV 2198 depends on $(cc-option,-fsanitize-coverage=trace-cmp) 2199 help 2200 KCOV also exposes operands of every comparison in the instrumented 2201 code along with operand sizes and PCs of the comparison instructions. 2202 These operands can be used by fuzzing engines to improve the quality 2203 of fuzzing coverage. 2204 2205config KCOV_INSTRUMENT_ALL 2206 bool "Instrument all code by default" 2207 depends on KCOV 2208 default y 2209 help 2210 If you are doing generic system call fuzzing (like e.g. syzkaller), 2211 then you will want to instrument the whole kernel and you should 2212 say y here. If you are doing more targeted fuzzing (like e.g. 2213 filesystem fuzzing with AFL) then you will want to enable coverage 2214 for more specific subsets of files, and should say n here. 2215 2216config KCOV_IRQ_AREA_SIZE 2217 hex "Size of interrupt coverage collection area in words" 2218 depends on KCOV 2219 default 0x40000 2220 help 2221 KCOV uses preallocated per-cpu areas to collect coverage from 2222 soft interrupts. This specifies the size of those areas in the 2223 number of unsigned long words. 2224 2225config KCOV_SELFTEST 2226 bool "Perform short selftests on boot" 2227 depends on KCOV 2228 help 2229 Run short KCOV coverage collection selftests on boot. 2230 On test failure, causes the kernel to panic. Recommended to be 2231 enabled, ensuring critical functionality works as intended. 2232 2233menuconfig RUNTIME_TESTING_MENU 2234 bool "Runtime Testing" 2235 default y 2236 2237if RUNTIME_TESTING_MENU 2238 2239config TEST_DHRY 2240 tristate "Dhrystone benchmark test" 2241 help 2242 Enable this to include the Dhrystone 2.1 benchmark. This test 2243 calculates the number of Dhrystones per second, and the number of 2244 DMIPS (Dhrystone MIPS) obtained when the Dhrystone score is divided 2245 by 1757 (the number of Dhrystones per second obtained on the VAX 2246 11/780, nominally a 1 MIPS machine). 2247 2248 To run the benchmark, it needs to be enabled explicitly, either from 2249 the kernel command line (when built-in), or from userspace (when 2250 built-in or modular). 2251 2252 Run once during kernel boot: 2253 2254 test_dhry.run 2255 2256 Set number of iterations from kernel command line: 2257 2258 test_dhry.iterations=<n> 2259 2260 Set number of iterations from userspace: 2261 2262 echo <n> > /sys/module/test_dhry/parameters/iterations 2263 2264 Trigger manual run from userspace: 2265 2266 echo y > /sys/module/test_dhry/parameters/run 2267 2268 If the number of iterations is <= 0, the test will devise a suitable 2269 number of iterations (test runs for at least 2s) automatically. 2270 This process takes ca. 4s. 2271 2272 If unsure, say N. 2273 2274config LKDTM 2275 tristate "Linux Kernel Dump Test Tool Module" 2276 depends on DEBUG_FS 2277 help 2278 This module enables testing of the different dumping mechanisms by 2279 inducing system failures at predefined crash points. 2280 If you don't need it: say N 2281 Choose M here to compile this code as a module. The module will be 2282 called lkdtm. 2283 2284 Documentation on how to use the module can be found in 2285 Documentation/fault-injection/provoke-crashes.rst 2286 2287config CPUMASK_KUNIT_TEST 2288 tristate "KUnit test for cpumask" if !KUNIT_ALL_TESTS 2289 depends on KUNIT 2290 default KUNIT_ALL_TESTS 2291 help 2292 Enable to turn on cpumask tests, running at boot or module load time. 2293 2294 For more information on KUnit and unit tests in general, please refer 2295 to the KUnit documentation in Documentation/dev-tools/kunit/. 2296 2297 If unsure, say N. 2298 2299config TEST_LIST_SORT 2300 tristate "Linked list sorting test" if !KUNIT_ALL_TESTS 2301 depends on KUNIT 2302 default KUNIT_ALL_TESTS 2303 help 2304 Enable this to turn on 'list_sort()' function test. This test is 2305 executed only once during system boot (so affects only boot time), 2306 or at module load time. 2307 2308 If unsure, say N. 2309 2310config TEST_MIN_HEAP 2311 tristate "Min heap test" 2312 depends on DEBUG_KERNEL || m 2313 help 2314 Enable this to turn on min heap function tests. This test is 2315 executed only once during system boot (so affects only boot time), 2316 or at module load time. 2317 2318 If unsure, say N. 2319 2320config TEST_SORT 2321 tristate "Array-based sort test" if !KUNIT_ALL_TESTS 2322 depends on KUNIT 2323 default KUNIT_ALL_TESTS 2324 help 2325 This option enables the self-test function of 'sort()' at boot, 2326 or at module load time. 2327 2328 If unsure, say N. 2329 2330config TEST_DIV64 2331 tristate "64bit/32bit division and modulo test" 2332 depends on DEBUG_KERNEL || m 2333 help 2334 Enable this to turn on 'do_div()' function test. This test is 2335 executed only once during system boot (so affects only boot time), 2336 or at module load time. 2337 2338 If unsure, say N. 2339 2340config TEST_MULDIV64 2341 tristate "mul_u64_u64_div_u64() test" 2342 depends on DEBUG_KERNEL || m 2343 help 2344 Enable this to turn on 'mul_u64_u64_div_u64()' function test. 2345 This test is executed only once during system boot (so affects 2346 only boot time), or at module load time. 2347 2348 If unsure, say N. 2349 2350config TEST_IOV_ITER 2351 tristate "Test iov_iter operation" if !KUNIT_ALL_TESTS 2352 depends on KUNIT 2353 depends on MMU 2354 default KUNIT_ALL_TESTS 2355 help 2356 Enable this to turn on testing of the operation of the I/O iterator 2357 (iov_iter). This test is executed only once during system boot (so 2358 affects only boot time), or at module load time. 2359 2360 If unsure, say N. 2361 2362config KPROBES_SANITY_TEST 2363 tristate "Kprobes sanity tests" if !KUNIT_ALL_TESTS 2364 depends on DEBUG_KERNEL 2365 depends on KPROBES 2366 depends on KUNIT 2367 select STACKTRACE if ARCH_CORRECT_STACKTRACE_ON_KRETPROBE 2368 default KUNIT_ALL_TESTS 2369 help 2370 This option provides for testing basic kprobes functionality on 2371 boot. Samples of kprobe and kretprobe are inserted and 2372 verified for functionality. 2373 2374 Say N if you are unsure. 2375 2376config FPROBE_SANITY_TEST 2377 bool "Self test for fprobe" 2378 depends on DEBUG_KERNEL 2379 depends on FPROBE 2380 depends on KUNIT=y 2381 help 2382 This option will enable testing the fprobe when the system boot. 2383 A series of tests are made to verify that the fprobe is functioning 2384 properly. 2385 2386 Say N if you are unsure. 2387 2388config BACKTRACE_SELF_TEST 2389 tristate "Self test for the backtrace code" 2390 depends on DEBUG_KERNEL 2391 help 2392 This option provides a kernel module that can be used to test 2393 the kernel stack backtrace code. This option is not useful 2394 for distributions or general kernels, but only for kernel 2395 developers working on architecture code. 2396 2397 Note that if you want to also test saved backtraces, you will 2398 have to enable STACKTRACE as well. 2399 2400 Say N if you are unsure. 2401 2402config TEST_REF_TRACKER 2403 tristate "Self test for reference tracker" 2404 depends on DEBUG_KERNEL && STACKTRACE_SUPPORT 2405 select REF_TRACKER 2406 help 2407 This option provides a kernel module performing tests 2408 using reference tracker infrastructure. 2409 2410 Say N if you are unsure. 2411 2412config RBTREE_TEST 2413 tristate "Red-Black tree test" 2414 depends on DEBUG_KERNEL 2415 help 2416 A benchmark measuring the performance of the rbtree library. 2417 Also includes rbtree invariant checks. 2418 2419config REED_SOLOMON_TEST 2420 tristate "Reed-Solomon library test" 2421 depends on DEBUG_KERNEL || m 2422 select REED_SOLOMON 2423 select REED_SOLOMON_ENC16 2424 select REED_SOLOMON_DEC16 2425 help 2426 This option enables the self-test function of rslib at boot, 2427 or at module load time. 2428 2429 If unsure, say N. 2430 2431config INTERVAL_TREE_TEST 2432 tristate "Interval tree test" 2433 depends on DEBUG_KERNEL 2434 select INTERVAL_TREE 2435 help 2436 A benchmark measuring the performance of the interval tree library 2437 2438config PERCPU_TEST 2439 tristate "Per cpu operations test" 2440 depends on m && DEBUG_KERNEL 2441 help 2442 Enable this option to build test module which validates per-cpu 2443 operations. 2444 2445 If unsure, say N. 2446 2447config ATOMIC64_SELFTEST 2448 tristate "Perform an atomic64_t self-test" 2449 help 2450 Enable this option to test the atomic64_t functions at boot or 2451 at module load time. 2452 2453 If unsure, say N. 2454 2455config ASYNC_RAID6_TEST 2456 tristate "Self test for hardware accelerated raid6 recovery" 2457 depends on ASYNC_RAID6_RECOV 2458 select ASYNC_MEMCPY 2459 help 2460 This is a one-shot self test that permutes through the 2461 recovery of all the possible two disk failure scenarios for a 2462 N-disk array. Recovery is performed with the asynchronous 2463 raid6 recovery routines, and will optionally use an offload 2464 engine if one is available. 2465 2466 If unsure, say N. 2467 2468config TEST_HEXDUMP 2469 tristate "Test functions located in the hexdump module at runtime" 2470 2471config PRINTF_KUNIT_TEST 2472 tristate "KUnit test printf() family of functions at runtime" if !KUNIT_ALL_TESTS 2473 depends on KUNIT 2474 default KUNIT_ALL_TESTS 2475 help 2476 Enable this option to test the printf functions at runtime. 2477 2478 If unsure, say N. 2479 2480config SCANF_KUNIT_TEST 2481 tristate "KUnit test scanf() family of functions at runtime" if !KUNIT_ALL_TESTS 2482 depends on KUNIT 2483 default KUNIT_ALL_TESTS 2484 help 2485 Enable this option to test the scanf functions at runtime. 2486 2487 If unsure, say N. 2488 2489config SEQ_BUF_KUNIT_TEST 2490 tristate "KUnit test for seq_buf" if !KUNIT_ALL_TESTS 2491 depends on KUNIT 2492 default KUNIT_ALL_TESTS 2493 help 2494 This builds unit tests for the seq_buf library. 2495 2496 If unsure, say N. 2497 2498config STRING_KUNIT_TEST 2499 tristate "KUnit test string functions at runtime" if !KUNIT_ALL_TESTS 2500 depends on KUNIT 2501 default KUNIT_ALL_TESTS 2502 2503config STRING_HELPERS_KUNIT_TEST 2504 tristate "KUnit test string helpers at runtime" if !KUNIT_ALL_TESTS 2505 depends on KUNIT 2506 default KUNIT_ALL_TESTS 2507 2508config FFS_KUNIT_TEST 2509 tristate "KUnit test ffs-family functions at runtime" if !KUNIT_ALL_TESTS 2510 depends on KUNIT 2511 default KUNIT_ALL_TESTS 2512 help 2513 This builds KUnit tests for ffs-family bit manipulation functions 2514 including ffs(), __ffs(), fls(), __fls(), fls64(), and __ffs64(). 2515 2516 These tests validate mathematical correctness, edge case handling, 2517 and cross-architecture consistency of bit scanning functions. 2518 2519 For more information on KUnit and unit tests in general, 2520 please refer to Documentation/dev-tools/kunit/. 2521 2522config TEST_KSTRTOX 2523 tristate "Test kstrto*() family of functions at runtime" 2524 2525config TEST_BITMAP 2526 tristate "Test bitmap_*() family of functions at runtime" 2527 help 2528 Enable this option to test the bitmap functions at boot. 2529 2530 If unsure, say N. 2531 2532config TEST_UUID 2533 tristate "Test functions located in the uuid module at runtime" 2534 2535config TEST_XARRAY 2536 tristate "Test the XArray code at runtime" 2537 2538config TEST_MAPLE_TREE 2539 tristate "Test the Maple Tree code at runtime or module load" 2540 help 2541 Enable this option to test the maple tree code functions at boot, or 2542 when the module is loaded. Enable "Debug Maple Trees" will enable 2543 more verbose output on failures. 2544 2545 If unsure, say N. 2546 2547config TEST_RHASHTABLE 2548 tristate "Perform selftest on resizable hash table" 2549 help 2550 Enable this option to test the rhashtable functions at boot. 2551 2552 If unsure, say N. 2553 2554config TEST_IDA 2555 tristate "Perform selftest on IDA functions" 2556 2557config TEST_MISC_MINOR 2558 bool "miscdevice KUnit test" if !KUNIT_ALL_TESTS 2559 depends on KUNIT=y 2560 default KUNIT_ALL_TESTS 2561 help 2562 Kunit test for miscdevice API, specially its behavior in respect to 2563 static and dynamic minor numbers. 2564 2565 KUnit tests run during boot and output the results to the debug log 2566 in TAP format (https://testanything.org/). Only useful for kernel devs 2567 running the KUnit test harness, and not intended for inclusion into a 2568 production build. 2569 2570 For more information on KUnit and unit tests in general please refer 2571 to the KUnit documentation in Documentation/dev-tools/kunit/. 2572 2573 If unsure, say N. 2574 2575config TEST_PARMAN 2576 tristate "Perform selftest on priority array manager" 2577 depends on PARMAN 2578 help 2579 Enable this option to test priority array manager on boot 2580 (or module load). 2581 2582 If unsure, say N. 2583 2584config TEST_IRQ_TIMINGS 2585 bool "IRQ timings selftest" 2586 depends on IRQ_TIMINGS 2587 help 2588 Enable this option to test the irq timings code on boot. 2589 2590 If unsure, say N. 2591 2592config TEST_LKM 2593 tristate "Test module loading with 'hello world' module" 2594 depends on m 2595 help 2596 This builds the "test_module" module that emits "Hello, world" 2597 on printk when loaded. It is designed to be used for basic 2598 evaluation of the module loading subsystem (for example when 2599 validating module verification). It lacks any extra dependencies, 2600 and will not normally be loaded by the system unless explicitly 2601 requested by name. 2602 2603 If unsure, say N. 2604 2605config TEST_BITOPS 2606 tristate "Test module for compilation of bitops operations" 2607 help 2608 This builds the "test_bitops" module that is much like the 2609 TEST_LKM module except that it does a basic exercise of the 2610 set/clear_bit macros and get_count_order/long to make sure there are 2611 no compiler warnings from C=1 sparse checker or -Wextra 2612 compilations. It has no dependencies and doesn't run or load unless 2613 explicitly requested by name. for example: modprobe test_bitops. 2614 2615 If unsure, say N. 2616 2617config TEST_VMALLOC 2618 tristate "Test module for stress/performance analysis of vmalloc allocator" 2619 default n 2620 depends on MMU 2621 help 2622 This builds the "test_vmalloc" module that should be used for 2623 stress and performance analysis. So, any new change for vmalloc 2624 subsystem can be evaluated from performance and stability point 2625 of view. 2626 2627 If unsure, say N. 2628 2629config TEST_BPF 2630 tristate "Test BPF filter functionality" 2631 depends on m && NET 2632 help 2633 This builds the "test_bpf" module that runs various test vectors 2634 against the BPF interpreter or BPF JIT compiler depending on the 2635 current setting. This is in particular useful for BPF JIT compiler 2636 development, but also to run regression tests against changes in 2637 the interpreter code. It also enables test stubs for eBPF maps and 2638 verifier used by user space verifier testsuite. 2639 2640 If unsure, say N. 2641 2642config FIND_BIT_BENCHMARK 2643 tristate "Test find_bit functions" 2644 help 2645 This builds the "test_find_bit" module that measure find_*_bit() 2646 functions performance. 2647 2648 If unsure, say N. 2649 2650config FIND_BIT_BENCHMARK_RUST 2651 tristate "Test find_bit functions in Rust" 2652 depends on RUST 2653 help 2654 This builds the "find_bit_benchmark_rust" module. It is a micro 2655 benchmark that measures the performance of Rust functions that 2656 correspond to the find_*_bit() operations in C. It follows the 2657 FIND_BIT_BENCHMARK closely but will in general not yield same 2658 numbers due to extra bounds checks and overhead of foreign 2659 function calls. 2660 2661 If unsure, say N. 2662 2663config TEST_FIRMWARE 2664 tristate "Test firmware loading via userspace interface" 2665 depends on FW_LOADER 2666 help 2667 This builds the "test_firmware" module that creates a userspace 2668 interface for testing firmware loading. This can be used to 2669 control the triggering of firmware loading without needing an 2670 actual firmware-using device. The contents can be rechecked by 2671 userspace. 2672 2673 If unsure, say N. 2674 2675config TEST_SYSCTL 2676 tristate "sysctl test driver" 2677 depends on PROC_SYSCTL 2678 help 2679 This builds the "test_sysctl" module. This driver enables to test the 2680 proc sysctl interfaces available to drivers safely without affecting 2681 production knobs which might alter system functionality. 2682 2683 If unsure, say N. 2684 2685config BITFIELD_KUNIT 2686 tristate "KUnit test bitfield functions at runtime" if !KUNIT_ALL_TESTS 2687 depends on KUNIT 2688 default KUNIT_ALL_TESTS 2689 help 2690 Enable this option to test the bitfield functions at boot. 2691 2692 KUnit tests run during boot and output the results to the debug log 2693 in TAP format (http://testanything.org/). Only useful for kernel devs 2694 running the KUnit test harness, and not intended for inclusion into a 2695 production build. 2696 2697 For more information on KUnit and unit tests in general please refer 2698 to the KUnit documentation in Documentation/dev-tools/kunit/. 2699 2700 If unsure, say N. 2701 2702config CHECKSUM_KUNIT 2703 tristate "KUnit test checksum functions at runtime" if !KUNIT_ALL_TESTS 2704 depends on KUNIT 2705 default KUNIT_ALL_TESTS 2706 help 2707 Enable this option to test the checksum functions at boot. 2708 2709 KUnit tests run during boot and output the results to the debug log 2710 in TAP format (http://testanything.org/). Only useful for kernel devs 2711 running the KUnit test harness, and not intended for inclusion into a 2712 production build. 2713 2714 For more information on KUnit and unit tests in general please refer 2715 to the KUnit documentation in Documentation/dev-tools/kunit/. 2716 2717 If unsure, say N. 2718 2719config UTIL_MACROS_KUNIT 2720 tristate "KUnit test util_macros.h functions at runtime" if !KUNIT_ALL_TESTS 2721 depends on KUNIT 2722 default KUNIT_ALL_TESTS 2723 help 2724 Enable this option to test the util_macros.h function at boot. 2725 2726 KUnit tests run during boot and output the results to the debug log 2727 in TAP format (http://testanything.org/). Only useful for kernel devs 2728 running the KUnit test harness, and not intended for inclusion into a 2729 production build. 2730 2731 For more information on KUnit and unit tests in general please refer 2732 to the KUnit documentation in Documentation/dev-tools/kunit/. 2733 2734 If unsure, say N. 2735 2736config HASH_KUNIT_TEST 2737 tristate "KUnit Test for integer hash functions" if !KUNIT_ALL_TESTS 2738 depends on KUNIT 2739 default KUNIT_ALL_TESTS 2740 help 2741 Enable this option to test the kernel's string (<linux/stringhash.h>), and 2742 integer (<linux/hash.h>) hash functions on boot. 2743 2744 KUnit tests run during boot and output the results to the debug log 2745 in TAP format (https://testanything.org/). Only useful for kernel devs 2746 running the KUnit test harness, and not intended for inclusion into a 2747 production build. 2748 2749 For more information on KUnit and unit tests in general please refer 2750 to the KUnit documentation in Documentation/dev-tools/kunit/. 2751 2752 This is intended to help people writing architecture-specific 2753 optimized versions. If unsure, say N. 2754 2755config RESOURCE_KUNIT_TEST 2756 tristate "KUnit test for resource API" if !KUNIT_ALL_TESTS 2757 depends on KUNIT 2758 default KUNIT_ALL_TESTS 2759 select GET_FREE_REGION 2760 help 2761 This builds the resource API unit test. 2762 Tests the logic of API provided by resource.c and ioport.h. 2763 For more information on KUnit and unit tests in general please refer 2764 to the KUnit documentation in Documentation/dev-tools/kunit/. 2765 2766 If unsure, say N. 2767 2768config SYSCTL_KUNIT_TEST 2769 tristate "KUnit test for sysctl" if !KUNIT_ALL_TESTS 2770 depends on KUNIT 2771 default KUNIT_ALL_TESTS 2772 help 2773 This builds the proc sysctl unit test, which runs on boot. 2774 Tests the API contract and implementation correctness of sysctl. 2775 For more information on KUnit and unit tests in general please refer 2776 to the KUnit documentation in Documentation/dev-tools/kunit/. 2777 2778 If unsure, say N. 2779 2780config KFIFO_KUNIT_TEST 2781 tristate "KUnit Test for the generic kernel FIFO implementation" if !KUNIT_ALL_TESTS 2782 depends on KUNIT 2783 default KUNIT_ALL_TESTS 2784 help 2785 This builds the generic FIFO implementation KUnit test suite. 2786 It tests that the API and basic functionality of the kfifo type 2787 and associated macros. 2788 2789 For more information on KUnit and unit tests in general please refer 2790 to the KUnit documentation in Documentation/dev-tools/kunit/. 2791 2792 If unsure, say N. 2793 2794config LIST_KUNIT_TEST 2795 tristate "KUnit Test for Kernel Linked-list structures" if !KUNIT_ALL_TESTS 2796 depends on KUNIT 2797 default KUNIT_ALL_TESTS 2798 help 2799 This builds the linked list KUnit test suite. 2800 It tests that the API and basic functionality of the list_head type 2801 and associated macros. 2802 2803 KUnit tests run during boot and output the results to the debug log 2804 in TAP format (https://testanything.org/). Only useful for kernel devs 2805 running the KUnit test harness, and not intended for inclusion into a 2806 production build. 2807 2808 For more information on KUnit and unit tests in general please refer 2809 to the KUnit documentation in Documentation/dev-tools/kunit/. 2810 2811 If unsure, say N. 2812 2813config HASHTABLE_KUNIT_TEST 2814 tristate "KUnit Test for Kernel Hashtable structures" if !KUNIT_ALL_TESTS 2815 depends on KUNIT 2816 default KUNIT_ALL_TESTS 2817 help 2818 This builds the hashtable KUnit test suite. 2819 It tests the basic functionality of the API defined in 2820 include/linux/hashtable.h. For more information on KUnit and 2821 unit tests in general please refer to the KUnit documentation 2822 in Documentation/dev-tools/kunit/. 2823 2824 If unsure, say N. 2825 2826config LINEAR_RANGES_TEST 2827 tristate "KUnit test for linear_ranges" 2828 depends on KUNIT 2829 select LINEAR_RANGES 2830 help 2831 This builds the linear_ranges unit test, which runs on boot. 2832 Tests the linear_ranges logic correctness. 2833 For more information on KUnit and unit tests in general please refer 2834 to the KUnit documentation in Documentation/dev-tools/kunit/. 2835 2836 If unsure, say N. 2837 2838config CONTEXT_ANALYSIS_TEST 2839 bool "Compiler context-analysis warnings test" 2840 depends on EXPERT 2841 help 2842 This builds the test for compiler-based context analysis. The test 2843 does not add executable code to the kernel, but is meant to test that 2844 common patterns supported by the analysis do not result in false 2845 positive warnings. 2846 2847 When adding support for new context locks, it is strongly recommended 2848 to add supported patterns to this test. 2849 2850 If unsure, say N. 2851 2852config CMDLINE_KUNIT_TEST 2853 tristate "KUnit test for cmdline API" if !KUNIT_ALL_TESTS 2854 depends on KUNIT 2855 default KUNIT_ALL_TESTS 2856 help 2857 This builds the cmdline API unit test. 2858 Tests the logic of API provided by cmdline.c. 2859 For more information on KUnit and unit tests in general please refer 2860 to the KUnit documentation in Documentation/dev-tools/kunit/. 2861 2862 If unsure, say N. 2863 2864config BASE64_KUNIT 2865 tristate "KUnit test for base64 decoding and encoding" if !KUNIT_ALL_TESTS 2866 depends on KUNIT 2867 default KUNIT_ALL_TESTS 2868 help 2869 This builds the base64 unit tests. 2870 2871 The tests cover the encoding and decoding logic of Base64 functions 2872 in the kernel. 2873 In addition to correctness checks, simple performance benchmarks 2874 for both encoding and decoding are also included. 2875 2876 For more information on KUnit and unit tests in general please refer 2877 to the KUnit documentation in Documentation/dev-tools/kunit/. 2878 2879 If unsure, say N. 2880 2881config BITS_TEST 2882 tristate "KUnit test for bit functions and macros" if !KUNIT_ALL_TESTS 2883 depends on KUNIT 2884 default KUNIT_ALL_TESTS 2885 help 2886 This builds the bits unit test. 2887 Tests the logic of macros defined in bits.h. 2888 For more information on KUnit and unit tests in general please refer 2889 to the KUnit documentation in Documentation/dev-tools/kunit/. 2890 2891 If unsure, say N. 2892 2893config SLUB_KUNIT_TEST 2894 tristate "KUnit test for SLUB cache error detection" if !KUNIT_ALL_TESTS 2895 depends on SLUB_DEBUG && KUNIT 2896 default KUNIT_ALL_TESTS 2897 help 2898 This builds SLUB allocator unit test. 2899 Tests SLUB cache debugging functionality. 2900 For more information on KUnit and unit tests in general please refer 2901 to the KUnit documentation in Documentation/dev-tools/kunit/. 2902 2903 If unsure, say N. 2904 2905config RATIONAL_KUNIT_TEST 2906 tristate "KUnit test for rational.c" if !KUNIT_ALL_TESTS 2907 depends on KUNIT && RATIONAL 2908 default KUNIT_ALL_TESTS 2909 help 2910 This builds the rational math unit test. 2911 For more information on KUnit and unit tests in general please refer 2912 to the KUnit documentation in Documentation/dev-tools/kunit/. 2913 2914 If unsure, say N. 2915 2916config MEMCPY_KUNIT_TEST 2917 tristate "Test memcpy(), memmove(), and memset() functions at runtime" if !KUNIT_ALL_TESTS 2918 depends on KUNIT 2919 default KUNIT_ALL_TESTS 2920 help 2921 Builds unit tests for memcpy(), memmove(), and memset() functions. 2922 For more information on KUnit and unit tests in general please refer 2923 to the KUnit documentation in Documentation/dev-tools/kunit/. 2924 2925 If unsure, say N. 2926 2927config IS_SIGNED_TYPE_KUNIT_TEST 2928 tristate "Test is_signed_type() macro" if !KUNIT_ALL_TESTS 2929 depends on KUNIT 2930 default KUNIT_ALL_TESTS 2931 help 2932 Builds unit tests for the is_signed_type() macro. 2933 2934 For more information on KUnit and unit tests in general please refer 2935 to the KUnit documentation in Documentation/dev-tools/kunit/. 2936 2937 If unsure, say N. 2938 2939config OVERFLOW_KUNIT_TEST 2940 tristate "Test check_*_overflow() functions at runtime" if !KUNIT_ALL_TESTS 2941 depends on KUNIT 2942 default KUNIT_ALL_TESTS 2943 help 2944 Builds unit tests for the check_*_overflow(), size_*(), allocation, and 2945 related functions. 2946 2947 For more information on KUnit and unit tests in general please refer 2948 to the KUnit documentation in Documentation/dev-tools/kunit/. 2949 2950 If unsure, say N. 2951 2952config RANDSTRUCT_KUNIT_TEST 2953 tristate "Test randstruct structure layout randomization at runtime" if !KUNIT_ALL_TESTS 2954 depends on KUNIT 2955 default KUNIT_ALL_TESTS 2956 help 2957 Builds unit tests for the checking CONFIG_RANDSTRUCT=y, which 2958 randomizes structure layouts. 2959 2960config STACKINIT_KUNIT_TEST 2961 tristate "Test level of stack variable initialization" if !KUNIT_ALL_TESTS 2962 depends on KUNIT 2963 default KUNIT_ALL_TESTS 2964 help 2965 Test if the kernel is zero-initializing stack variables and 2966 padding. Coverage is controlled by compiler flags, 2967 CONFIG_INIT_STACK_ALL_PATTERN or CONFIG_INIT_STACK_ALL_ZERO. 2968 2969config FORTIFY_KUNIT_TEST 2970 tristate "Test fortified str*() and mem*() function internals at runtime" if !KUNIT_ALL_TESTS 2971 depends on KUNIT 2972 default KUNIT_ALL_TESTS 2973 help 2974 Builds unit tests for checking internals of FORTIFY_SOURCE as used 2975 by the str*() and mem*() family of functions. For testing runtime 2976 traps of FORTIFY_SOURCE, see LKDTM's "FORTIFY_*" tests. 2977 2978config LONGEST_SYM_KUNIT_TEST 2979 tristate "Test the longest symbol possible" if !KUNIT_ALL_TESTS 2980 depends on KUNIT && KPROBES 2981 depends on !PREFIX_SYMBOLS && !CFI && !GCOV_KERNEL 2982 default KUNIT_ALL_TESTS 2983 help 2984 Tests the longest symbol possible 2985 2986 If unsure, say N. 2987 2988config HW_BREAKPOINT_KUNIT_TEST 2989 bool "Test hw_breakpoint constraints accounting" if !KUNIT_ALL_TESTS 2990 depends on HAVE_HW_BREAKPOINT 2991 depends on KUNIT=y 2992 default KUNIT_ALL_TESTS 2993 help 2994 Tests for hw_breakpoint constraints accounting. 2995 2996 If unsure, say N. 2997 2998config SIPHASH_KUNIT_TEST 2999 tristate "Perform selftest on siphash functions" if !KUNIT_ALL_TESTS 3000 depends on KUNIT 3001 default KUNIT_ALL_TESTS 3002 help 3003 Enable this option to test the kernel's siphash (<linux/siphash.h>) hash 3004 functions on boot (or module load). 3005 3006 This is intended to help people writing architecture-specific 3007 optimized versions. If unsure, say N. 3008 3009config USERCOPY_KUNIT_TEST 3010 tristate "KUnit Test for user/kernel boundary protections" 3011 depends on KUNIT 3012 default KUNIT_ALL_TESTS 3013 help 3014 This builds the "usercopy_kunit" module that runs sanity checks 3015 on the copy_to/from_user infrastructure, making sure basic 3016 user/kernel boundary testing is working. 3017 3018config BLACKHOLE_DEV_KUNIT_TEST 3019 tristate "Test blackhole netdev functionality" if !KUNIT_ALL_TESTS 3020 depends on NET 3021 depends on KUNIT 3022 default KUNIT_ALL_TESTS 3023 help 3024 This builds the "blackhole_dev_kunit" module that validates the 3025 data path through this blackhole netdev. 3026 3027 If unsure, say N. 3028 3029config TEST_UDELAY 3030 tristate "udelay test driver" 3031 help 3032 This builds the "udelay_test" module that helps to make sure 3033 that udelay() is working properly. 3034 3035 If unsure, say N. 3036 3037config TEST_STATIC_KEYS 3038 tristate "Test static keys" 3039 depends on m 3040 help 3041 Test the static key interfaces. 3042 3043 If unsure, say N. 3044 3045config TEST_DYNAMIC_DEBUG 3046 tristate "Test DYNAMIC_DEBUG" 3047 depends on DYNAMIC_DEBUG 3048 help 3049 This module registers a tracer callback to count enabled 3050 pr_debugs in a 'do_debugging' function, then alters their 3051 enablements, calls the function, and compares counts. 3052 3053 If unsure, say N. 3054 3055config TEST_KMOD 3056 tristate "kmod stress tester" 3057 depends on m 3058 select TEST_LKM 3059 help 3060 Test the kernel's module loading mechanism: kmod. kmod implements 3061 support to load modules using the Linux kernel's usermode helper. 3062 This test provides a series of tests against kmod. 3063 3064 Although technically you can either build test_kmod as a module or 3065 into the kernel we disallow building it into the kernel since 3066 it stress tests request_module() and this will very likely cause 3067 some issues by taking over precious threads available from other 3068 module load requests, ultimately this could be fatal. 3069 3070 To run tests run: 3071 3072 tools/testing/selftests/kmod/kmod.sh --help 3073 3074 If unsure, say N. 3075 3076config TEST_RUNTIME 3077 bool 3078 3079config TEST_RUNTIME_MODULE 3080 bool 3081 3082config TEST_KALLSYMS 3083 tristate "module kallsyms find_symbol() test" 3084 depends on m 3085 select TEST_RUNTIME 3086 select TEST_RUNTIME_MODULE 3087 select TEST_KALLSYMS_A 3088 select TEST_KALLSYMS_B 3089 select TEST_KALLSYMS_C 3090 select TEST_KALLSYMS_D 3091 help 3092 This allows us to stress test find_symbol() through the kallsyms 3093 used to place symbols on the kernel ELF kallsyms and modules kallsyms 3094 where we place kernel symbols such as exported symbols. 3095 3096 We have four test modules: 3097 3098 A: has KALLSYSMS_NUMSYMS exported symbols 3099 B: uses one of A's symbols 3100 C: adds KALLSYMS_SCALE_FACTOR * KALLSYSMS_NUMSYMS exported 3101 D: adds 2 * the symbols than C 3102 3103 We stress test find_symbol() through two means: 3104 3105 1) Upon load of B it will trigger simplify_symbols() to look for the 3106 one symbol it uses from the module A with tons of symbols. This is an 3107 indirect way for us to have B call resolve_symbol_wait() upon module 3108 load. This will eventually call find_symbol() which will eventually 3109 try to find the symbols used with find_exported_symbol_in_section(). 3110 find_exported_symbol_in_section() uses bsearch() so a binary search 3111 for each symbol. Binary search will at worst be O(log(n)) so the 3112 larger TEST_MODULE_KALLSYSMS the worse the search. 3113 3114 2) The selftests should load C first, before B. Upon B's load towards 3115 the end right before we call module B's init routine we get 3116 complete_formation() called on the module. That will first check 3117 for duplicate symbols with the call to verify_exported_symbols(). 3118 That is when we'll force iteration on module C's insane symbol list. 3119 Since it has 10 * KALLSYMS_NUMSYMS it means we can first test 3120 just loading B without C. The amount of time it takes to load C Vs 3121 B can give us an idea of the impact growth of the symbol space and 3122 give us projection. Module A only uses one symbol from B so to allow 3123 this scaling in module C to be proportional, if it used more symbols 3124 then the first test would be doing more and increasing just the 3125 search space would be slightly different. The last module, module D 3126 will just increase the search space by twice the number of symbols in 3127 C so to allow for full projects. 3128 3129 tools/testing/selftests/module/find_symbol.sh 3130 3131 The current defaults will incur a build delay of about 7 minutes 3132 on an x86_64 with only 8 cores. Enable this only if you want to 3133 stress test find_symbol() with thousands of symbols. At the same 3134 time this is also useful to test building modules with thousands of 3135 symbols, and if BTF is enabled this also stress tests adding BTF 3136 information for each module. Currently enabling many more symbols 3137 will segfault the build system. 3138 3139 If unsure, say N. 3140 3141if TEST_KALLSYMS 3142 3143config TEST_KALLSYMS_A 3144 tristate 3145 depends on m 3146 3147config TEST_KALLSYMS_B 3148 tristate 3149 depends on m 3150 3151config TEST_KALLSYMS_C 3152 tristate 3153 depends on m 3154 3155config TEST_KALLSYMS_D 3156 tristate 3157 depends on m 3158 3159choice 3160 prompt "Kallsym test range" 3161 default TEST_KALLSYMS_LARGE 3162 help 3163 Selecting something other than "Fast" will enable tests which slow 3164 down the build and may crash your build. 3165 3166config TEST_KALLSYMS_FAST 3167 bool "Fast builds" 3168 help 3169 You won't really be testing kallsysms, so this just helps fast builds 3170 when allmodconfig is used.. 3171 3172config TEST_KALLSYMS_LARGE 3173 bool "Enable testing kallsyms with large exports" 3174 help 3175 This will enable larger number of symbols. This will slow down 3176 your build considerably. 3177 3178config TEST_KALLSYMS_MAX 3179 bool "Known kallsysms limits" 3180 help 3181 This will enable exports to the point we know we'll start crashing 3182 builds. 3183 3184endchoice 3185 3186config TEST_KALLSYMS_NUMSYMS 3187 int "test kallsyms number of symbols" 3188 range 2 10000 3189 default 2 if TEST_KALLSYMS_FAST 3190 default 100 if TEST_KALLSYMS_LARGE 3191 default 10000 if TEST_KALLSYMS_MAX 3192 help 3193 The number of symbols to create on TEST_KALLSYMS_A, only one of which 3194 module TEST_KALLSYMS_B will use. This also will be used 3195 for how many symbols TEST_KALLSYMS_C will have, scaled up by 3196 TEST_KALLSYMS_SCALE_FACTOR. Note that setting this to 10,000 will 3197 trigger a segfault today, don't use anything close to it unless 3198 you are aware that this should not be used for automated build tests. 3199 3200config TEST_KALLSYMS_SCALE_FACTOR 3201 int "test kallsyms scale factor" 3202 default 8 3203 help 3204 How many more unusued symbols will TEST_KALLSYSMS_C have than 3205 TEST_KALLSYMS_A. If 8, then module C will have 8 * syms 3206 than module A. Then TEST_KALLSYMS_D will have double the amount 3207 of symbols than C so to allow projections. 3208 3209endif # TEST_KALLSYMS 3210 3211config TEST_DEBUG_VIRTUAL 3212 tristate "Test CONFIG_DEBUG_VIRTUAL feature" 3213 depends on DEBUG_VIRTUAL 3214 help 3215 Test the kernel's ability to detect incorrect calls to 3216 virt_to_phys() done against the non-linear part of the 3217 kernel's virtual address map. 3218 3219 If unsure, say N. 3220 3221config TEST_MEMCAT_P 3222 tristate "Test memcat_p() helper function" 3223 help 3224 Test the memcat_p() helper for correctly merging two 3225 pointer arrays together. 3226 3227 If unsure, say N. 3228 3229config TEST_OBJAGG 3230 tristate "Perform selftest on object aggreration manager" 3231 default n 3232 depends on OBJAGG 3233 help 3234 Enable this option to test object aggregation manager on boot 3235 (or module load). 3236 3237config TEST_MEMINIT 3238 tristate "Test heap/page initialization" 3239 help 3240 Test if the kernel is zero-initializing heap and page allocations. 3241 This can be useful to test init_on_alloc and init_on_free features. 3242 3243 If unsure, say N. 3244 3245config TEST_HMM 3246 tristate "Test HMM (Heterogeneous Memory Management)" 3247 depends on TRANSPARENT_HUGEPAGE 3248 depends on DEVICE_PRIVATE 3249 select HMM_MIRROR 3250 select MMU_NOTIFIER 3251 help 3252 This is a pseudo device driver solely for testing HMM. 3253 Say M here if you want to build the HMM test module. 3254 Doing so will allow you to run tools/testing/selftest/vm/hmm-tests. 3255 3256 If unsure, say N. 3257 3258config TEST_FREE_PAGES 3259 tristate "Test freeing pages" 3260 help 3261 Test that a memory leak does not occur due to a race between 3262 freeing a block of pages and a speculative page reference. 3263 Loading this module is safe if your kernel has the bug fixed. 3264 If the bug is not fixed, it will leak gigabytes of memory and 3265 probably OOM your system. 3266 3267config TEST_FPU 3268 tristate "Test floating point operations in kernel space" 3269 depends on ARCH_HAS_KERNEL_FPU_SUPPORT && !KCOV_INSTRUMENT_ALL 3270 help 3271 Enable this option to add /sys/kernel/debug/selftest_helpers/test_fpu 3272 which will trigger a sequence of floating point operations. This is used 3273 for self-testing floating point control register setting in 3274 kernel_fpu_begin(). 3275 3276 If unsure, say N. 3277 3278config TEST_CLOCKSOURCE_WATCHDOG 3279 tristate "Test clocksource watchdog in kernel space" 3280 depends on CLOCKSOURCE_WATCHDOG 3281 help 3282 Enable this option to create a kernel module that will trigger 3283 a test of the clocksource watchdog. This module may be loaded 3284 via modprobe or insmod in which case it will run upon being 3285 loaded, or it may be built in, in which case it will run 3286 shortly after boot. 3287 3288 If unsure, say N. 3289 3290config TEST_OBJPOOL 3291 tristate "Test module for correctness and stress of objpool" 3292 default n 3293 depends on m && DEBUG_KERNEL 3294 help 3295 This builds the "test_objpool" module that should be used for 3296 correctness verification and concurrent testings of objects 3297 allocation and reclamation. 3298 3299 If unsure, say N. 3300 3301config TEST_KEXEC_HANDOVER 3302 bool "Test for Kexec HandOver" 3303 default n 3304 depends on KEXEC_HANDOVER 3305 help 3306 This option enables test for Kexec HandOver (KHO). 3307 The test consists of two parts: saving kernel data before kexec and 3308 restoring the data after kexec and verifying that it was properly 3309 handed over. This test module creates and saves data on the boot of 3310 the first kernel and restores and verifies the data on the boot of 3311 kexec'ed kernel. 3312 3313 For detailed documentation about KHO, see Documentation/core-api/kho. 3314 3315 To run the test run: 3316 3317 tools/testing/selftests/kho/vmtest.sh -h 3318 3319 If unsure, say N. 3320 3321config RATELIMIT_KUNIT_TEST 3322 tristate "KUnit Test for correctness and stress of ratelimit" if !KUNIT_ALL_TESTS 3323 depends on KUNIT 3324 default KUNIT_ALL_TESTS 3325 help 3326 This builds the "test_ratelimit" module that should be used 3327 for correctness verification and concurrent testings of rate 3328 limiting. 3329 3330 If unsure, say N. 3331 3332config INT_POW_KUNIT_TEST 3333 tristate "Integer exponentiation (int_pow) test" if !KUNIT_ALL_TESTS 3334 depends on KUNIT 3335 default KUNIT_ALL_TESTS 3336 help 3337 This option enables the KUnit test suite for the int_pow function, 3338 which performs integer exponentiation. The test suite is designed to 3339 verify that the implementation of int_pow correctly computes the power 3340 of a given base raised to a given exponent. 3341 3342 Enabling this option will include tests that check various scenarios 3343 and edge cases to ensure the accuracy and reliability of the exponentiation 3344 function. 3345 3346 If unsure, say N 3347 3348config INT_SQRT_KUNIT_TEST 3349 tristate "Integer square root test" if !KUNIT_ALL_TESTS 3350 depends on KUNIT 3351 default KUNIT_ALL_TESTS 3352 help 3353 This option enables the KUnit test suite for the int_sqrt() function, 3354 which performs square root calculation. The test suite checks 3355 various scenarios, including edge cases, to ensure correctness. 3356 3357 Enabling this option will include tests that check various scenarios 3358 and edge cases to ensure the accuracy and reliability of the square root 3359 function. 3360 3361 If unsure, say N 3362 3363config INT_LOG_KUNIT_TEST 3364 tristate "Integer log (int_log) test" if !KUNIT_ALL_TESTS 3365 depends on KUNIT 3366 default KUNIT_ALL_TESTS 3367 help 3368 This option enables the KUnit test suite for the int_log library, which 3369 provides two functions to compute the integer logarithm in base 2 and 3370 base 10, called respectively as intlog2 and intlog10. 3371 3372 If unsure, say N 3373 3374config GCD_KUNIT_TEST 3375 tristate "Greatest common divisor test" if !KUNIT_ALL_TESTS 3376 depends on KUNIT 3377 default KUNIT_ALL_TESTS 3378 help 3379 This option enables the KUnit test suite for the gcd() function, 3380 which computes the greatest common divisor of two numbers. 3381 3382 This test suite verifies the correctness of gcd() across various 3383 scenarios, including edge cases. 3384 3385 If unsure, say N 3386 3387config PRIME_NUMBERS_KUNIT_TEST 3388 tristate "Prime number generator test" if !KUNIT_ALL_TESTS 3389 depends on KUNIT 3390 depends on PRIME_NUMBERS 3391 default KUNIT_ALL_TESTS 3392 help 3393 This option enables the KUnit test suite for the {is,next}_prime_number 3394 functions. 3395 3396 Enabling this option will include tests that compare the prime number 3397 generator functions against a brute force implementation. 3398 3399 If unsure, say N 3400 3401endif # RUNTIME_TESTING_MENU 3402 3403config ARCH_USE_MEMTEST 3404 bool 3405 help 3406 An architecture should select this when it uses early_memtest() 3407 during boot process. 3408 3409config MEMTEST 3410 bool "Memtest" 3411 depends on ARCH_USE_MEMTEST 3412 help 3413 This option adds a kernel parameter 'memtest', which allows memtest 3414 to be set and executed. 3415 memtest=0, mean disabled; -- default 3416 memtest=1, mean do 1 test pattern; 3417 ... 3418 memtest=17, mean do 17 test patterns. 3419 If you are unsure how to answer this question, answer N. 3420 3421 3422 3423config HYPERV_TESTING 3424 bool "Microsoft Hyper-V driver testing" 3425 default n 3426 depends on HYPERV && DEBUG_FS 3427 help 3428 Select this option to enable Hyper-V vmbus testing. 3429 3430endmenu # "Kernel Testing and Coverage" 3431 3432menu "Rust hacking" 3433 3434config RUST_DEBUG_ASSERTIONS 3435 bool "Debug assertions" 3436 depends on RUST 3437 help 3438 Enables rustc's `-Cdebug-assertions` codegen option. 3439 3440 This flag lets you turn `cfg(debug_assertions)` conditional 3441 compilation on or off. This can be used to enable extra debugging 3442 code in development but not in production. For example, it controls 3443 the behavior of the standard library's `debug_assert!` macro. 3444 3445 Note that this will apply to all Rust code, including `core`. 3446 3447 If unsure, say N. 3448 3449config RUST_OVERFLOW_CHECKS 3450 bool "Overflow checks" 3451 default y 3452 depends on RUST 3453 help 3454 Enables rustc's `-Coverflow-checks` codegen option. 3455 3456 This flag allows you to control the behavior of runtime integer 3457 overflow. When overflow-checks are enabled, a Rust panic will occur 3458 on overflow. 3459 3460 Note that this will apply to all Rust code, including `core`. 3461 3462 If unsure, say Y. 3463 3464config RUST_BUILD_ASSERT_ALLOW 3465 bool "Allow unoptimized build-time assertions" 3466 depends on RUST 3467 help 3468 Controls how `build_error!` and `build_assert!` are handled during the build. 3469 3470 If calls to them exist in the binary, it may indicate a violated invariant 3471 or that the optimizer failed to verify the invariant during compilation. 3472 3473 This should not happen, thus by default the build is aborted. However, 3474 as an escape hatch, you can choose Y here to ignore them during build 3475 and let the check be carried at runtime (with `panic!` being called if 3476 the check fails). 3477 3478 If unsure, say N. 3479 3480config RUST_KERNEL_DOCTESTS 3481 bool "Doctests for the `kernel` crate" if !KUNIT_ALL_TESTS 3482 depends on RUST && KUNIT=y 3483 default KUNIT_ALL_TESTS 3484 help 3485 This builds the documentation tests of the `kernel` crate 3486 as KUnit tests. 3487 3488 For more information on KUnit and unit tests in general, 3489 please refer to the KUnit documentation in Documentation/dev-tools/kunit/. 3490 3491 If unsure, say N. 3492 3493endmenu # "Rust" 3494 3495endmenu # Kernel hacking 3496