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 CONSOLE_LOGLEVEL_DEFAULT 39 int "Default console loglevel (1-15)" 40 range 1 15 41 default "7" 42 help 43 Default loglevel to determine what will be printed on the console. 44 45 Setting a default here is equivalent to passing in loglevel=<x> in 46 the kernel bootargs. loglevel=<x> continues to override whatever 47 value is specified here as well. 48 49 Note: This does not affect the log level of un-prefixed printk() 50 usage in the kernel. That is controlled by the MESSAGE_LOGLEVEL_DEFAULT 51 option. 52 53config CONSOLE_LOGLEVEL_QUIET 54 int "quiet console loglevel (1-15)" 55 range 1 15 56 default "4" 57 help 58 loglevel to use when "quiet" is passed on the kernel commandline. 59 60 When "quiet" is passed on the kernel commandline this loglevel 61 will be used as the loglevel. IOW passing "quiet" will be the 62 equivalent of passing "loglevel=<CONSOLE_LOGLEVEL_QUIET>" 63 64config MESSAGE_LOGLEVEL_DEFAULT 65 int "Default message log level (1-7)" 66 range 1 7 67 default "4" 68 help 69 Default log level for printk statements with no specified priority. 70 71 This was hard-coded to KERN_WARNING since at least 2.6.10 but folks 72 that are auditing their logs closely may want to set it to a lower 73 priority. 74 75 Note: This does not affect what message level gets printed on the console 76 by default. To change that, use loglevel=<x> in the kernel bootargs, 77 or pick a different CONSOLE_LOGLEVEL_DEFAULT configuration value. 78 79config BOOT_PRINTK_DELAY 80 bool "Delay each boot printk message by N milliseconds" 81 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY 82 help 83 This build option allows you to read kernel boot messages 84 by inserting a short delay after each one. The delay is 85 specified in milliseconds on the kernel command line, 86 using "boot_delay=N". 87 88 It is likely that you would also need to use "lpj=M" to preset 89 the "loops per jiffie" value. 90 See a previous boot log for the "lpj" value to use for your 91 system, and then set "lpj=M" before setting "boot_delay=N". 92 NOTE: Using this option may adversely affect SMP systems. 93 I.e., processors other than the first one may not boot up. 94 BOOT_PRINTK_DELAY also may cause LOCKUP_DETECTOR to detect 95 what it believes to be lockup conditions. 96 97config DYNAMIC_DEBUG 98 bool "Enable dynamic printk() support" 99 default n 100 depends on PRINTK 101 depends on (DEBUG_FS || PROC_FS) 102 select DYNAMIC_DEBUG_CORE 103 help 104 105 Compiles debug level messages into the kernel, which would not 106 otherwise be available at runtime. These messages can then be 107 enabled/disabled based on various levels of scope - per source file, 108 function, module, format string, and line number. This mechanism 109 implicitly compiles in all pr_debug() and dev_dbg() calls, which 110 enlarges the kernel text size by about 2%. 111 112 If a source file is compiled with DEBUG flag set, any 113 pr_debug() calls in it are enabled by default, but can be 114 disabled at runtime as below. Note that DEBUG flag is 115 turned on by many CONFIG_*DEBUG* options. 116 117 Usage: 118 119 Dynamic debugging is controlled via the 'dynamic_debug/control' file, 120 which is contained in the 'debugfs' filesystem or procfs. 121 Thus, the debugfs or procfs filesystem must first be mounted before 122 making use of this feature. 123 We refer the control file as: <debugfs>/dynamic_debug/control. This 124 file contains a list of the debug statements that can be enabled. The 125 format for each line of the file is: 126 127 filename:lineno [module]function flags format 128 129 filename : source file of the debug statement 130 lineno : line number of the debug statement 131 module : module that contains the debug statement 132 function : function that contains the debug statement 133 flags : '=p' means the line is turned 'on' for printing 134 format : the format used for the debug statement 135 136 From a live system: 137 138 nullarbor:~ # cat <debugfs>/dynamic_debug/control 139 # filename:lineno [module]function flags format 140 fs/aio.c:222 [aio]__put_ioctx =_ "__put_ioctx:\040freeing\040%p\012" 141 fs/aio.c:248 [aio]ioctx_alloc =_ "ENOMEM:\040nr_events\040too\040high\012" 142 fs/aio.c:1770 [aio]sys_io_cancel =_ "calling\040cancel\012" 143 144 Example usage: 145 146 // enable the message at line 1603 of file svcsock.c 147 nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' > 148 <debugfs>/dynamic_debug/control 149 150 // enable all the messages in file svcsock.c 151 nullarbor:~ # echo -n 'file svcsock.c +p' > 152 <debugfs>/dynamic_debug/control 153 154 // enable all the messages in the NFS server module 155 nullarbor:~ # echo -n 'module nfsd +p' > 156 <debugfs>/dynamic_debug/control 157 158 // enable all 12 messages in the function svc_process() 159 nullarbor:~ # echo -n 'func svc_process +p' > 160 <debugfs>/dynamic_debug/control 161 162 // disable all 12 messages in the function svc_process() 163 nullarbor:~ # echo -n 'func svc_process -p' > 164 <debugfs>/dynamic_debug/control 165 166 See Documentation/admin-guide/dynamic-debug-howto.rst for additional 167 information. 168 169config DYNAMIC_DEBUG_CORE 170 bool "Enable core function of dynamic debug support" 171 depends on PRINTK 172 depends on (DEBUG_FS || PROC_FS) 173 help 174 Enable core functional support of dynamic debug. It is useful 175 when you want to tie dynamic debug to your kernel modules with 176 DYNAMIC_DEBUG_MODULE defined for each of them, especially for 177 the case of embedded system where the kernel image size is 178 sensitive for people. 179 180config SYMBOLIC_ERRNAME 181 bool "Support symbolic error names in printf" 182 default y if PRINTK 183 help 184 If you say Y here, the kernel's printf implementation will 185 be able to print symbolic error names such as ENOSPC instead 186 of the number 28. It makes the kernel image slightly larger 187 (about 3KB), but can make the kernel logs easier to read. 188 189config DEBUG_BUGVERBOSE 190 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT 191 depends on BUG && (GENERIC_BUG || HAVE_DEBUG_BUGVERBOSE) 192 default y 193 help 194 Say Y here to make BUG() panics output the file name and line number 195 of the BUG call as well as the EIP and oops trace. This aids 196 debugging but costs about 70-100K of memory. 197 198endmenu # "printk and dmesg options" 199 200menu "Compile-time checks and compiler options" 201 202config DEBUG_INFO 203 bool "Compile the kernel with debug info" 204 depends on DEBUG_KERNEL && !COMPILE_TEST 205 help 206 If you say Y here the resulting kernel image will include 207 debugging info resulting in a larger kernel image. 208 This adds debug symbols to the kernel and modules (gcc -g), and 209 is needed if you intend to use kernel crashdump or binary object 210 tools like crash, kgdb, LKCD, gdb, etc on the kernel. 211 Say Y here only if you plan to debug the kernel. 212 213 If unsure, say N. 214 215config DEBUG_INFO_REDUCED 216 bool "Reduce debugging information" 217 depends on DEBUG_INFO 218 help 219 If you say Y here gcc is instructed to generate less debugging 220 information for structure types. This means that tools that 221 need full debugging information (like kgdb or systemtap) won't 222 be happy. But if you merely need debugging information to 223 resolve line numbers there is no loss. Advantage is that 224 build directory object sizes shrink dramatically over a full 225 DEBUG_INFO build and compile times are reduced too. 226 Only works with newer gcc versions. 227 228config DEBUG_INFO_COMPRESSED 229 bool "Compressed debugging information" 230 depends on DEBUG_INFO 231 depends on $(cc-option,-gz=zlib) 232 depends on $(as-option,-Wa$(comma)--compress-debug-sections=zlib) 233 depends on $(ld-option,--compress-debug-sections=zlib) 234 help 235 Compress the debug information using zlib. Requires GCC 5.0+ or Clang 236 5.0+, binutils 2.26+, and zlib. 237 238 Users of dpkg-deb via scripts/package/builddeb may find an increase in 239 size of their debug .deb packages with this config set, due to the 240 debug info being compressed with zlib, then the object files being 241 recompressed with a different compression scheme. But this is still 242 preferable to setting $KDEB_COMPRESS to "none" which would be even 243 larger. 244 245config DEBUG_INFO_SPLIT 246 bool "Produce split debuginfo in .dwo files" 247 depends on DEBUG_INFO 248 depends on $(cc-option,-gsplit-dwarf) 249 help 250 Generate debug info into separate .dwo files. This significantly 251 reduces the build directory size for builds with DEBUG_INFO, 252 because it stores the information only once on disk in .dwo 253 files instead of multiple times in object files and executables. 254 In addition the debug information is also compressed. 255 256 Requires recent gcc (4.7+) and recent gdb/binutils. 257 Any tool that packages or reads debug information would need 258 to know about the .dwo files and include them. 259 Incompatible with older versions of ccache. 260 261config DEBUG_INFO_DWARF4 262 bool "Generate dwarf4 debuginfo" 263 depends on DEBUG_INFO 264 depends on $(cc-option,-gdwarf-4) 265 help 266 Generate dwarf4 debug info. This requires recent versions 267 of gcc and gdb. It makes the debug information larger. 268 But it significantly improves the success of resolving 269 variables in gdb on optimized code. 270 271config DEBUG_INFO_BTF 272 bool "Generate BTF typeinfo" 273 depends on DEBUG_INFO 274 depends on !DEBUG_INFO_SPLIT && !DEBUG_INFO_REDUCED 275 depends on !GCC_PLUGIN_RANDSTRUCT || COMPILE_TEST 276 help 277 Generate deduplicated BTF type information from DWARF debug info. 278 Turning this on expects presence of pahole tool, which will convert 279 DWARF type info into equivalent deduplicated BTF type info. 280 281config GDB_SCRIPTS 282 bool "Provide GDB scripts for kernel debugging" 283 depends on DEBUG_INFO 284 help 285 This creates the required links to GDB helper scripts in the 286 build directory. If you load vmlinux into gdb, the helper 287 scripts will be automatically imported by gdb as well, and 288 additional functions are available to analyze a Linux kernel 289 instance. See Documentation/dev-tools/gdb-kernel-debugging.rst 290 for further details. 291 292config ENABLE_MUST_CHECK 293 bool "Enable __must_check logic" 294 default y 295 help 296 Enable the __must_check logic in the kernel build. Disable this to 297 suppress the "warning: ignoring return value of 'foo', declared with 298 attribute warn_unused_result" messages. 299 300config FRAME_WARN 301 int "Warn for stack frames larger than" 302 range 0 8192 303 default 2048 if GCC_PLUGIN_LATENT_ENTROPY 304 default 1280 if (!64BIT && PARISC) 305 default 1024 if (!64BIT && !PARISC) 306 default 2048 if 64BIT 307 help 308 Tell gcc to warn at build time for stack frames larger than this. 309 Setting this too low will cause a lot of warnings. 310 Setting it to 0 disables the warning. 311 312config STRIP_ASM_SYMS 313 bool "Strip assembler-generated symbols during link" 314 default n 315 help 316 Strip internal assembler-generated symbols during a link (symbols 317 that look like '.Lxxx') so they don't pollute the output of 318 get_wchan() and suchlike. 319 320config READABLE_ASM 321 bool "Generate readable assembler code" 322 depends on DEBUG_KERNEL 323 help 324 Disable some compiler optimizations that tend to generate human unreadable 325 assembler output. This may make the kernel slightly slower, but it helps 326 to keep kernel developers who have to stare a lot at assembler listings 327 sane. 328 329config HEADERS_INSTALL 330 bool "Install uapi headers to usr/include" 331 depends on !UML 332 help 333 This option will install uapi headers (headers exported to user-space) 334 into the usr/include directory for use during the kernel build. 335 This is unneeded for building the kernel itself, but needed for some 336 user-space program samples. It is also needed by some features such 337 as uapi header sanity checks. 338 339config DEBUG_SECTION_MISMATCH 340 bool "Enable full Section mismatch analysis" 341 help 342 The section mismatch analysis checks if there are illegal 343 references from one section to another section. 344 During linktime or runtime, some sections are dropped; 345 any use of code/data previously in these sections would 346 most likely result in an oops. 347 In the code, functions and variables are annotated with 348 __init,, etc. (see the full list in include/linux/init.h), 349 which results in the code/data being placed in specific sections. 350 The section mismatch analysis is always performed after a full 351 kernel build, and enabling this option causes the following 352 additional step to occur: 353 - Add the option -fno-inline-functions-called-once to gcc commands. 354 When inlining a function annotated with __init in a non-init 355 function, we would lose the section information and thus 356 the analysis would not catch the illegal reference. 357 This option tells gcc to inline less (but it does result in 358 a larger kernel). 359 360config SECTION_MISMATCH_WARN_ONLY 361 bool "Make section mismatch errors non-fatal" 362 default y 363 help 364 If you say N here, the build process will fail if there are any 365 section mismatch, instead of just throwing warnings. 366 367 If unsure, say Y. 368 369# 370# Select this config option from the architecture Kconfig, if it 371# is preferred to always offer frame pointers as a config 372# option on the architecture (regardless of KERNEL_DEBUG): 373# 374config ARCH_WANT_FRAME_POINTERS 375 bool 376 377config FRAME_POINTER 378 bool "Compile the kernel with frame pointers" 379 depends on DEBUG_KERNEL && (M68K || UML || SUPERH) || ARCH_WANT_FRAME_POINTERS 380 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS 381 help 382 If you say Y here the resulting kernel image will be slightly 383 larger and slower, but it gives very useful debugging information 384 in case of kernel bugs. (precise oopses/stacktraces/warnings) 385 386config STACK_VALIDATION 387 bool "Compile-time stack metadata validation" 388 depends on HAVE_STACK_VALIDATION 389 default n 390 help 391 Add compile-time checks to validate stack metadata, including frame 392 pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure 393 that runtime stack traces are more reliable. 394 395 This is also a prerequisite for generation of ORC unwind data, which 396 is needed for CONFIG_UNWINDER_ORC. 397 398 For more information, see 399 tools/objtool/Documentation/stack-validation.txt. 400 401config VMLINUX_VALIDATION 402 bool 403 depends on STACK_VALIDATION && DEBUG_ENTRY && !PARAVIRT 404 default y 405 406config DEBUG_FORCE_WEAK_PER_CPU 407 bool "Force weak per-cpu definitions" 408 depends on DEBUG_KERNEL 409 help 410 s390 and alpha require percpu variables in modules to be 411 defined weak to work around addressing range issue which 412 puts the following two restrictions on percpu variable 413 definitions. 414 415 1. percpu symbols must be unique whether static or not 416 2. percpu variables can't be defined inside a function 417 418 To ensure that generic code follows the above rules, this 419 option forces all percpu variables to be defined as weak. 420 421endmenu # "Compiler options" 422 423menu "Generic Kernel Debugging Instruments" 424 425config MAGIC_SYSRQ 426 bool "Magic SysRq key" 427 depends on !UML 428 help 429 If you say Y here, you will have some control over the system even 430 if the system crashes for example during kernel debugging (e.g., you 431 will be able to flush the buffer cache to disk, reboot the system 432 immediately or dump some status information). This is accomplished 433 by pressing various keys while holding SysRq (Alt+PrintScreen). It 434 also works on a serial console (on PC hardware at least), if you 435 send a BREAK and then within 5 seconds a command keypress. The 436 keys are documented in <file:Documentation/admin-guide/sysrq.rst>. 437 Don't say Y unless you really know what this hack does. 438 439config MAGIC_SYSRQ_DEFAULT_ENABLE 440 hex "Enable magic SysRq key functions by default" 441 depends on MAGIC_SYSRQ 442 default 0x1 443 help 444 Specifies which SysRq key functions are enabled by default. 445 This may be set to 1 or 0 to enable or disable them all, or 446 to a bitmask as described in Documentation/admin-guide/sysrq.rst. 447 448config MAGIC_SYSRQ_SERIAL 449 bool "Enable magic SysRq key over serial" 450 depends on MAGIC_SYSRQ 451 default y 452 help 453 Many embedded boards have a disconnected TTL level serial which can 454 generate some garbage that can lead to spurious false sysrq detects. 455 This option allows you to decide whether you want to enable the 456 magic SysRq key. 457 458config MAGIC_SYSRQ_SERIAL_SEQUENCE 459 string "Char sequence that enables magic SysRq over serial" 460 depends on MAGIC_SYSRQ_SERIAL 461 default "" 462 help 463 Specifies a sequence of characters that can follow BREAK to enable 464 SysRq on a serial console. 465 466 If unsure, leave an empty string and the option will not be enabled. 467 468config DEBUG_FS 469 bool "Debug Filesystem" 470 help 471 debugfs is a virtual file system that kernel developers use to put 472 debugging files into. Enable this option to be able to read and 473 write to these files. 474 475 For detailed documentation on the debugfs API, see 476 Documentation/filesystems/. 477 478 If unsure, say N. 479 480source "lib/Kconfig.kgdb" 481 482source "lib/Kconfig.ubsan" 483 484endmenu 485 486config DEBUG_KERNEL 487 bool "Kernel debugging" 488 help 489 Say Y here if you are developing drivers or trying to debug and 490 identify kernel problems. 491 492config DEBUG_MISC 493 bool "Miscellaneous debug code" 494 default DEBUG_KERNEL 495 depends on DEBUG_KERNEL 496 help 497 Say Y here if you need to enable miscellaneous debug code that should 498 be under a more specific debug option but isn't. 499 500 501menu "Memory Debugging" 502 503source "mm/Kconfig.debug" 504 505config DEBUG_OBJECTS 506 bool "Debug object operations" 507 depends on DEBUG_KERNEL 508 help 509 If you say Y here, additional code will be inserted into the 510 kernel to track the life time of various objects and validate 511 the operations on those objects. 512 513config DEBUG_OBJECTS_SELFTEST 514 bool "Debug objects selftest" 515 depends on DEBUG_OBJECTS 516 help 517 This enables the selftest of the object debug code. 518 519config DEBUG_OBJECTS_FREE 520 bool "Debug objects in freed memory" 521 depends on DEBUG_OBJECTS 522 help 523 This enables checks whether a k/v free operation frees an area 524 which contains an object which has not been deactivated 525 properly. This can make kmalloc/kfree-intensive workloads 526 much slower. 527 528config DEBUG_OBJECTS_TIMERS 529 bool "Debug timer objects" 530 depends on DEBUG_OBJECTS 531 help 532 If you say Y here, additional code will be inserted into the 533 timer routines to track the life time of timer objects and 534 validate the timer operations. 535 536config DEBUG_OBJECTS_WORK 537 bool "Debug work objects" 538 depends on DEBUG_OBJECTS 539 help 540 If you say Y here, additional code will be inserted into the 541 work queue routines to track the life time of work objects and 542 validate the work operations. 543 544config DEBUG_OBJECTS_RCU_HEAD 545 bool "Debug RCU callbacks objects" 546 depends on DEBUG_OBJECTS 547 help 548 Enable this to turn on debugging of RCU list heads (call_rcu() usage). 549 550config DEBUG_OBJECTS_PERCPU_COUNTER 551 bool "Debug percpu counter objects" 552 depends on DEBUG_OBJECTS 553 help 554 If you say Y here, additional code will be inserted into the 555 percpu counter routines to track the life time of percpu counter 556 objects and validate the percpu counter operations. 557 558config DEBUG_OBJECTS_ENABLE_DEFAULT 559 int "debug_objects bootup default value (0-1)" 560 range 0 1 561 default "1" 562 depends on DEBUG_OBJECTS 563 help 564 Debug objects boot parameter default value 565 566config DEBUG_SLAB 567 bool "Debug slab memory allocations" 568 depends on DEBUG_KERNEL && SLAB 569 help 570 Say Y here to have the kernel do limited verification on memory 571 allocation as well as poisoning memory on free to catch use of freed 572 memory. This can make kmalloc/kfree-intensive workloads much slower. 573 574config SLUB_DEBUG_ON 575 bool "SLUB debugging on by default" 576 depends on SLUB && SLUB_DEBUG 577 default n 578 help 579 Boot with debugging on by default. SLUB boots by default with 580 the runtime debug capabilities switched off. Enabling this is 581 equivalent to specifying the "slub_debug" parameter on boot. 582 There is no support for more fine grained debug control like 583 possible with slub_debug=xxx. SLUB debugging may be switched 584 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying 585 "slub_debug=-". 586 587config SLUB_STATS 588 default n 589 bool "Enable SLUB performance statistics" 590 depends on SLUB && SYSFS 591 help 592 SLUB statistics are useful to debug SLUBs allocation behavior in 593 order find ways to optimize the allocator. This should never be 594 enabled for production use since keeping statistics slows down 595 the allocator by a few percentage points. The slabinfo command 596 supports the determination of the most active slabs to figure 597 out which slabs are relevant to a particular load. 598 Try running: slabinfo -DA 599 600config HAVE_DEBUG_KMEMLEAK 601 bool 602 603config DEBUG_KMEMLEAK 604 bool "Kernel memory leak detector" 605 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK 606 select DEBUG_FS 607 select STACKTRACE if STACKTRACE_SUPPORT 608 select KALLSYMS 609 select CRC32 610 help 611 Say Y here if you want to enable the memory leak 612 detector. The memory allocation/freeing is traced in a way 613 similar to the Boehm's conservative garbage collector, the 614 difference being that the orphan objects are not freed but 615 only shown in /sys/kernel/debug/kmemleak. Enabling this 616 feature will introduce an overhead to memory 617 allocations. See Documentation/dev-tools/kmemleak.rst for more 618 details. 619 620 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances 621 of finding leaks due to the slab objects poisoning. 622 623 In order to access the kmemleak file, debugfs needs to be 624 mounted (usually at /sys/kernel/debug). 625 626config DEBUG_KMEMLEAK_MEM_POOL_SIZE 627 int "Kmemleak memory pool size" 628 depends on DEBUG_KMEMLEAK 629 range 200 1000000 630 default 16000 631 help 632 Kmemleak must track all the memory allocations to avoid 633 reporting false positives. Since memory may be allocated or 634 freed before kmemleak is fully initialised, use a static pool 635 of metadata objects to track such callbacks. After kmemleak is 636 fully initialised, this memory pool acts as an emergency one 637 if slab allocations fail. 638 639config DEBUG_KMEMLEAK_TEST 640 tristate "Simple test for the kernel memory leak detector" 641 depends on DEBUG_KMEMLEAK && m 642 help 643 This option enables a module that explicitly leaks memory. 644 645 If unsure, say N. 646 647config DEBUG_KMEMLEAK_DEFAULT_OFF 648 bool "Default kmemleak to off" 649 depends on DEBUG_KMEMLEAK 650 help 651 Say Y here to disable kmemleak by default. It can then be enabled 652 on the command line via kmemleak=on. 653 654config DEBUG_KMEMLEAK_AUTO_SCAN 655 bool "Enable kmemleak auto scan thread on boot up" 656 default y 657 depends on DEBUG_KMEMLEAK 658 help 659 Depending on the cpu, kmemleak scan may be cpu intensive and can 660 stall user tasks at times. This option enables/disables automatic 661 kmemleak scan at boot up. 662 663 Say N here to disable kmemleak auto scan thread to stop automatic 664 scanning. Disabling this option disables automatic reporting of 665 memory leaks. 666 667 If unsure, say Y. 668 669config DEBUG_STACK_USAGE 670 bool "Stack utilization instrumentation" 671 depends on DEBUG_KERNEL && !IA64 672 help 673 Enables the display of the minimum amount of free stack which each 674 task has ever had available in the sysrq-T and sysrq-P debug output. 675 676 This option will slow down process creation somewhat. 677 678config SCHED_STACK_END_CHECK 679 bool "Detect stack corruption on calls to schedule()" 680 depends on DEBUG_KERNEL 681 default n 682 help 683 This option checks for a stack overrun on calls to schedule(). 684 If the stack end location is found to be over written always panic as 685 the content of the corrupted region can no longer be trusted. 686 This is to ensure no erroneous behaviour occurs which could result in 687 data corruption or a sporadic crash at a later stage once the region 688 is examined. The runtime overhead introduced is minimal. 689 690config ARCH_HAS_DEBUG_VM_PGTABLE 691 bool 692 help 693 An architecture should select this when it can successfully 694 build and run DEBUG_VM_PGTABLE. 695 696config DEBUG_VM 697 bool "Debug VM" 698 depends on DEBUG_KERNEL 699 help 700 Enable this to turn on extended checks in the virtual-memory system 701 that may impact performance. 702 703 If unsure, say N. 704 705config DEBUG_VM_VMACACHE 706 bool "Debug VMA caching" 707 depends on DEBUG_VM 708 help 709 Enable this to turn on VMA caching debug information. Doing so 710 can cause significant overhead, so only enable it in non-production 711 environments. 712 713 If unsure, say N. 714 715config DEBUG_VM_RB 716 bool "Debug VM red-black trees" 717 depends on DEBUG_VM 718 help 719 Enable VM red-black tree debugging information and extra validations. 720 721 If unsure, say N. 722 723config DEBUG_VM_PGFLAGS 724 bool "Debug page-flags operations" 725 depends on DEBUG_VM 726 help 727 Enables extra validation on page flags operations. 728 729 If unsure, say N. 730 731config DEBUG_VM_PGTABLE 732 bool "Debug arch page table for semantics compliance" 733 depends on MMU 734 depends on ARCH_HAS_DEBUG_VM_PGTABLE 735 default y if DEBUG_VM 736 help 737 This option provides a debug method which can be used to test 738 architecture page table helper functions on various platforms in 739 verifying if they comply with expected generic MM semantics. This 740 will help architecture code in making sure that any changes or 741 new additions of these helpers still conform to expected 742 semantics of the generic MM. Platforms will have to opt in for 743 this through ARCH_HAS_DEBUG_VM_PGTABLE. 744 745 If unsure, say N. 746 747config ARCH_HAS_DEBUG_VIRTUAL 748 bool 749 750config DEBUG_VIRTUAL 751 bool "Debug VM translations" 752 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL 753 help 754 Enable some costly sanity checks in virtual to page code. This can 755 catch mistakes with virt_to_page() and friends. 756 757 If unsure, say N. 758 759config DEBUG_NOMMU_REGIONS 760 bool "Debug the global anon/private NOMMU mapping region tree" 761 depends on DEBUG_KERNEL && !MMU 762 help 763 This option causes the global tree of anonymous and private mapping 764 regions to be regularly checked for invalid topology. 765 766config DEBUG_MEMORY_INIT 767 bool "Debug memory initialisation" if EXPERT 768 default !EXPERT 769 help 770 Enable this for additional checks during memory initialisation. 771 The sanity checks verify aspects of the VM such as the memory model 772 and other information provided by the architecture. Verbose 773 information will be printed at KERN_DEBUG loglevel depending 774 on the mminit_loglevel= command-line option. 775 776 If unsure, say Y 777 778config MEMORY_NOTIFIER_ERROR_INJECT 779 tristate "Memory hotplug notifier error injection module" 780 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION 781 help 782 This option provides the ability to inject artificial errors to 783 memory hotplug notifier chain callbacks. It is controlled through 784 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory 785 786 If the notifier call chain should be failed with some events 787 notified, write the error code to "actions/<notifier event>/error". 788 789 Example: Inject memory hotplug offline error (-12 == -ENOMEM) 790 791 # cd /sys/kernel/debug/notifier-error-inject/memory 792 # echo -12 > actions/MEM_GOING_OFFLINE/error 793 # echo offline > /sys/devices/system/memory/memoryXXX/state 794 bash: echo: write error: Cannot allocate memory 795 796 To compile this code as a module, choose M here: the module will 797 be called memory-notifier-error-inject. 798 799 If unsure, say N. 800 801config DEBUG_PER_CPU_MAPS 802 bool "Debug access to per_cpu maps" 803 depends on DEBUG_KERNEL 804 depends on SMP 805 help 806 Say Y to verify that the per_cpu map being accessed has 807 been set up. This adds a fair amount of code to kernel memory 808 and decreases performance. 809 810 Say N if unsure. 811 812config DEBUG_HIGHMEM 813 bool "Highmem debugging" 814 depends on DEBUG_KERNEL && HIGHMEM 815 help 816 This option enables additional error checking for high memory 817 systems. Disable for production systems. 818 819config HAVE_DEBUG_STACKOVERFLOW 820 bool 821 822config DEBUG_STACKOVERFLOW 823 bool "Check for stack overflows" 824 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW 825 help 826 Say Y here if you want to check for overflows of kernel, IRQ 827 and exception stacks (if your architecture uses them). This 828 option will show detailed messages if free stack space drops 829 below a certain limit. 830 831 These kinds of bugs usually occur when call-chains in the 832 kernel get too deep, especially when interrupts are 833 involved. 834 835 Use this in cases where you see apparently random memory 836 corruption, especially if it appears in 'struct thread_info' 837 838 If in doubt, say "N". 839 840source "lib/Kconfig.kasan" 841 842endmenu # "Memory Debugging" 843 844config DEBUG_SHIRQ 845 bool "Debug shared IRQ handlers" 846 depends on DEBUG_KERNEL 847 help 848 Enable this to generate a spurious interrupt as soon as a shared 849 interrupt handler is registered, and just before one is deregistered. 850 Drivers ought to be able to handle interrupts coming in at those 851 points; some don't and need to be caught. 852 853menu "Debug Oops, Lockups and Hangs" 854 855config PANIC_ON_OOPS 856 bool "Panic on Oops" 857 help 858 Say Y here to enable the kernel to panic when it oopses. This 859 has the same effect as setting oops=panic on the kernel command 860 line. 861 862 This feature is useful to ensure that the kernel does not do 863 anything erroneous after an oops which could result in data 864 corruption or other issues. 865 866 Say N if unsure. 867 868config PANIC_ON_OOPS_VALUE 869 int 870 range 0 1 871 default 0 if !PANIC_ON_OOPS 872 default 1 if PANIC_ON_OOPS 873 874config PANIC_TIMEOUT 875 int "panic timeout" 876 default 0 877 help 878 Set the timeout value (in seconds) until a reboot occurs when the 879 the kernel panics. If n = 0, then we wait forever. A timeout 880 value n > 0 will wait n seconds before rebooting, while a timeout 881 value n < 0 will reboot immediately. 882 883config LOCKUP_DETECTOR 884 bool 885 886config SOFTLOCKUP_DETECTOR 887 bool "Detect Soft Lockups" 888 depends on DEBUG_KERNEL && !S390 889 select LOCKUP_DETECTOR 890 help 891 Say Y here to enable the kernel to act as a watchdog to detect 892 soft lockups. 893 894 Softlockups are bugs that cause the kernel to loop in kernel 895 mode for more than 20 seconds, without giving other tasks a 896 chance to run. The current stack trace is displayed upon 897 detection and the system will stay locked up. 898 899config BOOTPARAM_SOFTLOCKUP_PANIC 900 bool "Panic (Reboot) On Soft Lockups" 901 depends on SOFTLOCKUP_DETECTOR 902 help 903 Say Y here to enable the kernel to panic on "soft lockups", 904 which are bugs that cause the kernel to loop in kernel 905 mode for more than 20 seconds (configurable using the watchdog_thresh 906 sysctl), without giving other tasks a chance to run. 907 908 The panic can be used in combination with panic_timeout, 909 to cause the system to reboot automatically after a 910 lockup has been detected. This feature is useful for 911 high-availability systems that have uptime guarantees and 912 where a lockup must be resolved ASAP. 913 914 Say N if unsure. 915 916config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE 917 int 918 depends on SOFTLOCKUP_DETECTOR 919 range 0 1 920 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC 921 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC 922 923config HARDLOCKUP_DETECTOR_PERF 924 bool 925 select SOFTLOCKUP_DETECTOR 926 927# 928# Enables a timestamp based low pass filter to compensate for perf based 929# hard lockup detection which runs too fast due to turbo modes. 930# 931config HARDLOCKUP_CHECK_TIMESTAMP 932 bool 933 934# 935# arch/ can define HAVE_HARDLOCKUP_DETECTOR_ARCH to provide their own hard 936# lockup detector rather than the perf based detector. 937# 938config HARDLOCKUP_DETECTOR 939 bool "Detect Hard Lockups" 940 depends on DEBUG_KERNEL && !S390 941 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_ARCH 942 select LOCKUP_DETECTOR 943 select HARDLOCKUP_DETECTOR_PERF if HAVE_HARDLOCKUP_DETECTOR_PERF 944 select HARDLOCKUP_DETECTOR_ARCH if HAVE_HARDLOCKUP_DETECTOR_ARCH 945 help 946 Say Y here to enable the kernel to act as a watchdog to detect 947 hard lockups. 948 949 Hardlockups are bugs that cause the CPU to loop in kernel mode 950 for more than 10 seconds, without letting other interrupts have a 951 chance to run. The current stack trace is displayed upon detection 952 and the system will stay locked up. 953 954config BOOTPARAM_HARDLOCKUP_PANIC 955 bool "Panic (Reboot) On Hard Lockups" 956 depends on HARDLOCKUP_DETECTOR 957 help 958 Say Y here to enable the kernel to panic on "hard lockups", 959 which are bugs that cause the kernel to loop in kernel 960 mode with interrupts disabled for more than 10 seconds (configurable 961 using the watchdog_thresh sysctl). 962 963 Say N if unsure. 964 965config BOOTPARAM_HARDLOCKUP_PANIC_VALUE 966 int 967 depends on HARDLOCKUP_DETECTOR 968 range 0 1 969 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC 970 default 1 if BOOTPARAM_HARDLOCKUP_PANIC 971 972config DETECT_HUNG_TASK 973 bool "Detect Hung Tasks" 974 depends on DEBUG_KERNEL 975 default SOFTLOCKUP_DETECTOR 976 help 977 Say Y here to enable the kernel to detect "hung tasks", 978 which are bugs that cause the task to be stuck in 979 uninterruptible "D" state indefinitely. 980 981 When a hung task is detected, the kernel will print the 982 current stack trace (which you should report), but the 983 task will stay in uninterruptible state. If lockdep is 984 enabled then all held locks will also be reported. This 985 feature has negligible overhead. 986 987config DEFAULT_HUNG_TASK_TIMEOUT 988 int "Default timeout for hung task detection (in seconds)" 989 depends on DETECT_HUNG_TASK 990 default 120 991 help 992 This option controls the default timeout (in seconds) used 993 to determine when a task has become non-responsive and should 994 be considered hung. 995 996 It can be adjusted at runtime via the kernel.hung_task_timeout_secs 997 sysctl or by writing a value to 998 /proc/sys/kernel/hung_task_timeout_secs. 999 1000 A timeout of 0 disables the check. The default is two minutes. 1001 Keeping the default should be fine in most cases. 1002 1003config BOOTPARAM_HUNG_TASK_PANIC 1004 bool "Panic (Reboot) On Hung Tasks" 1005 depends on DETECT_HUNG_TASK 1006 help 1007 Say Y here to enable the kernel to panic on "hung tasks", 1008 which are bugs that cause the kernel to leave a task stuck 1009 in uninterruptible "D" state. 1010 1011 The panic can be used in combination with panic_timeout, 1012 to cause the system to reboot automatically after a 1013 hung task has been detected. This feature is useful for 1014 high-availability systems that have uptime guarantees and 1015 where a hung tasks must be resolved ASAP. 1016 1017 Say N if unsure. 1018 1019config BOOTPARAM_HUNG_TASK_PANIC_VALUE 1020 int 1021 depends on DETECT_HUNG_TASK 1022 range 0 1 1023 default 0 if !BOOTPARAM_HUNG_TASK_PANIC 1024 default 1 if BOOTPARAM_HUNG_TASK_PANIC 1025 1026config WQ_WATCHDOG 1027 bool "Detect Workqueue Stalls" 1028 depends on DEBUG_KERNEL 1029 help 1030 Say Y here to enable stall detection on workqueues. If a 1031 worker pool doesn't make forward progress on a pending work 1032 item for over a given amount of time, 30s by default, a 1033 warning message is printed along with dump of workqueue 1034 state. This can be configured through kernel parameter 1035 "workqueue.watchdog_thresh" and its sysfs counterpart. 1036 1037config TEST_LOCKUP 1038 tristate "Test module to generate lockups" 1039 help 1040 This builds the "test_lockup" module that helps to make sure 1041 that watchdogs and lockup detectors are working properly. 1042 1043 Depending on module parameters it could emulate soft or hard 1044 lockup, "hung task", or locking arbitrary lock for a long time. 1045 Also it could generate series of lockups with cooling-down periods. 1046 1047 If unsure, say N. 1048 1049endmenu # "Debug lockups and hangs" 1050 1051menu "Scheduler Debugging" 1052 1053config SCHED_DEBUG 1054 bool "Collect scheduler debugging info" 1055 depends on DEBUG_KERNEL && PROC_FS 1056 default y 1057 help 1058 If you say Y here, the /proc/sched_debug file will be provided 1059 that can help debug the scheduler. The runtime overhead of this 1060 option is minimal. 1061 1062config SCHED_INFO 1063 bool 1064 default n 1065 1066config SCHEDSTATS 1067 bool "Collect scheduler statistics" 1068 depends on DEBUG_KERNEL && PROC_FS 1069 select SCHED_INFO 1070 help 1071 If you say Y here, additional code will be inserted into the 1072 scheduler and related routines to collect statistics about 1073 scheduler behavior and provide them in /proc/schedstat. These 1074 stats may be useful for both tuning and debugging the scheduler 1075 If you aren't debugging the scheduler or trying to tune a specific 1076 application, you can say N to avoid the very slight overhead 1077 this adds. 1078 1079endmenu 1080 1081config DEBUG_TIMEKEEPING 1082 bool "Enable extra timekeeping sanity checking" 1083 help 1084 This option will enable additional timekeeping sanity checks 1085 which may be helpful when diagnosing issues where timekeeping 1086 problems are suspected. 1087 1088 This may include checks in the timekeeping hotpaths, so this 1089 option may have a (very small) performance impact to some 1090 workloads. 1091 1092 If unsure, say N. 1093 1094config DEBUG_PREEMPT 1095 bool "Debug preemptible kernel" 1096 depends on DEBUG_KERNEL && PREEMPTION && TRACE_IRQFLAGS_SUPPORT 1097 default y 1098 help 1099 If you say Y here then the kernel will use a debug variant of the 1100 commonly used smp_processor_id() function and will print warnings 1101 if kernel code uses it in a preemption-unsafe way. Also, the kernel 1102 will detect preemption count underflows. 1103 1104menu "Lock Debugging (spinlocks, mutexes, etc...)" 1105 1106config LOCK_DEBUGGING_SUPPORT 1107 bool 1108 depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1109 default y 1110 1111config PROVE_LOCKING 1112 bool "Lock debugging: prove locking correctness" 1113 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1114 select LOCKDEP 1115 select DEBUG_SPINLOCK 1116 select DEBUG_MUTEXES 1117 select DEBUG_RT_MUTEXES if RT_MUTEXES 1118 select DEBUG_RWSEMS 1119 select DEBUG_WW_MUTEX_SLOWPATH 1120 select DEBUG_LOCK_ALLOC 1121 select TRACE_IRQFLAGS 1122 default n 1123 help 1124 This feature enables the kernel to prove that all locking 1125 that occurs in the kernel runtime is mathematically 1126 correct: that under no circumstance could an arbitrary (and 1127 not yet triggered) combination of observed locking 1128 sequences (on an arbitrary number of CPUs, running an 1129 arbitrary number of tasks and interrupt contexts) cause a 1130 deadlock. 1131 1132 In short, this feature enables the kernel to report locking 1133 related deadlocks before they actually occur. 1134 1135 The proof does not depend on how hard and complex a 1136 deadlock scenario would be to trigger: how many 1137 participant CPUs, tasks and irq-contexts would be needed 1138 for it to trigger. The proof also does not depend on 1139 timing: if a race and a resulting deadlock is possible 1140 theoretically (no matter how unlikely the race scenario 1141 is), it will be proven so and will immediately be 1142 reported by the kernel (once the event is observed that 1143 makes the deadlock theoretically possible). 1144 1145 If a deadlock is impossible (i.e. the locking rules, as 1146 observed by the kernel, are mathematically correct), the 1147 kernel reports nothing. 1148 1149 NOTE: this feature can also be enabled for rwlocks, mutexes 1150 and rwsems - in which case all dependencies between these 1151 different locking variants are observed and mapped too, and 1152 the proof of observed correctness is also maintained for an 1153 arbitrary combination of these separate locking variants. 1154 1155 For more details, see Documentation/locking/lockdep-design.rst. 1156 1157config PROVE_RAW_LOCK_NESTING 1158 bool "Enable raw_spinlock - spinlock nesting checks" 1159 depends on PROVE_LOCKING 1160 default n 1161 help 1162 Enable the raw_spinlock vs. spinlock nesting checks which ensure 1163 that the lock nesting rules for PREEMPT_RT enabled kernels are 1164 not violated. 1165 1166 NOTE: There are known nesting problems. So if you enable this 1167 option expect lockdep splats until these problems have been fully 1168 addressed which is work in progress. This config switch allows to 1169 identify and analyze these problems. It will be removed and the 1170 check permanentely enabled once the main issues have been fixed. 1171 1172 If unsure, select N. 1173 1174config LOCK_STAT 1175 bool "Lock usage statistics" 1176 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1177 select LOCKDEP 1178 select DEBUG_SPINLOCK 1179 select DEBUG_MUTEXES 1180 select DEBUG_RT_MUTEXES if RT_MUTEXES 1181 select DEBUG_LOCK_ALLOC 1182 default n 1183 help 1184 This feature enables tracking lock contention points 1185 1186 For more details, see Documentation/locking/lockstat.rst 1187 1188 This also enables lock events required by "perf lock", 1189 subcommand of perf. 1190 If you want to use "perf lock", you also need to turn on 1191 CONFIG_EVENT_TRACING. 1192 1193 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events. 1194 (CONFIG_LOCKDEP defines "acquire" and "release" events.) 1195 1196config DEBUG_RT_MUTEXES 1197 bool "RT Mutex debugging, deadlock detection" 1198 depends on DEBUG_KERNEL && RT_MUTEXES 1199 help 1200 This allows rt mutex semantics violations and rt mutex related 1201 deadlocks (lockups) to be detected and reported automatically. 1202 1203config DEBUG_SPINLOCK 1204 bool "Spinlock and rw-lock debugging: basic checks" 1205 depends on DEBUG_KERNEL 1206 select UNINLINE_SPIN_UNLOCK 1207 help 1208 Say Y here and build SMP to catch missing spinlock initialization 1209 and certain other kinds of spinlock errors commonly made. This is 1210 best used in conjunction with the NMI watchdog so that spinlock 1211 deadlocks are also debuggable. 1212 1213config DEBUG_MUTEXES 1214 bool "Mutex debugging: basic checks" 1215 depends on DEBUG_KERNEL 1216 help 1217 This feature allows mutex semantics violations to be detected and 1218 reported. 1219 1220config DEBUG_WW_MUTEX_SLOWPATH 1221 bool "Wait/wound mutex debugging: Slowpath testing" 1222 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1223 select DEBUG_LOCK_ALLOC 1224 select DEBUG_SPINLOCK 1225 select DEBUG_MUTEXES 1226 help 1227 This feature enables slowpath testing for w/w mutex users by 1228 injecting additional -EDEADLK wound/backoff cases. Together with 1229 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this 1230 will test all possible w/w mutex interface abuse with the 1231 exception of simply not acquiring all the required locks. 1232 Note that this feature can introduce significant overhead, so 1233 it really should not be enabled in a production or distro kernel, 1234 even a debug kernel. If you are a driver writer, enable it. If 1235 you are a distro, do not. 1236 1237config DEBUG_RWSEMS 1238 bool "RW Semaphore debugging: basic checks" 1239 depends on DEBUG_KERNEL 1240 help 1241 This debugging feature allows mismatched rw semaphore locks 1242 and unlocks to be detected and reported. 1243 1244config DEBUG_LOCK_ALLOC 1245 bool "Lock debugging: detect incorrect freeing of live locks" 1246 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1247 select DEBUG_SPINLOCK 1248 select DEBUG_MUTEXES 1249 select DEBUG_RT_MUTEXES if RT_MUTEXES 1250 select LOCKDEP 1251 help 1252 This feature will check whether any held lock (spinlock, rwlock, 1253 mutex or rwsem) is incorrectly freed by the kernel, via any of the 1254 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(), 1255 vfree(), etc.), whether a live lock is incorrectly reinitialized via 1256 spin_lock_init()/mutex_init()/etc., or whether there is any lock 1257 held during task exit. 1258 1259config LOCKDEP 1260 bool 1261 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1262 select STACKTRACE 1263 select FRAME_POINTER if !MIPS && !PPC && !ARM && !S390 && !MICROBLAZE && !ARC && !X86 1264 select KALLSYMS 1265 select KALLSYMS_ALL 1266 1267config LOCKDEP_SMALL 1268 bool 1269 1270config DEBUG_LOCKDEP 1271 bool "Lock dependency engine debugging" 1272 depends on DEBUG_KERNEL && LOCKDEP 1273 help 1274 If you say Y here, the lock dependency engine will do 1275 additional runtime checks to debug itself, at the price 1276 of more runtime overhead. 1277 1278config DEBUG_ATOMIC_SLEEP 1279 bool "Sleep inside atomic section checking" 1280 select PREEMPT_COUNT 1281 depends on DEBUG_KERNEL 1282 depends on !ARCH_NO_PREEMPT 1283 help 1284 If you say Y here, various routines which may sleep will become very 1285 noisy if they are called inside atomic sections: when a spinlock is 1286 held, inside an rcu read side critical section, inside preempt disabled 1287 sections, inside an interrupt, etc... 1288 1289config DEBUG_LOCKING_API_SELFTESTS 1290 bool "Locking API boot-time self-tests" 1291 depends on DEBUG_KERNEL 1292 help 1293 Say Y here if you want the kernel to run a short self-test during 1294 bootup. The self-test checks whether common types of locking bugs 1295 are detected by debugging mechanisms or not. (if you disable 1296 lock debugging then those bugs wont be detected of course.) 1297 The following locking APIs are covered: spinlocks, rwlocks, 1298 mutexes and rwsems. 1299 1300config LOCK_TORTURE_TEST 1301 tristate "torture tests for locking" 1302 depends on DEBUG_KERNEL 1303 select TORTURE_TEST 1304 help 1305 This option provides a kernel module that runs torture tests 1306 on kernel locking primitives. The kernel module may be built 1307 after the fact on the running kernel to be tested, if desired. 1308 1309 Say Y here if you want kernel locking-primitive torture tests 1310 to be built into the kernel. 1311 Say M if you want these torture tests to build as a module. 1312 Say N if you are unsure. 1313 1314config WW_MUTEX_SELFTEST 1315 tristate "Wait/wound mutex selftests" 1316 help 1317 This option provides a kernel module that runs tests on the 1318 on the struct ww_mutex locking API. 1319 1320 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction 1321 with this test harness. 1322 1323 Say M if you want these self tests to build as a module. 1324 Say N if you are unsure. 1325 1326endmenu # lock debugging 1327 1328config TRACE_IRQFLAGS 1329 bool 1330 help 1331 Enables hooks to interrupt enabling and disabling for 1332 either tracing or lock debugging. 1333 1334config STACKTRACE 1335 bool "Stack backtrace support" 1336 depends on STACKTRACE_SUPPORT 1337 help 1338 This option causes the kernel to create a /proc/pid/stack for 1339 every process, showing its current stack trace. 1340 It is also used by various kernel debugging features that require 1341 stack trace generation. 1342 1343config WARN_ALL_UNSEEDED_RANDOM 1344 bool "Warn for all uses of unseeded randomness" 1345 default n 1346 help 1347 Some parts of the kernel contain bugs relating to their use of 1348 cryptographically secure random numbers before it's actually possible 1349 to generate those numbers securely. This setting ensures that these 1350 flaws don't go unnoticed, by enabling a message, should this ever 1351 occur. This will allow people with obscure setups to know when things 1352 are going wrong, so that they might contact developers about fixing 1353 it. 1354 1355 Unfortunately, on some models of some architectures getting 1356 a fully seeded CRNG is extremely difficult, and so this can 1357 result in dmesg getting spammed for a surprisingly long 1358 time. This is really bad from a security perspective, and 1359 so architecture maintainers really need to do what they can 1360 to get the CRNG seeded sooner after the system is booted. 1361 However, since users cannot do anything actionable to 1362 address this, by default the kernel will issue only a single 1363 warning for the first use of unseeded randomness. 1364 1365 Say Y here if you want to receive warnings for all uses of 1366 unseeded randomness. This will be of use primarily for 1367 those developers interested in improving the security of 1368 Linux kernels running on their architecture (or 1369 subarchitecture). 1370 1371config DEBUG_KOBJECT 1372 bool "kobject debugging" 1373 depends on DEBUG_KERNEL 1374 help 1375 If you say Y here, some extra kobject debugging messages will be sent 1376 to the syslog. 1377 1378config DEBUG_KOBJECT_RELEASE 1379 bool "kobject release debugging" 1380 depends on DEBUG_OBJECTS_TIMERS 1381 help 1382 kobjects are reference counted objects. This means that their 1383 last reference count put is not predictable, and the kobject can 1384 live on past the point at which a driver decides to drop it's 1385 initial reference to the kobject gained on allocation. An 1386 example of this would be a struct device which has just been 1387 unregistered. 1388 1389 However, some buggy drivers assume that after such an operation, 1390 the memory backing the kobject can be immediately freed. This 1391 goes completely against the principles of a refcounted object. 1392 1393 If you say Y here, the kernel will delay the release of kobjects 1394 on the last reference count to improve the visibility of this 1395 kind of kobject release bug. 1396 1397config HAVE_DEBUG_BUGVERBOSE 1398 bool 1399 1400menu "Debug kernel data structures" 1401 1402config DEBUG_LIST 1403 bool "Debug linked list manipulation" 1404 depends on DEBUG_KERNEL || BUG_ON_DATA_CORRUPTION 1405 help 1406 Enable this to turn on extended checks in the linked-list 1407 walking routines. 1408 1409 If unsure, say N. 1410 1411config DEBUG_PLIST 1412 bool "Debug priority linked list manipulation" 1413 depends on DEBUG_KERNEL 1414 help 1415 Enable this to turn on extended checks in the priority-ordered 1416 linked-list (plist) walking routines. This checks the entire 1417 list multiple times during each manipulation. 1418 1419 If unsure, say N. 1420 1421config DEBUG_SG 1422 bool "Debug SG table operations" 1423 depends on DEBUG_KERNEL 1424 help 1425 Enable this to turn on checks on scatter-gather tables. This can 1426 help find problems with drivers that do not properly initialize 1427 their sg tables. 1428 1429 If unsure, say N. 1430 1431config DEBUG_NOTIFIERS 1432 bool "Debug notifier call chains" 1433 depends on DEBUG_KERNEL 1434 help 1435 Enable this to turn on sanity checking for notifier call chains. 1436 This is most useful for kernel developers to make sure that 1437 modules properly unregister themselves from notifier chains. 1438 This is a relatively cheap check but if you care about maximum 1439 performance, say N. 1440 1441config BUG_ON_DATA_CORRUPTION 1442 bool "Trigger a BUG when data corruption is detected" 1443 select DEBUG_LIST 1444 help 1445 Select this option if the kernel should BUG when it encounters 1446 data corruption in kernel memory structures when they get checked 1447 for validity. 1448 1449 If unsure, say N. 1450 1451endmenu 1452 1453config DEBUG_CREDENTIALS 1454 bool "Debug credential management" 1455 depends on DEBUG_KERNEL 1456 help 1457 Enable this to turn on some debug checking for credential 1458 management. The additional code keeps track of the number of 1459 pointers from task_structs to any given cred struct, and checks to 1460 see that this number never exceeds the usage count of the cred 1461 struct. 1462 1463 Furthermore, if SELinux is enabled, this also checks that the 1464 security pointer in the cred struct is never seen to be invalid. 1465 1466 If unsure, say N. 1467 1468source "kernel/rcu/Kconfig.debug" 1469 1470config DEBUG_WQ_FORCE_RR_CPU 1471 bool "Force round-robin CPU selection for unbound work items" 1472 depends on DEBUG_KERNEL 1473 default n 1474 help 1475 Workqueue used to implicitly guarantee that work items queued 1476 without explicit CPU specified are put on the local CPU. This 1477 guarantee is no longer true and while local CPU is still 1478 preferred work items may be put on foreign CPUs. Kernel 1479 parameter "workqueue.debug_force_rr_cpu" is added to force 1480 round-robin CPU selection to flush out usages which depend on the 1481 now broken guarantee. This config option enables the debug 1482 feature by default. When enabled, memory and cache locality will 1483 be impacted. 1484 1485config DEBUG_BLOCK_EXT_DEVT 1486 bool "Force extended block device numbers and spread them" 1487 depends on DEBUG_KERNEL 1488 depends on BLOCK 1489 default n 1490 help 1491 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON 1492 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT 1493 YOU ARE DOING. Distros, please enable this and fix whatever 1494 is broken. 1495 1496 Conventionally, block device numbers are allocated from 1497 predetermined contiguous area. However, extended block area 1498 may introduce non-contiguous block device numbers. This 1499 option forces most block device numbers to be allocated from 1500 the extended space and spreads them to discover kernel or 1501 userland code paths which assume predetermined contiguous 1502 device number allocation. 1503 1504 Note that turning on this debug option shuffles all the 1505 device numbers for all IDE and SCSI devices including libata 1506 ones, so root partition specified using device number 1507 directly (via rdev or root=MAJ:MIN) won't work anymore. 1508 Textual device names (root=/dev/sdXn) will continue to work. 1509 1510 Say N if you are unsure. 1511 1512config CPU_HOTPLUG_STATE_CONTROL 1513 bool "Enable CPU hotplug state control" 1514 depends on DEBUG_KERNEL 1515 depends on HOTPLUG_CPU 1516 default n 1517 help 1518 Allows to write steps between "offline" and "online" to the CPUs 1519 sysfs target file so states can be stepped granular. This is a debug 1520 option for now as the hotplug machinery cannot be stopped and 1521 restarted at arbitrary points yet. 1522 1523 Say N if your are unsure. 1524 1525config LATENCYTOP 1526 bool "Latency measuring infrastructure" 1527 depends on DEBUG_KERNEL 1528 depends on STACKTRACE_SUPPORT 1529 depends on PROC_FS 1530 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM && !ARC && !X86 1531 select KALLSYMS 1532 select KALLSYMS_ALL 1533 select STACKTRACE 1534 select SCHEDSTATS 1535 select SCHED_DEBUG 1536 help 1537 Enable this option if you want to use the LatencyTOP tool 1538 to find out which userspace is blocking on what kernel operations. 1539 1540source "kernel/trace/Kconfig" 1541 1542config PROVIDE_OHCI1394_DMA_INIT 1543 bool "Remote debugging over FireWire early on boot" 1544 depends on PCI && X86 1545 help 1546 If you want to debug problems which hang or crash the kernel early 1547 on boot and the crashing machine has a FireWire port, you can use 1548 this feature to remotely access the memory of the crashed machine 1549 over FireWire. This employs remote DMA as part of the OHCI1394 1550 specification which is now the standard for FireWire controllers. 1551 1552 With remote DMA, you can monitor the printk buffer remotely using 1553 firescope and access all memory below 4GB using fireproxy from gdb. 1554 Even controlling a kernel debugger is possible using remote DMA. 1555 1556 Usage: 1557 1558 If ohci1394_dma=early is used as boot parameter, it will initialize 1559 all OHCI1394 controllers which are found in the PCI config space. 1560 1561 As all changes to the FireWire bus such as enabling and disabling 1562 devices cause a bus reset and thereby disable remote DMA for all 1563 devices, be sure to have the cable plugged and FireWire enabled on 1564 the debugging host before booting the debug target for debugging. 1565 1566 This code (~1k) is freed after boot. By then, the firewire stack 1567 in charge of the OHCI-1394 controllers should be used instead. 1568 1569 See Documentation/core-api/debugging-via-ohci1394.rst for more information. 1570 1571source "samples/Kconfig" 1572 1573config ARCH_HAS_DEVMEM_IS_ALLOWED 1574 bool 1575 1576config STRICT_DEVMEM 1577 bool "Filter access to /dev/mem" 1578 depends on MMU && DEVMEM 1579 depends on ARCH_HAS_DEVMEM_IS_ALLOWED 1580 default y if PPC || X86 || ARM64 1581 help 1582 If this option is disabled, you allow userspace (root) access to all 1583 of memory, including kernel and userspace memory. Accidental 1584 access to this is obviously disastrous, but specific access can 1585 be used by people debugging the kernel. Note that with PAT support 1586 enabled, even in this case there are restrictions on /dev/mem 1587 use due to the cache aliasing requirements. 1588 1589 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem 1590 file only allows userspace access to PCI space and the BIOS code and 1591 data regions. This is sufficient for dosemu and X and all common 1592 users of /dev/mem. 1593 1594 If in doubt, say Y. 1595 1596config IO_STRICT_DEVMEM 1597 bool "Filter I/O access to /dev/mem" 1598 depends on STRICT_DEVMEM 1599 help 1600 If this option is disabled, you allow userspace (root) access to all 1601 io-memory regardless of whether a driver is actively using that 1602 range. Accidental access to this is obviously disastrous, but 1603 specific access can be used by people debugging kernel drivers. 1604 1605 If this option is switched on, the /dev/mem file only allows 1606 userspace access to *idle* io-memory ranges (see /proc/iomem) This 1607 may break traditional users of /dev/mem (dosemu, legacy X, etc...) 1608 if the driver using a given range cannot be disabled. 1609 1610 If in doubt, say Y. 1611 1612menu "$(SRCARCH) Debugging" 1613 1614source "arch/$(SRCARCH)/Kconfig.debug" 1615 1616endmenu 1617 1618menu "Kernel Testing and Coverage" 1619 1620source "lib/kunit/Kconfig" 1621 1622config NOTIFIER_ERROR_INJECTION 1623 tristate "Notifier error injection" 1624 depends on DEBUG_KERNEL 1625 select DEBUG_FS 1626 help 1627 This option provides the ability to inject artificial errors to 1628 specified notifier chain callbacks. It is useful to test the error 1629 handling of notifier call chain failures. 1630 1631 Say N if unsure. 1632 1633config PM_NOTIFIER_ERROR_INJECT 1634 tristate "PM notifier error injection module" 1635 depends on PM && NOTIFIER_ERROR_INJECTION 1636 default m if PM_DEBUG 1637 help 1638 This option provides the ability to inject artificial errors to 1639 PM notifier chain callbacks. It is controlled through debugfs 1640 interface /sys/kernel/debug/notifier-error-inject/pm 1641 1642 If the notifier call chain should be failed with some events 1643 notified, write the error code to "actions/<notifier event>/error". 1644 1645 Example: Inject PM suspend error (-12 = -ENOMEM) 1646 1647 # cd /sys/kernel/debug/notifier-error-inject/pm/ 1648 # echo -12 > actions/PM_SUSPEND_PREPARE/error 1649 # echo mem > /sys/power/state 1650 bash: echo: write error: Cannot allocate memory 1651 1652 To compile this code as a module, choose M here: the module will 1653 be called pm-notifier-error-inject. 1654 1655 If unsure, say N. 1656 1657config OF_RECONFIG_NOTIFIER_ERROR_INJECT 1658 tristate "OF reconfig notifier error injection module" 1659 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION 1660 help 1661 This option provides the ability to inject artificial errors to 1662 OF reconfig notifier chain callbacks. It is controlled 1663 through debugfs interface under 1664 /sys/kernel/debug/notifier-error-inject/OF-reconfig/ 1665 1666 If the notifier call chain should be failed with some events 1667 notified, write the error code to "actions/<notifier event>/error". 1668 1669 To compile this code as a module, choose M here: the module will 1670 be called of-reconfig-notifier-error-inject. 1671 1672 If unsure, say N. 1673 1674config NETDEV_NOTIFIER_ERROR_INJECT 1675 tristate "Netdev notifier error injection module" 1676 depends on NET && NOTIFIER_ERROR_INJECTION 1677 help 1678 This option provides the ability to inject artificial errors to 1679 netdevice notifier chain callbacks. It is controlled through debugfs 1680 interface /sys/kernel/debug/notifier-error-inject/netdev 1681 1682 If the notifier call chain should be failed with some events 1683 notified, write the error code to "actions/<notifier event>/error". 1684 1685 Example: Inject netdevice mtu change error (-22 = -EINVAL) 1686 1687 # cd /sys/kernel/debug/notifier-error-inject/netdev 1688 # echo -22 > actions/NETDEV_CHANGEMTU/error 1689 # ip link set eth0 mtu 1024 1690 RTNETLINK answers: Invalid argument 1691 1692 To compile this code as a module, choose M here: the module will 1693 be called netdev-notifier-error-inject. 1694 1695 If unsure, say N. 1696 1697config FUNCTION_ERROR_INJECTION 1698 def_bool y 1699 depends on HAVE_FUNCTION_ERROR_INJECTION && KPROBES 1700 1701config FAULT_INJECTION 1702 bool "Fault-injection framework" 1703 depends on DEBUG_KERNEL 1704 help 1705 Provide fault-injection framework. 1706 For more details, see Documentation/fault-injection/. 1707 1708config FAILSLAB 1709 bool "Fault-injection capability for kmalloc" 1710 depends on FAULT_INJECTION 1711 depends on SLAB || SLUB 1712 help 1713 Provide fault-injection capability for kmalloc. 1714 1715config FAIL_PAGE_ALLOC 1716 bool "Fault-injection capability for alloc_pages()" 1717 depends on FAULT_INJECTION 1718 help 1719 Provide fault-injection capability for alloc_pages(). 1720 1721config FAIL_MAKE_REQUEST 1722 bool "Fault-injection capability for disk IO" 1723 depends on FAULT_INJECTION && BLOCK 1724 help 1725 Provide fault-injection capability for disk IO. 1726 1727config FAIL_IO_TIMEOUT 1728 bool "Fault-injection capability for faking disk interrupts" 1729 depends on FAULT_INJECTION && BLOCK 1730 help 1731 Provide fault-injection capability on end IO handling. This 1732 will make the block layer "forget" an interrupt as configured, 1733 thus exercising the error handling. 1734 1735 Only works with drivers that use the generic timeout handling, 1736 for others it wont do anything. 1737 1738config FAIL_FUTEX 1739 bool "Fault-injection capability for futexes" 1740 select DEBUG_FS 1741 depends on FAULT_INJECTION && FUTEX 1742 help 1743 Provide fault-injection capability for futexes. 1744 1745config FAULT_INJECTION_DEBUG_FS 1746 bool "Debugfs entries for fault-injection capabilities" 1747 depends on FAULT_INJECTION && SYSFS && DEBUG_FS 1748 help 1749 Enable configuration of fault-injection capabilities via debugfs. 1750 1751config FAIL_FUNCTION 1752 bool "Fault-injection capability for functions" 1753 depends on FAULT_INJECTION_DEBUG_FS && FUNCTION_ERROR_INJECTION 1754 help 1755 Provide function-based fault-injection capability. 1756 This will allow you to override a specific function with a return 1757 with given return value. As a result, function caller will see 1758 an error value and have to handle it. This is useful to test the 1759 error handling in various subsystems. 1760 1761config FAIL_MMC_REQUEST 1762 bool "Fault-injection capability for MMC IO" 1763 depends on FAULT_INJECTION_DEBUG_FS && MMC 1764 help 1765 Provide fault-injection capability for MMC IO. 1766 This will make the mmc core return data errors. This is 1767 useful to test the error handling in the mmc block device 1768 and to test how the mmc host driver handles retries from 1769 the block device. 1770 1771config FAULT_INJECTION_STACKTRACE_FILTER 1772 bool "stacktrace filter for fault-injection capabilities" 1773 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT 1774 depends on !X86_64 1775 select STACKTRACE 1776 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM && !ARC && !X86 1777 help 1778 Provide stacktrace filter for fault-injection capabilities 1779 1780config ARCH_HAS_KCOV 1781 bool 1782 help 1783 An architecture should select this when it can successfully 1784 build and run with CONFIG_KCOV. This typically requires 1785 disabling instrumentation for some early boot code. 1786 1787config CC_HAS_SANCOV_TRACE_PC 1788 def_bool $(cc-option,-fsanitize-coverage=trace-pc) 1789 1790 1791config KCOV 1792 bool "Code coverage for fuzzing" 1793 depends on ARCH_HAS_KCOV 1794 depends on CC_HAS_SANCOV_TRACE_PC || GCC_PLUGINS 1795 select DEBUG_FS 1796 select GCC_PLUGIN_SANCOV if !CC_HAS_SANCOV_TRACE_PC 1797 help 1798 KCOV exposes kernel code coverage information in a form suitable 1799 for coverage-guided fuzzing (randomized testing). 1800 1801 If RANDOMIZE_BASE is enabled, PC values will not be stable across 1802 different machines and across reboots. If you need stable PC values, 1803 disable RANDOMIZE_BASE. 1804 1805 For more details, see Documentation/dev-tools/kcov.rst. 1806 1807config KCOV_ENABLE_COMPARISONS 1808 bool "Enable comparison operands collection by KCOV" 1809 depends on KCOV 1810 depends on $(cc-option,-fsanitize-coverage=trace-cmp) 1811 help 1812 KCOV also exposes operands of every comparison in the instrumented 1813 code along with operand sizes and PCs of the comparison instructions. 1814 These operands can be used by fuzzing engines to improve the quality 1815 of fuzzing coverage. 1816 1817config KCOV_INSTRUMENT_ALL 1818 bool "Instrument all code by default" 1819 depends on KCOV 1820 default y 1821 help 1822 If you are doing generic system call fuzzing (like e.g. syzkaller), 1823 then you will want to instrument the whole kernel and you should 1824 say y here. If you are doing more targeted fuzzing (like e.g. 1825 filesystem fuzzing with AFL) then you will want to enable coverage 1826 for more specific subsets of files, and should say n here. 1827 1828config KCOV_IRQ_AREA_SIZE 1829 hex "Size of interrupt coverage collection area in words" 1830 depends on KCOV 1831 default 0x40000 1832 help 1833 KCOV uses preallocated per-cpu areas to collect coverage from 1834 soft interrupts. This specifies the size of those areas in the 1835 number of unsigned long words. 1836 1837menuconfig RUNTIME_TESTING_MENU 1838 bool "Runtime Testing" 1839 def_bool y 1840 1841if RUNTIME_TESTING_MENU 1842 1843config LKDTM 1844 tristate "Linux Kernel Dump Test Tool Module" 1845 depends on DEBUG_FS 1846 help 1847 This module enables testing of the different dumping mechanisms by 1848 inducing system failures at predefined crash points. 1849 If you don't need it: say N 1850 Choose M here to compile this code as a module. The module will be 1851 called lkdtm. 1852 1853 Documentation on how to use the module can be found in 1854 Documentation/fault-injection/provoke-crashes.rst 1855 1856config TEST_LIST_SORT 1857 tristate "Linked list sorting test" 1858 depends on DEBUG_KERNEL || m 1859 help 1860 Enable this to turn on 'list_sort()' function test. This test is 1861 executed only once during system boot (so affects only boot time), 1862 or at module load time. 1863 1864 If unsure, say N. 1865 1866config TEST_MIN_HEAP 1867 tristate "Min heap test" 1868 depends on DEBUG_KERNEL || m 1869 help 1870 Enable this to turn on min heap function tests. This test is 1871 executed only once during system boot (so affects only boot time), 1872 or at module load time. 1873 1874 If unsure, say N. 1875 1876config TEST_SORT 1877 tristate "Array-based sort test" 1878 depends on DEBUG_KERNEL || m 1879 help 1880 This option enables the self-test function of 'sort()' at boot, 1881 or at module load time. 1882 1883 If unsure, say N. 1884 1885config KPROBES_SANITY_TEST 1886 bool "Kprobes sanity tests" 1887 depends on DEBUG_KERNEL 1888 depends on KPROBES 1889 help 1890 This option provides for testing basic kprobes functionality on 1891 boot. Samples of kprobe and kretprobe are inserted and 1892 verified for functionality. 1893 1894 Say N if you are unsure. 1895 1896config BACKTRACE_SELF_TEST 1897 tristate "Self test for the backtrace code" 1898 depends on DEBUG_KERNEL 1899 help 1900 This option provides a kernel module that can be used to test 1901 the kernel stack backtrace code. This option is not useful 1902 for distributions or general kernels, but only for kernel 1903 developers working on architecture code. 1904 1905 Note that if you want to also test saved backtraces, you will 1906 have to enable STACKTRACE as well. 1907 1908 Say N if you are unsure. 1909 1910config RBTREE_TEST 1911 tristate "Red-Black tree test" 1912 depends on DEBUG_KERNEL 1913 help 1914 A benchmark measuring the performance of the rbtree library. 1915 Also includes rbtree invariant checks. 1916 1917config REED_SOLOMON_TEST 1918 tristate "Reed-Solomon library test" 1919 depends on DEBUG_KERNEL || m 1920 select REED_SOLOMON 1921 select REED_SOLOMON_ENC16 1922 select REED_SOLOMON_DEC16 1923 help 1924 This option enables the self-test function of rslib at boot, 1925 or at module load time. 1926 1927 If unsure, say N. 1928 1929config INTERVAL_TREE_TEST 1930 tristate "Interval tree test" 1931 depends on DEBUG_KERNEL 1932 select INTERVAL_TREE 1933 help 1934 A benchmark measuring the performance of the interval tree library 1935 1936config PERCPU_TEST 1937 tristate "Per cpu operations test" 1938 depends on m && DEBUG_KERNEL 1939 help 1940 Enable this option to build test module which validates per-cpu 1941 operations. 1942 1943 If unsure, say N. 1944 1945config ATOMIC64_SELFTEST 1946 tristate "Perform an atomic64_t self-test" 1947 help 1948 Enable this option to test the atomic64_t functions at boot or 1949 at module load time. 1950 1951 If unsure, say N. 1952 1953config ASYNC_RAID6_TEST 1954 tristate "Self test for hardware accelerated raid6 recovery" 1955 depends on ASYNC_RAID6_RECOV 1956 select ASYNC_MEMCPY 1957 help 1958 This is a one-shot self test that permutes through the 1959 recovery of all the possible two disk failure scenarios for a 1960 N-disk array. Recovery is performed with the asynchronous 1961 raid6 recovery routines, and will optionally use an offload 1962 engine if one is available. 1963 1964 If unsure, say N. 1965 1966config TEST_HEXDUMP 1967 tristate "Test functions located in the hexdump module at runtime" 1968 1969config TEST_STRING_HELPERS 1970 tristate "Test functions located in the string_helpers module at runtime" 1971 1972config TEST_STRSCPY 1973 tristate "Test strscpy*() family of functions at runtime" 1974 1975config TEST_KSTRTOX 1976 tristate "Test kstrto*() family of functions at runtime" 1977 1978config TEST_PRINTF 1979 tristate "Test printf() family of functions at runtime" 1980 1981config TEST_BITMAP 1982 tristate "Test bitmap_*() family of functions at runtime" 1983 help 1984 Enable this option to test the bitmap functions at boot. 1985 1986 If unsure, say N. 1987 1988config TEST_BITFIELD 1989 tristate "Test bitfield functions at runtime" 1990 help 1991 Enable this option to test the bitfield functions at boot. 1992 1993 If unsure, say N. 1994 1995config TEST_UUID 1996 tristate "Test functions located in the uuid module at runtime" 1997 1998config TEST_XARRAY 1999 tristate "Test the XArray code at runtime" 2000 2001config TEST_OVERFLOW 2002 tristate "Test check_*_overflow() functions at runtime" 2003 2004config TEST_RHASHTABLE 2005 tristate "Perform selftest on resizable hash table" 2006 help 2007 Enable this option to test the rhashtable functions at boot. 2008 2009 If unsure, say N. 2010 2011config TEST_HASH 2012 tristate "Perform selftest on hash functions" 2013 help 2014 Enable this option to test the kernel's integer (<linux/hash.h>), 2015 string (<linux/stringhash.h>), and siphash (<linux/siphash.h>) 2016 hash functions on boot (or module load). 2017 2018 This is intended to help people writing architecture-specific 2019 optimized versions. If unsure, say N. 2020 2021config TEST_IDA 2022 tristate "Perform selftest on IDA functions" 2023 2024config TEST_PARMAN 2025 tristate "Perform selftest on priority array manager" 2026 depends on PARMAN 2027 help 2028 Enable this option to test priority array manager on boot 2029 (or module load). 2030 2031 If unsure, say N. 2032 2033config TEST_IRQ_TIMINGS 2034 bool "IRQ timings selftest" 2035 depends on IRQ_TIMINGS 2036 help 2037 Enable this option to test the irq timings code on boot. 2038 2039 If unsure, say N. 2040 2041config TEST_LKM 2042 tristate "Test module loading with 'hello world' module" 2043 depends on m 2044 help 2045 This builds the "test_module" module that emits "Hello, world" 2046 on printk when loaded. It is designed to be used for basic 2047 evaluation of the module loading subsystem (for example when 2048 validating module verification). It lacks any extra dependencies, 2049 and will not normally be loaded by the system unless explicitly 2050 requested by name. 2051 2052 If unsure, say N. 2053 2054config TEST_BITOPS 2055 tristate "Test module for compilation of clear_bit/set_bit operations" 2056 depends on m 2057 help 2058 This builds the "test_bitops" module that is much like the 2059 TEST_LKM module except that it does a basic exercise of the 2060 clear_bit and set_bit macros to make sure there are no compiler 2061 warnings from C=1 sparse checker or -Wextra compilations. It has 2062 no dependencies and doesn't run or load unless explicitly requested 2063 by name. for example: modprobe test_bitops. 2064 2065 If unsure, say N. 2066 2067config TEST_VMALLOC 2068 tristate "Test module for stress/performance analysis of vmalloc allocator" 2069 default n 2070 depends on MMU 2071 depends on m 2072 help 2073 This builds the "test_vmalloc" module that should be used for 2074 stress and performance analysis. So, any new change for vmalloc 2075 subsystem can be evaluated from performance and stability point 2076 of view. 2077 2078 If unsure, say N. 2079 2080config TEST_USER_COPY 2081 tristate "Test user/kernel boundary protections" 2082 depends on m 2083 help 2084 This builds the "test_user_copy" module that runs sanity checks 2085 on the copy_to/from_user infrastructure, making sure basic 2086 user/kernel boundary testing is working. If it fails to load, 2087 a regression has been detected in the user/kernel memory boundary 2088 protections. 2089 2090 If unsure, say N. 2091 2092config TEST_BPF 2093 tristate "Test BPF filter functionality" 2094 depends on m && NET 2095 help 2096 This builds the "test_bpf" module that runs various test vectors 2097 against the BPF interpreter or BPF JIT compiler depending on the 2098 current setting. This is in particular useful for BPF JIT compiler 2099 development, but also to run regression tests against changes in 2100 the interpreter code. It also enables test stubs for eBPF maps and 2101 verifier used by user space verifier testsuite. 2102 2103 If unsure, say N. 2104 2105config TEST_BLACKHOLE_DEV 2106 tristate "Test blackhole netdev functionality" 2107 depends on m && NET 2108 help 2109 This builds the "test_blackhole_dev" module that validates the 2110 data path through this blackhole netdev. 2111 2112 If unsure, say N. 2113 2114config FIND_BIT_BENCHMARK 2115 tristate "Test find_bit functions" 2116 help 2117 This builds the "test_find_bit" module that measure find_*_bit() 2118 functions performance. 2119 2120 If unsure, say N. 2121 2122config TEST_FIRMWARE 2123 tristate "Test firmware loading via userspace interface" 2124 depends on FW_LOADER 2125 help 2126 This builds the "test_firmware" module that creates a userspace 2127 interface for testing firmware loading. This can be used to 2128 control the triggering of firmware loading without needing an 2129 actual firmware-using device. The contents can be rechecked by 2130 userspace. 2131 2132 If unsure, say N. 2133 2134config TEST_SYSCTL 2135 tristate "sysctl test driver" 2136 depends on PROC_SYSCTL 2137 help 2138 This builds the "test_sysctl" module. This driver enables to test the 2139 proc sysctl interfaces available to drivers safely without affecting 2140 production knobs which might alter system functionality. 2141 2142 If unsure, say N. 2143 2144config SYSCTL_KUNIT_TEST 2145 tristate "KUnit test for sysctl" if !KUNIT_ALL_TESTS 2146 depends on KUNIT 2147 default KUNIT_ALL_TESTS 2148 help 2149 This builds the proc sysctl unit test, which runs on boot. 2150 Tests the API contract and implementation correctness of sysctl. 2151 For more information on KUnit and unit tests in general please refer 2152 to the KUnit documentation in Documentation/dev-tools/kunit/. 2153 2154 If unsure, say N. 2155 2156config LIST_KUNIT_TEST 2157 tristate "KUnit Test for Kernel Linked-list structures" if !KUNIT_ALL_TESTS 2158 depends on KUNIT 2159 default KUNIT_ALL_TESTS 2160 help 2161 This builds the linked list KUnit test suite. 2162 It tests that the API and basic functionality of the list_head type 2163 and associated macros. 2164 2165 KUnit tests run during boot and output the results to the debug log 2166 in TAP format (http://testanything.org/). Only useful for kernel devs 2167 running the KUnit test harness, and not intended for inclusion into a 2168 production build. 2169 2170 For more information on KUnit and unit tests in general please refer 2171 to the KUnit documentation in Documentation/dev-tools/kunit/. 2172 2173 If unsure, say N. 2174 2175config LINEAR_RANGES_TEST 2176 tristate "KUnit test for linear_ranges" 2177 depends on KUNIT 2178 select LINEAR_RANGES 2179 help 2180 This builds the linear_ranges unit test, which runs on boot. 2181 Tests the linear_ranges logic correctness. 2182 For more information on KUnit and unit tests in general please refer 2183 to the KUnit documentation in Documentation/dev-tools/kunit/. 2184 2185 If unsure, say N. 2186 2187config TEST_UDELAY 2188 tristate "udelay test driver" 2189 help 2190 This builds the "udelay_test" module that helps to make sure 2191 that udelay() is working properly. 2192 2193 If unsure, say N. 2194 2195config TEST_STATIC_KEYS 2196 tristate "Test static keys" 2197 depends on m 2198 help 2199 Test the static key interfaces. 2200 2201 If unsure, say N. 2202 2203config TEST_KMOD 2204 tristate "kmod stress tester" 2205 depends on m 2206 depends on NETDEVICES && NET_CORE && INET # for TUN 2207 depends on BLOCK 2208 select TEST_LKM 2209 select XFS_FS 2210 select TUN 2211 select BTRFS_FS 2212 help 2213 Test the kernel's module loading mechanism: kmod. kmod implements 2214 support to load modules using the Linux kernel's usermode helper. 2215 This test provides a series of tests against kmod. 2216 2217 Although technically you can either build test_kmod as a module or 2218 into the kernel we disallow building it into the kernel since 2219 it stress tests request_module() and this will very likely cause 2220 some issues by taking over precious threads available from other 2221 module load requests, ultimately this could be fatal. 2222 2223 To run tests run: 2224 2225 tools/testing/selftests/kmod/kmod.sh --help 2226 2227 If unsure, say N. 2228 2229config TEST_DEBUG_VIRTUAL 2230 tristate "Test CONFIG_DEBUG_VIRTUAL feature" 2231 depends on DEBUG_VIRTUAL 2232 help 2233 Test the kernel's ability to detect incorrect calls to 2234 virt_to_phys() done against the non-linear part of the 2235 kernel's virtual address map. 2236 2237 If unsure, say N. 2238 2239config TEST_MEMCAT_P 2240 tristate "Test memcat_p() helper function" 2241 help 2242 Test the memcat_p() helper for correctly merging two 2243 pointer arrays together. 2244 2245 If unsure, say N. 2246 2247config TEST_LIVEPATCH 2248 tristate "Test livepatching" 2249 default n 2250 depends on DYNAMIC_DEBUG 2251 depends on LIVEPATCH 2252 depends on m 2253 help 2254 Test kernel livepatching features for correctness. The tests will 2255 load test modules that will be livepatched in various scenarios. 2256 2257 To run all the livepatching tests: 2258 2259 make -C tools/testing/selftests TARGETS=livepatch run_tests 2260 2261 Alternatively, individual tests may be invoked: 2262 2263 tools/testing/selftests/livepatch/test-callbacks.sh 2264 tools/testing/selftests/livepatch/test-livepatch.sh 2265 tools/testing/selftests/livepatch/test-shadow-vars.sh 2266 2267 If unsure, say N. 2268 2269config TEST_OBJAGG 2270 tristate "Perform selftest on object aggreration manager" 2271 default n 2272 depends on OBJAGG 2273 help 2274 Enable this option to test object aggregation manager on boot 2275 (or module load). 2276 2277 2278config TEST_STACKINIT 2279 tristate "Test level of stack variable initialization" 2280 help 2281 Test if the kernel is zero-initializing stack variables and 2282 padding. Coverage is controlled by compiler flags, 2283 CONFIG_GCC_PLUGIN_STRUCTLEAK, CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF, 2284 or CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL. 2285 2286 If unsure, say N. 2287 2288config TEST_MEMINIT 2289 tristate "Test heap/page initialization" 2290 help 2291 Test if the kernel is zero-initializing heap and page allocations. 2292 This can be useful to test init_on_alloc and init_on_free features. 2293 2294 If unsure, say N. 2295 2296config TEST_HMM 2297 tristate "Test HMM (Heterogeneous Memory Management)" 2298 depends on TRANSPARENT_HUGEPAGE 2299 depends on DEVICE_PRIVATE 2300 select HMM_MIRROR 2301 select MMU_NOTIFIER 2302 help 2303 This is a pseudo device driver solely for testing HMM. 2304 Say M here if you want to build the HMM test module. 2305 Doing so will allow you to run tools/testing/selftest/vm/hmm-tests. 2306 2307 If unsure, say N. 2308 2309endif # RUNTIME_TESTING_MENU 2310 2311config MEMTEST 2312 bool "Memtest" 2313 help 2314 This option adds a kernel parameter 'memtest', which allows memtest 2315 to be set. 2316 memtest=0, mean disabled; -- default 2317 memtest=1, mean do 1 test pattern; 2318 ... 2319 memtest=17, mean do 17 test patterns. 2320 If you are unsure how to answer this question, answer N. 2321 2322 2323 2324config HYPERV_TESTING 2325 bool "Microsoft Hyper-V driver testing" 2326 default n 2327 depends on HYPERV && DEBUG_FS 2328 help 2329 Select this option to enable Hyper-V vmbus testing. 2330 2331endmenu # "Kernel Testing and Coverage" 2332 2333endmenu # Kernel hacking 2334