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 215if DEBUG_INFO 216 217config DEBUG_INFO_REDUCED 218 bool "Reduce debugging information" 219 help 220 If you say Y here gcc is instructed to generate less debugging 221 information for structure types. This means that tools that 222 need full debugging information (like kgdb or systemtap) won't 223 be happy. But if you merely need debugging information to 224 resolve line numbers there is no loss. Advantage is that 225 build directory object sizes shrink dramatically over a full 226 DEBUG_INFO build and compile times are reduced too. 227 Only works with newer gcc versions. 228 229config DEBUG_INFO_COMPRESSED 230 bool "Compressed debugging information" 231 depends on $(cc-option,-gz=zlib) 232 depends on $(ld-option,--compress-debug-sections=zlib) 233 help 234 Compress the debug information using zlib. Requires GCC 5.0+ or Clang 235 5.0+, binutils 2.26+, and zlib. 236 237 Users of dpkg-deb via scripts/package/builddeb may find an increase in 238 size of their debug .deb packages with this config set, due to the 239 debug info being compressed with zlib, then the object files being 240 recompressed with a different compression scheme. But this is still 241 preferable to setting $KDEB_COMPRESS to "none" which would be even 242 larger. 243 244config DEBUG_INFO_SPLIT 245 bool "Produce split debuginfo in .dwo files" 246 depends on $(cc-option,-gsplit-dwarf) 247 help 248 Generate debug info into separate .dwo files. This significantly 249 reduces the build directory size for builds with DEBUG_INFO, 250 because it stores the information only once on disk in .dwo 251 files instead of multiple times in object files and executables. 252 In addition the debug information is also compressed. 253 254 Requires recent gcc (4.7+) and recent gdb/binutils. 255 Any tool that packages or reads debug information would need 256 to know about the .dwo files and include them. 257 Incompatible with older versions of ccache. 258 259config DEBUG_INFO_DWARF4 260 bool "Generate dwarf4 debuginfo" 261 depends on $(cc-option,-gdwarf-4) 262 help 263 Generate dwarf4 debug info. This requires recent versions 264 of gcc and gdb. It makes the debug information larger. 265 But it significantly improves the success of resolving 266 variables in gdb on optimized code. 267 268config DEBUG_INFO_BTF 269 bool "Generate BTF typeinfo" 270 depends on !DEBUG_INFO_SPLIT && !DEBUG_INFO_REDUCED 271 depends on !GCC_PLUGIN_RANDSTRUCT || COMPILE_TEST 272 help 273 Generate deduplicated BTF type information from DWARF debug info. 274 Turning this on expects presence of pahole tool, which will convert 275 DWARF type info into equivalent deduplicated BTF type info. 276 277config PAHOLE_HAS_SPLIT_BTF 278 def_bool $(success, test `$(PAHOLE) --version | sed -E 's/v([0-9]+)\.([0-9]+)/\1\2/'` -ge "119") 279 280config DEBUG_INFO_BTF_MODULES 281 def_bool y 282 depends on DEBUG_INFO_BTF && MODULES && PAHOLE_HAS_SPLIT_BTF 283 help 284 Generate compact split BTF type information for kernel modules. 285 286config GDB_SCRIPTS 287 bool "Provide GDB scripts for kernel debugging" 288 help 289 This creates the required links to GDB helper scripts in the 290 build directory. If you load vmlinux into gdb, the helper 291 scripts will be automatically imported by gdb as well, and 292 additional functions are available to analyze a Linux kernel 293 instance. See Documentation/dev-tools/gdb-kernel-debugging.rst 294 for further details. 295 296endif # DEBUG_INFO 297 298config FRAME_WARN 299 int "Warn for stack frames larger than" 300 range 0 8192 301 default 2048 if GCC_PLUGIN_LATENT_ENTROPY 302 default 1280 if (!64BIT && PARISC) 303 default 1024 if (!64BIT && !PARISC) 304 default 2048 if 64BIT 305 help 306 Tell gcc to warn at build time for stack frames larger than this. 307 Setting this too low will cause a lot of warnings. 308 Setting it to 0 disables the warning. 309 310config STRIP_ASM_SYMS 311 bool "Strip assembler-generated symbols during link" 312 default n 313 help 314 Strip internal assembler-generated symbols during a link (symbols 315 that look like '.Lxxx') so they don't pollute the output of 316 get_wchan() and suchlike. 317 318config READABLE_ASM 319 bool "Generate readable assembler code" 320 depends on DEBUG_KERNEL 321 help 322 Disable some compiler optimizations that tend to generate human unreadable 323 assembler output. This may make the kernel slightly slower, but it helps 324 to keep kernel developers who have to stare a lot at assembler listings 325 sane. 326 327config HEADERS_INSTALL 328 bool "Install uapi headers to usr/include" 329 depends on !UML 330 help 331 This option will install uapi headers (headers exported to user-space) 332 into the usr/include directory for use during the kernel build. 333 This is unneeded for building the kernel itself, but needed for some 334 user-space program samples. It is also needed by some features such 335 as uapi header sanity checks. 336 337config DEBUG_SECTION_MISMATCH 338 bool "Enable full Section mismatch analysis" 339 help 340 The section mismatch analysis checks if there are illegal 341 references from one section to another section. 342 During linktime or runtime, some sections are dropped; 343 any use of code/data previously in these sections would 344 most likely result in an oops. 345 In the code, functions and variables are annotated with 346 __init,, etc. (see the full list in include/linux/init.h), 347 which results in the code/data being placed in specific sections. 348 The section mismatch analysis is always performed after a full 349 kernel build, and enabling this option causes the following 350 additional step to occur: 351 - Add the option -fno-inline-functions-called-once to gcc commands. 352 When inlining a function annotated with __init in a non-init 353 function, we would lose the section information and thus 354 the analysis would not catch the illegal reference. 355 This option tells gcc to inline less (but it does result in 356 a larger kernel). 357 358config SECTION_MISMATCH_WARN_ONLY 359 bool "Make section mismatch errors non-fatal" 360 default y 361 help 362 If you say N here, the build process will fail if there are any 363 section mismatch, instead of just throwing warnings. 364 365 If unsure, say Y. 366 367config DEBUG_FORCE_FUNCTION_ALIGN_32B 368 bool "Force all function address 32B aligned" if EXPERT 369 help 370 There are cases that a commit from one domain changes the function 371 address alignment of other domains, and cause magic performance 372 bump (regression or improvement). Enable this option will help to 373 verify if the bump is caused by function alignment changes, while 374 it will slightly increase the kernel size and affect icache usage. 375 376 It is mainly for debug and performance tuning use. 377 378# 379# Select this config option from the architecture Kconfig, if it 380# is preferred to always offer frame pointers as a config 381# option on the architecture (regardless of KERNEL_DEBUG): 382# 383config ARCH_WANT_FRAME_POINTERS 384 bool 385 386config FRAME_POINTER 387 bool "Compile the kernel with frame pointers" 388 depends on DEBUG_KERNEL && (M68K || UML || SUPERH) || ARCH_WANT_FRAME_POINTERS 389 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS 390 help 391 If you say Y here the resulting kernel image will be slightly 392 larger and slower, but it gives very useful debugging information 393 in case of kernel bugs. (precise oopses/stacktraces/warnings) 394 395config STACK_VALIDATION 396 bool "Compile-time stack metadata validation" 397 depends on HAVE_STACK_VALIDATION 398 default n 399 help 400 Add compile-time checks to validate stack metadata, including frame 401 pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure 402 that runtime stack traces are more reliable. 403 404 This is also a prerequisite for generation of ORC unwind data, which 405 is needed for CONFIG_UNWINDER_ORC. 406 407 For more information, see 408 tools/objtool/Documentation/stack-validation.txt. 409 410config VMLINUX_VALIDATION 411 bool 412 depends on STACK_VALIDATION && DEBUG_ENTRY && !PARAVIRT 413 default y 414 415config DEBUG_FORCE_WEAK_PER_CPU 416 bool "Force weak per-cpu definitions" 417 depends on DEBUG_KERNEL 418 help 419 s390 and alpha require percpu variables in modules to be 420 defined weak to work around addressing range issue which 421 puts the following two restrictions on percpu variable 422 definitions. 423 424 1. percpu symbols must be unique whether static or not 425 2. percpu variables can't be defined inside a function 426 427 To ensure that generic code follows the above rules, this 428 option forces all percpu variables to be defined as weak. 429 430endmenu # "Compiler options" 431 432menu "Generic Kernel Debugging Instruments" 433 434config MAGIC_SYSRQ 435 bool "Magic SysRq key" 436 depends on !UML 437 help 438 If you say Y here, you will have some control over the system even 439 if the system crashes for example during kernel debugging (e.g., you 440 will be able to flush the buffer cache to disk, reboot the system 441 immediately or dump some status information). This is accomplished 442 by pressing various keys while holding SysRq (Alt+PrintScreen). It 443 also works on a serial console (on PC hardware at least), if you 444 send a BREAK and then within 5 seconds a command keypress. The 445 keys are documented in <file:Documentation/admin-guide/sysrq.rst>. 446 Don't say Y unless you really know what this hack does. 447 448config MAGIC_SYSRQ_DEFAULT_ENABLE 449 hex "Enable magic SysRq key functions by default" 450 depends on MAGIC_SYSRQ 451 default 0x1 452 help 453 Specifies which SysRq key functions are enabled by default. 454 This may be set to 1 or 0 to enable or disable them all, or 455 to a bitmask as described in Documentation/admin-guide/sysrq.rst. 456 457config MAGIC_SYSRQ_SERIAL 458 bool "Enable magic SysRq key over serial" 459 depends on MAGIC_SYSRQ 460 default y 461 help 462 Many embedded boards have a disconnected TTL level serial which can 463 generate some garbage that can lead to spurious false sysrq detects. 464 This option allows you to decide whether you want to enable the 465 magic SysRq key. 466 467config MAGIC_SYSRQ_SERIAL_SEQUENCE 468 string "Char sequence that enables magic SysRq over serial" 469 depends on MAGIC_SYSRQ_SERIAL 470 default "" 471 help 472 Specifies a sequence of characters that can follow BREAK to enable 473 SysRq on a serial console. 474 475 If unsure, leave an empty string and the option will not be enabled. 476 477config DEBUG_FS 478 bool "Debug Filesystem" 479 help 480 debugfs is a virtual file system that kernel developers use to put 481 debugging files into. Enable this option to be able to read and 482 write to these files. 483 484 For detailed documentation on the debugfs API, see 485 Documentation/filesystems/. 486 487 If unsure, say N. 488 489choice 490 prompt "Debugfs default access" 491 depends on DEBUG_FS 492 default DEBUG_FS_ALLOW_ALL 493 help 494 This selects the default access restrictions for debugfs. 495 It can be overridden with kernel command line option 496 debugfs=[on,no-mount,off]. The restrictions apply for API access 497 and filesystem registration. 498 499config DEBUG_FS_ALLOW_ALL 500 bool "Access normal" 501 help 502 No restrictions apply. Both API and filesystem registration 503 is on. This is the normal default operation. 504 505config DEBUG_FS_DISALLOW_MOUNT 506 bool "Do not register debugfs as filesystem" 507 help 508 The API is open but filesystem is not loaded. Clients can still do 509 their work and read with debug tools that do not need 510 debugfs filesystem. 511 512config DEBUG_FS_ALLOW_NONE 513 bool "No access" 514 help 515 Access is off. Clients get -PERM when trying to create nodes in 516 debugfs tree and debugfs is not registered as a filesystem. 517 Client can then back-off or continue without debugfs access. 518 519endchoice 520 521source "lib/Kconfig.kgdb" 522source "lib/Kconfig.ubsan" 523source "lib/Kconfig.kcsan" 524 525endmenu 526 527config DEBUG_KERNEL 528 bool "Kernel debugging" 529 help 530 Say Y here if you are developing drivers or trying to debug and 531 identify kernel problems. 532 533config DEBUG_MISC 534 bool "Miscellaneous debug code" 535 default DEBUG_KERNEL 536 depends on DEBUG_KERNEL 537 help 538 Say Y here if you need to enable miscellaneous debug code that should 539 be under a more specific debug option but isn't. 540 541 542menu "Memory Debugging" 543 544source "mm/Kconfig.debug" 545 546config DEBUG_OBJECTS 547 bool "Debug object operations" 548 depends on DEBUG_KERNEL 549 help 550 If you say Y here, additional code will be inserted into the 551 kernel to track the life time of various objects and validate 552 the operations on those objects. 553 554config DEBUG_OBJECTS_SELFTEST 555 bool "Debug objects selftest" 556 depends on DEBUG_OBJECTS 557 help 558 This enables the selftest of the object debug code. 559 560config DEBUG_OBJECTS_FREE 561 bool "Debug objects in freed memory" 562 depends on DEBUG_OBJECTS 563 help 564 This enables checks whether a k/v free operation frees an area 565 which contains an object which has not been deactivated 566 properly. This can make kmalloc/kfree-intensive workloads 567 much slower. 568 569config DEBUG_OBJECTS_TIMERS 570 bool "Debug timer objects" 571 depends on DEBUG_OBJECTS 572 help 573 If you say Y here, additional code will be inserted into the 574 timer routines to track the life time of timer objects and 575 validate the timer operations. 576 577config DEBUG_OBJECTS_WORK 578 bool "Debug work objects" 579 depends on DEBUG_OBJECTS 580 help 581 If you say Y here, additional code will be inserted into the 582 work queue routines to track the life time of work objects and 583 validate the work operations. 584 585config DEBUG_OBJECTS_RCU_HEAD 586 bool "Debug RCU callbacks objects" 587 depends on DEBUG_OBJECTS 588 help 589 Enable this to turn on debugging of RCU list heads (call_rcu() usage). 590 591config DEBUG_OBJECTS_PERCPU_COUNTER 592 bool "Debug percpu counter objects" 593 depends on DEBUG_OBJECTS 594 help 595 If you say Y here, additional code will be inserted into the 596 percpu counter routines to track the life time of percpu counter 597 objects and validate the percpu counter operations. 598 599config DEBUG_OBJECTS_ENABLE_DEFAULT 600 int "debug_objects bootup default value (0-1)" 601 range 0 1 602 default "1" 603 depends on DEBUG_OBJECTS 604 help 605 Debug objects boot parameter default value 606 607config DEBUG_SLAB 608 bool "Debug slab memory allocations" 609 depends on DEBUG_KERNEL && SLAB 610 help 611 Say Y here to have the kernel do limited verification on memory 612 allocation as well as poisoning memory on free to catch use of freed 613 memory. This can make kmalloc/kfree-intensive workloads much slower. 614 615config SLUB_DEBUG_ON 616 bool "SLUB debugging on by default" 617 depends on SLUB && SLUB_DEBUG 618 default n 619 help 620 Boot with debugging on by default. SLUB boots by default with 621 the runtime debug capabilities switched off. Enabling this is 622 equivalent to specifying the "slub_debug" parameter on boot. 623 There is no support for more fine grained debug control like 624 possible with slub_debug=xxx. SLUB debugging may be switched 625 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying 626 "slub_debug=-". 627 628config SLUB_STATS 629 default n 630 bool "Enable SLUB performance statistics" 631 depends on SLUB && SYSFS 632 help 633 SLUB statistics are useful to debug SLUBs allocation behavior in 634 order find ways to optimize the allocator. This should never be 635 enabled for production use since keeping statistics slows down 636 the allocator by a few percentage points. The slabinfo command 637 supports the determination of the most active slabs to figure 638 out which slabs are relevant to a particular load. 639 Try running: slabinfo -DA 640 641config HAVE_DEBUG_KMEMLEAK 642 bool 643 644config DEBUG_KMEMLEAK 645 bool "Kernel memory leak detector" 646 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK 647 select DEBUG_FS 648 select STACKTRACE if STACKTRACE_SUPPORT 649 select KALLSYMS 650 select CRC32 651 help 652 Say Y here if you want to enable the memory leak 653 detector. The memory allocation/freeing is traced in a way 654 similar to the Boehm's conservative garbage collector, the 655 difference being that the orphan objects are not freed but 656 only shown in /sys/kernel/debug/kmemleak. Enabling this 657 feature will introduce an overhead to memory 658 allocations. See Documentation/dev-tools/kmemleak.rst for more 659 details. 660 661 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances 662 of finding leaks due to the slab objects poisoning. 663 664 In order to access the kmemleak file, debugfs needs to be 665 mounted (usually at /sys/kernel/debug). 666 667config DEBUG_KMEMLEAK_MEM_POOL_SIZE 668 int "Kmemleak memory pool size" 669 depends on DEBUG_KMEMLEAK 670 range 200 1000000 671 default 16000 672 help 673 Kmemleak must track all the memory allocations to avoid 674 reporting false positives. Since memory may be allocated or 675 freed before kmemleak is fully initialised, use a static pool 676 of metadata objects to track such callbacks. After kmemleak is 677 fully initialised, this memory pool acts as an emergency one 678 if slab allocations fail. 679 680config DEBUG_KMEMLEAK_TEST 681 tristate "Simple test for the kernel memory leak detector" 682 depends on DEBUG_KMEMLEAK && m 683 help 684 This option enables a module that explicitly leaks memory. 685 686 If unsure, say N. 687 688config DEBUG_KMEMLEAK_DEFAULT_OFF 689 bool "Default kmemleak to off" 690 depends on DEBUG_KMEMLEAK 691 help 692 Say Y here to disable kmemleak by default. It can then be enabled 693 on the command line via kmemleak=on. 694 695config DEBUG_KMEMLEAK_AUTO_SCAN 696 bool "Enable kmemleak auto scan thread on boot up" 697 default y 698 depends on DEBUG_KMEMLEAK 699 help 700 Depending on the cpu, kmemleak scan may be cpu intensive and can 701 stall user tasks at times. This option enables/disables automatic 702 kmemleak scan at boot up. 703 704 Say N here to disable kmemleak auto scan thread to stop automatic 705 scanning. Disabling this option disables automatic reporting of 706 memory leaks. 707 708 If unsure, say Y. 709 710config DEBUG_STACK_USAGE 711 bool "Stack utilization instrumentation" 712 depends on DEBUG_KERNEL && !IA64 713 help 714 Enables the display of the minimum amount of free stack which each 715 task has ever had available in the sysrq-T and sysrq-P debug output. 716 717 This option will slow down process creation somewhat. 718 719config SCHED_STACK_END_CHECK 720 bool "Detect stack corruption on calls to schedule()" 721 depends on DEBUG_KERNEL 722 default n 723 help 724 This option checks for a stack overrun on calls to schedule(). 725 If the stack end location is found to be over written always panic as 726 the content of the corrupted region can no longer be trusted. 727 This is to ensure no erroneous behaviour occurs which could result in 728 data corruption or a sporadic crash at a later stage once the region 729 is examined. The runtime overhead introduced is minimal. 730 731config ARCH_HAS_DEBUG_VM_PGTABLE 732 bool 733 help 734 An architecture should select this when it can successfully 735 build and run DEBUG_VM_PGTABLE. 736 737config DEBUG_VM 738 bool "Debug VM" 739 depends on DEBUG_KERNEL 740 help 741 Enable this to turn on extended checks in the virtual-memory system 742 that may impact performance. 743 744 If unsure, say N. 745 746config DEBUG_VM_VMACACHE 747 bool "Debug VMA caching" 748 depends on DEBUG_VM 749 help 750 Enable this to turn on VMA caching debug information. Doing so 751 can cause significant overhead, so only enable it in non-production 752 environments. 753 754 If unsure, say N. 755 756config DEBUG_VM_RB 757 bool "Debug VM red-black trees" 758 depends on DEBUG_VM 759 help 760 Enable VM red-black tree debugging information and extra validations. 761 762 If unsure, say N. 763 764config DEBUG_VM_PGFLAGS 765 bool "Debug page-flags operations" 766 depends on DEBUG_VM 767 help 768 Enables extra validation on page flags operations. 769 770 If unsure, say N. 771 772config DEBUG_VM_PGTABLE 773 bool "Debug arch page table for semantics compliance" 774 depends on MMU 775 depends on ARCH_HAS_DEBUG_VM_PGTABLE 776 default y if DEBUG_VM 777 help 778 This option provides a debug method which can be used to test 779 architecture page table helper functions on various platforms in 780 verifying if they comply with expected generic MM semantics. This 781 will help architecture code in making sure that any changes or 782 new additions of these helpers still conform to expected 783 semantics of the generic MM. Platforms will have to opt in for 784 this through ARCH_HAS_DEBUG_VM_PGTABLE. 785 786 If unsure, say N. 787 788config ARCH_HAS_DEBUG_VIRTUAL 789 bool 790 791config DEBUG_VIRTUAL 792 bool "Debug VM translations" 793 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL 794 help 795 Enable some costly sanity checks in virtual to page code. This can 796 catch mistakes with virt_to_page() and friends. 797 798 If unsure, say N. 799 800config DEBUG_NOMMU_REGIONS 801 bool "Debug the global anon/private NOMMU mapping region tree" 802 depends on DEBUG_KERNEL && !MMU 803 help 804 This option causes the global tree of anonymous and private mapping 805 regions to be regularly checked for invalid topology. 806 807config DEBUG_MEMORY_INIT 808 bool "Debug memory initialisation" if EXPERT 809 default !EXPERT 810 help 811 Enable this for additional checks during memory initialisation. 812 The sanity checks verify aspects of the VM such as the memory model 813 and other information provided by the architecture. Verbose 814 information will be printed at KERN_DEBUG loglevel depending 815 on the mminit_loglevel= command-line option. 816 817 If unsure, say Y 818 819config MEMORY_NOTIFIER_ERROR_INJECT 820 tristate "Memory hotplug notifier error injection module" 821 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION 822 help 823 This option provides the ability to inject artificial errors to 824 memory hotplug notifier chain callbacks. It is controlled through 825 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory 826 827 If the notifier call chain should be failed with some events 828 notified, write the error code to "actions/<notifier event>/error". 829 830 Example: Inject memory hotplug offline error (-12 == -ENOMEM) 831 832 # cd /sys/kernel/debug/notifier-error-inject/memory 833 # echo -12 > actions/MEM_GOING_OFFLINE/error 834 # echo offline > /sys/devices/system/memory/memoryXXX/state 835 bash: echo: write error: Cannot allocate memory 836 837 To compile this code as a module, choose M here: the module will 838 be called memory-notifier-error-inject. 839 840 If unsure, say N. 841 842config DEBUG_PER_CPU_MAPS 843 bool "Debug access to per_cpu maps" 844 depends on DEBUG_KERNEL 845 depends on SMP 846 help 847 Say Y to verify that the per_cpu map being accessed has 848 been set up. This adds a fair amount of code to kernel memory 849 and decreases performance. 850 851 Say N if unsure. 852 853config DEBUG_KMAP_LOCAL 854 bool "Debug kmap_local temporary mappings" 855 depends on DEBUG_KERNEL && KMAP_LOCAL 856 help 857 This option enables additional error checking for the kmap_local 858 infrastructure. Disable for production use. 859 860config ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP 861 bool 862 863config DEBUG_KMAP_LOCAL_FORCE_MAP 864 bool "Enforce kmap_local temporary mappings" 865 depends on DEBUG_KERNEL && ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP 866 select KMAP_LOCAL 867 select DEBUG_KMAP_LOCAL 868 help 869 This option enforces temporary mappings through the kmap_local 870 mechanism for non-highmem pages and on non-highmem systems. 871 Disable this for production systems! 872 873config DEBUG_HIGHMEM 874 bool "Highmem debugging" 875 depends on DEBUG_KERNEL && HIGHMEM 876 select DEBUG_KMAP_LOCAL_FORCE_MAP if ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP 877 select DEBUG_KMAP_LOCAL 878 help 879 This option enables additional error checking for high memory 880 systems. Disable for production systems. 881 882config HAVE_DEBUG_STACKOVERFLOW 883 bool 884 885config DEBUG_STACKOVERFLOW 886 bool "Check for stack overflows" 887 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW 888 help 889 Say Y here if you want to check for overflows of kernel, IRQ 890 and exception stacks (if your architecture uses them). This 891 option will show detailed messages if free stack space drops 892 below a certain limit. 893 894 These kinds of bugs usually occur when call-chains in the 895 kernel get too deep, especially when interrupts are 896 involved. 897 898 Use this in cases where you see apparently random memory 899 corruption, especially if it appears in 'struct thread_info' 900 901 If in doubt, say "N". 902 903source "lib/Kconfig.kasan" 904 905endmenu # "Memory Debugging" 906 907config DEBUG_SHIRQ 908 bool "Debug shared IRQ handlers" 909 depends on DEBUG_KERNEL 910 help 911 Enable this to generate a spurious interrupt just before a shared 912 interrupt handler is deregistered (generating one when registering 913 is currently disabled). Drivers need to handle this correctly. Some 914 don't and need to be caught. 915 916menu "Debug Oops, Lockups and Hangs" 917 918config PANIC_ON_OOPS 919 bool "Panic on Oops" 920 help 921 Say Y here to enable the kernel to panic when it oopses. This 922 has the same effect as setting oops=panic on the kernel command 923 line. 924 925 This feature is useful to ensure that the kernel does not do 926 anything erroneous after an oops which could result in data 927 corruption or other issues. 928 929 Say N if unsure. 930 931config PANIC_ON_OOPS_VALUE 932 int 933 range 0 1 934 default 0 if !PANIC_ON_OOPS 935 default 1 if PANIC_ON_OOPS 936 937config PANIC_TIMEOUT 938 int "panic timeout" 939 default 0 940 help 941 Set the timeout value (in seconds) until a reboot occurs when 942 the kernel panics. If n = 0, then we wait forever. A timeout 943 value n > 0 will wait n seconds before rebooting, while a timeout 944 value n < 0 will reboot immediately. 945 946config LOCKUP_DETECTOR 947 bool 948 949config SOFTLOCKUP_DETECTOR 950 bool "Detect Soft Lockups" 951 depends on DEBUG_KERNEL && !S390 952 select LOCKUP_DETECTOR 953 help 954 Say Y here to enable the kernel to act as a watchdog to detect 955 soft lockups. 956 957 Softlockups are bugs that cause the kernel to loop in kernel 958 mode for more than 20 seconds, without giving other tasks a 959 chance to run. The current stack trace is displayed upon 960 detection and the system will stay locked up. 961 962config BOOTPARAM_SOFTLOCKUP_PANIC 963 bool "Panic (Reboot) On Soft Lockups" 964 depends on SOFTLOCKUP_DETECTOR 965 help 966 Say Y here to enable the kernel to panic on "soft lockups", 967 which are bugs that cause the kernel to loop in kernel 968 mode for more than 20 seconds (configurable using the watchdog_thresh 969 sysctl), without giving other tasks a chance to run. 970 971 The panic can be used in combination with panic_timeout, 972 to cause the system to reboot automatically after a 973 lockup has been detected. This feature is useful for 974 high-availability systems that have uptime guarantees and 975 where a lockup must be resolved ASAP. 976 977 Say N if unsure. 978 979config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE 980 int 981 depends on SOFTLOCKUP_DETECTOR 982 range 0 1 983 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC 984 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC 985 986config HARDLOCKUP_DETECTOR_PERF 987 bool 988 select SOFTLOCKUP_DETECTOR 989 990# 991# Enables a timestamp based low pass filter to compensate for perf based 992# hard lockup detection which runs too fast due to turbo modes. 993# 994config HARDLOCKUP_CHECK_TIMESTAMP 995 bool 996 997# 998# arch/ can define HAVE_HARDLOCKUP_DETECTOR_ARCH to provide their own hard 999# lockup detector rather than the perf based detector. 1000# 1001config HARDLOCKUP_DETECTOR 1002 bool "Detect Hard Lockups" 1003 depends on DEBUG_KERNEL && !S390 1004 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_ARCH 1005 select LOCKUP_DETECTOR 1006 select HARDLOCKUP_DETECTOR_PERF if HAVE_HARDLOCKUP_DETECTOR_PERF 1007 select HARDLOCKUP_DETECTOR_ARCH if HAVE_HARDLOCKUP_DETECTOR_ARCH 1008 help 1009 Say Y here to enable the kernel to act as a watchdog to detect 1010 hard lockups. 1011 1012 Hardlockups are bugs that cause the CPU to loop in kernel mode 1013 for more than 10 seconds, without letting other interrupts have a 1014 chance to run. The current stack trace is displayed upon detection 1015 and the system will stay locked up. 1016 1017config BOOTPARAM_HARDLOCKUP_PANIC 1018 bool "Panic (Reboot) On Hard Lockups" 1019 depends on HARDLOCKUP_DETECTOR 1020 help 1021 Say Y here to enable the kernel to panic on "hard lockups", 1022 which are bugs that cause the kernel to loop in kernel 1023 mode with interrupts disabled for more than 10 seconds (configurable 1024 using the watchdog_thresh sysctl). 1025 1026 Say N if unsure. 1027 1028config BOOTPARAM_HARDLOCKUP_PANIC_VALUE 1029 int 1030 depends on HARDLOCKUP_DETECTOR 1031 range 0 1 1032 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC 1033 default 1 if BOOTPARAM_HARDLOCKUP_PANIC 1034 1035config DETECT_HUNG_TASK 1036 bool "Detect Hung Tasks" 1037 depends on DEBUG_KERNEL 1038 default SOFTLOCKUP_DETECTOR 1039 help 1040 Say Y here to enable the kernel to detect "hung tasks", 1041 which are bugs that cause the task to be stuck in 1042 uninterruptible "D" state indefinitely. 1043 1044 When a hung task is detected, the kernel will print the 1045 current stack trace (which you should report), but the 1046 task will stay in uninterruptible state. If lockdep is 1047 enabled then all held locks will also be reported. This 1048 feature has negligible overhead. 1049 1050config DEFAULT_HUNG_TASK_TIMEOUT 1051 int "Default timeout for hung task detection (in seconds)" 1052 depends on DETECT_HUNG_TASK 1053 default 120 1054 help 1055 This option controls the default timeout (in seconds) used 1056 to determine when a task has become non-responsive and should 1057 be considered hung. 1058 1059 It can be adjusted at runtime via the kernel.hung_task_timeout_secs 1060 sysctl or by writing a value to 1061 /proc/sys/kernel/hung_task_timeout_secs. 1062 1063 A timeout of 0 disables the check. The default is two minutes. 1064 Keeping the default should be fine in most cases. 1065 1066config BOOTPARAM_HUNG_TASK_PANIC 1067 bool "Panic (Reboot) On Hung Tasks" 1068 depends on DETECT_HUNG_TASK 1069 help 1070 Say Y here to enable the kernel to panic on "hung tasks", 1071 which are bugs that cause the kernel to leave a task stuck 1072 in uninterruptible "D" state. 1073 1074 The panic can be used in combination with panic_timeout, 1075 to cause the system to reboot automatically after a 1076 hung task has been detected. This feature is useful for 1077 high-availability systems that have uptime guarantees and 1078 where a hung tasks must be resolved ASAP. 1079 1080 Say N if unsure. 1081 1082config BOOTPARAM_HUNG_TASK_PANIC_VALUE 1083 int 1084 depends on DETECT_HUNG_TASK 1085 range 0 1 1086 default 0 if !BOOTPARAM_HUNG_TASK_PANIC 1087 default 1 if BOOTPARAM_HUNG_TASK_PANIC 1088 1089config WQ_WATCHDOG 1090 bool "Detect Workqueue Stalls" 1091 depends on DEBUG_KERNEL 1092 help 1093 Say Y here to enable stall detection on workqueues. If a 1094 worker pool doesn't make forward progress on a pending work 1095 item for over a given amount of time, 30s by default, a 1096 warning message is printed along with dump of workqueue 1097 state. This can be configured through kernel parameter 1098 "workqueue.watchdog_thresh" and its sysfs counterpart. 1099 1100config TEST_LOCKUP 1101 tristate "Test module to generate lockups" 1102 depends on m 1103 help 1104 This builds the "test_lockup" module that helps to make sure 1105 that watchdogs and lockup detectors are working properly. 1106 1107 Depending on module parameters it could emulate soft or hard 1108 lockup, "hung task", or locking arbitrary lock for a long time. 1109 Also it could generate series of lockups with cooling-down periods. 1110 1111 If unsure, say N. 1112 1113endmenu # "Debug lockups and hangs" 1114 1115menu "Scheduler Debugging" 1116 1117config SCHED_DEBUG 1118 bool "Collect scheduler debugging info" 1119 depends on DEBUG_KERNEL && PROC_FS 1120 default y 1121 help 1122 If you say Y here, the /proc/sched_debug file will be provided 1123 that can help debug the scheduler. The runtime overhead of this 1124 option is minimal. 1125 1126config SCHED_INFO 1127 bool 1128 default n 1129 1130config SCHEDSTATS 1131 bool "Collect scheduler statistics" 1132 depends on DEBUG_KERNEL && PROC_FS 1133 select SCHED_INFO 1134 help 1135 If you say Y here, additional code will be inserted into the 1136 scheduler and related routines to collect statistics about 1137 scheduler behavior and provide them in /proc/schedstat. These 1138 stats may be useful for both tuning and debugging the scheduler 1139 If you aren't debugging the scheduler or trying to tune a specific 1140 application, you can say N to avoid the very slight overhead 1141 this adds. 1142 1143endmenu 1144 1145config DEBUG_TIMEKEEPING 1146 bool "Enable extra timekeeping sanity checking" 1147 help 1148 This option will enable additional timekeeping sanity checks 1149 which may be helpful when diagnosing issues where timekeeping 1150 problems are suspected. 1151 1152 This may include checks in the timekeeping hotpaths, so this 1153 option may have a (very small) performance impact to some 1154 workloads. 1155 1156 If unsure, say N. 1157 1158config DEBUG_PREEMPT 1159 bool "Debug preemptible kernel" 1160 depends on DEBUG_KERNEL && PREEMPTION && TRACE_IRQFLAGS_SUPPORT 1161 default y 1162 help 1163 If you say Y here then the kernel will use a debug variant of the 1164 commonly used smp_processor_id() function and will print warnings 1165 if kernel code uses it in a preemption-unsafe way. Also, the kernel 1166 will detect preemption count underflows. 1167 1168menu "Lock Debugging (spinlocks, mutexes, etc...)" 1169 1170config LOCK_DEBUGGING_SUPPORT 1171 bool 1172 depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1173 default y 1174 1175config PROVE_LOCKING 1176 bool "Lock debugging: prove locking correctness" 1177 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1178 select LOCKDEP 1179 select DEBUG_SPINLOCK 1180 select DEBUG_MUTEXES 1181 select DEBUG_RT_MUTEXES if RT_MUTEXES 1182 select DEBUG_RWSEMS 1183 select DEBUG_WW_MUTEX_SLOWPATH 1184 select DEBUG_LOCK_ALLOC 1185 select PREEMPT_COUNT if !ARCH_NO_PREEMPT 1186 select TRACE_IRQFLAGS 1187 default n 1188 help 1189 This feature enables the kernel to prove that all locking 1190 that occurs in the kernel runtime is mathematically 1191 correct: that under no circumstance could an arbitrary (and 1192 not yet triggered) combination of observed locking 1193 sequences (on an arbitrary number of CPUs, running an 1194 arbitrary number of tasks and interrupt contexts) cause a 1195 deadlock. 1196 1197 In short, this feature enables the kernel to report locking 1198 related deadlocks before they actually occur. 1199 1200 The proof does not depend on how hard and complex a 1201 deadlock scenario would be to trigger: how many 1202 participant CPUs, tasks and irq-contexts would be needed 1203 for it to trigger. The proof also does not depend on 1204 timing: if a race and a resulting deadlock is possible 1205 theoretically (no matter how unlikely the race scenario 1206 is), it will be proven so and will immediately be 1207 reported by the kernel (once the event is observed that 1208 makes the deadlock theoretically possible). 1209 1210 If a deadlock is impossible (i.e. the locking rules, as 1211 observed by the kernel, are mathematically correct), the 1212 kernel reports nothing. 1213 1214 NOTE: this feature can also be enabled for rwlocks, mutexes 1215 and rwsems - in which case all dependencies between these 1216 different locking variants are observed and mapped too, and 1217 the proof of observed correctness is also maintained for an 1218 arbitrary combination of these separate locking variants. 1219 1220 For more details, see Documentation/locking/lockdep-design.rst. 1221 1222config PROVE_RAW_LOCK_NESTING 1223 bool "Enable raw_spinlock - spinlock nesting checks" 1224 depends on PROVE_LOCKING 1225 default n 1226 help 1227 Enable the raw_spinlock vs. spinlock nesting checks which ensure 1228 that the lock nesting rules for PREEMPT_RT enabled kernels are 1229 not violated. 1230 1231 NOTE: There are known nesting problems. So if you enable this 1232 option expect lockdep splats until these problems have been fully 1233 addressed which is work in progress. This config switch allows to 1234 identify and analyze these problems. It will be removed and the 1235 check permanentely enabled once the main issues have been fixed. 1236 1237 If unsure, select N. 1238 1239config LOCK_STAT 1240 bool "Lock usage statistics" 1241 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1242 select LOCKDEP 1243 select DEBUG_SPINLOCK 1244 select DEBUG_MUTEXES 1245 select DEBUG_RT_MUTEXES if RT_MUTEXES 1246 select DEBUG_LOCK_ALLOC 1247 default n 1248 help 1249 This feature enables tracking lock contention points 1250 1251 For more details, see Documentation/locking/lockstat.rst 1252 1253 This also enables lock events required by "perf lock", 1254 subcommand of perf. 1255 If you want to use "perf lock", you also need to turn on 1256 CONFIG_EVENT_TRACING. 1257 1258 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events. 1259 (CONFIG_LOCKDEP defines "acquire" and "release" events.) 1260 1261config DEBUG_RT_MUTEXES 1262 bool "RT Mutex debugging, deadlock detection" 1263 depends on DEBUG_KERNEL && RT_MUTEXES 1264 help 1265 This allows rt mutex semantics violations and rt mutex related 1266 deadlocks (lockups) to be detected and reported automatically. 1267 1268config DEBUG_SPINLOCK 1269 bool "Spinlock and rw-lock debugging: basic checks" 1270 depends on DEBUG_KERNEL 1271 select UNINLINE_SPIN_UNLOCK 1272 help 1273 Say Y here and build SMP to catch missing spinlock initialization 1274 and certain other kinds of spinlock errors commonly made. This is 1275 best used in conjunction with the NMI watchdog so that spinlock 1276 deadlocks are also debuggable. 1277 1278config DEBUG_MUTEXES 1279 bool "Mutex debugging: basic checks" 1280 depends on DEBUG_KERNEL 1281 help 1282 This feature allows mutex semantics violations to be detected and 1283 reported. 1284 1285config DEBUG_WW_MUTEX_SLOWPATH 1286 bool "Wait/wound mutex debugging: Slowpath testing" 1287 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1288 select DEBUG_LOCK_ALLOC 1289 select DEBUG_SPINLOCK 1290 select DEBUG_MUTEXES 1291 help 1292 This feature enables slowpath testing for w/w mutex users by 1293 injecting additional -EDEADLK wound/backoff cases. Together with 1294 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this 1295 will test all possible w/w mutex interface abuse with the 1296 exception of simply not acquiring all the required locks. 1297 Note that this feature can introduce significant overhead, so 1298 it really should not be enabled in a production or distro kernel, 1299 even a debug kernel. If you are a driver writer, enable it. If 1300 you are a distro, do not. 1301 1302config DEBUG_RWSEMS 1303 bool "RW Semaphore debugging: basic checks" 1304 depends on DEBUG_KERNEL 1305 help 1306 This debugging feature allows mismatched rw semaphore locks 1307 and unlocks to be detected and reported. 1308 1309config DEBUG_LOCK_ALLOC 1310 bool "Lock debugging: detect incorrect freeing of live locks" 1311 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1312 select DEBUG_SPINLOCK 1313 select DEBUG_MUTEXES 1314 select DEBUG_RT_MUTEXES if RT_MUTEXES 1315 select LOCKDEP 1316 help 1317 This feature will check whether any held lock (spinlock, rwlock, 1318 mutex or rwsem) is incorrectly freed by the kernel, via any of the 1319 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(), 1320 vfree(), etc.), whether a live lock is incorrectly reinitialized via 1321 spin_lock_init()/mutex_init()/etc., or whether there is any lock 1322 held during task exit. 1323 1324config LOCKDEP 1325 bool 1326 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1327 select STACKTRACE 1328 select FRAME_POINTER if !MIPS && !PPC && !ARM && !S390 && !MICROBLAZE && !ARC && !X86 1329 select KALLSYMS 1330 select KALLSYMS_ALL 1331 1332config LOCKDEP_SMALL 1333 bool 1334 1335config DEBUG_LOCKDEP 1336 bool "Lock dependency engine debugging" 1337 depends on DEBUG_KERNEL && LOCKDEP 1338 help 1339 If you say Y here, the lock dependency engine will do 1340 additional runtime checks to debug itself, at the price 1341 of more runtime overhead. 1342 1343config DEBUG_ATOMIC_SLEEP 1344 bool "Sleep inside atomic section checking" 1345 select PREEMPT_COUNT 1346 depends on DEBUG_KERNEL 1347 depends on !ARCH_NO_PREEMPT 1348 help 1349 If you say Y here, various routines which may sleep will become very 1350 noisy if they are called inside atomic sections: when a spinlock is 1351 held, inside an rcu read side critical section, inside preempt disabled 1352 sections, inside an interrupt, etc... 1353 1354config DEBUG_LOCKING_API_SELFTESTS 1355 bool "Locking API boot-time self-tests" 1356 depends on DEBUG_KERNEL 1357 help 1358 Say Y here if you want the kernel to run a short self-test during 1359 bootup. The self-test checks whether common types of locking bugs 1360 are detected by debugging mechanisms or not. (if you disable 1361 lock debugging then those bugs wont be detected of course.) 1362 The following locking APIs are covered: spinlocks, rwlocks, 1363 mutexes and rwsems. 1364 1365config LOCK_TORTURE_TEST 1366 tristate "torture tests for locking" 1367 depends on DEBUG_KERNEL 1368 select TORTURE_TEST 1369 help 1370 This option provides a kernel module that runs torture tests 1371 on kernel locking primitives. The kernel module may be built 1372 after the fact on the running kernel to be tested, if desired. 1373 1374 Say Y here if you want kernel locking-primitive torture tests 1375 to be built into the kernel. 1376 Say M if you want these torture tests to build as a module. 1377 Say N if you are unsure. 1378 1379config WW_MUTEX_SELFTEST 1380 tristate "Wait/wound mutex selftests" 1381 help 1382 This option provides a kernel module that runs tests on the 1383 on the struct ww_mutex locking API. 1384 1385 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction 1386 with this test harness. 1387 1388 Say M if you want these self tests to build as a module. 1389 Say N if you are unsure. 1390 1391config SCF_TORTURE_TEST 1392 tristate "torture tests for smp_call_function*()" 1393 depends on DEBUG_KERNEL 1394 select TORTURE_TEST 1395 help 1396 This option provides a kernel module that runs torture tests 1397 on the smp_call_function() family of primitives. The kernel 1398 module may be built after the fact on the running kernel to 1399 be tested, if desired. 1400 1401config CSD_LOCK_WAIT_DEBUG 1402 bool "Debugging for csd_lock_wait(), called from smp_call_function*()" 1403 depends on DEBUG_KERNEL 1404 depends on 64BIT 1405 default n 1406 help 1407 This option enables debug prints when CPUs are slow to respond 1408 to the smp_call_function*() IPI wrappers. These debug prints 1409 include the IPI handler function currently executing (if any) 1410 and relevant stack traces. 1411 1412endmenu # lock debugging 1413 1414config TRACE_IRQFLAGS 1415 depends on TRACE_IRQFLAGS_SUPPORT 1416 bool 1417 help 1418 Enables hooks to interrupt enabling and disabling for 1419 either tracing or lock debugging. 1420 1421config TRACE_IRQFLAGS_NMI 1422 def_bool y 1423 depends on TRACE_IRQFLAGS 1424 depends on TRACE_IRQFLAGS_NMI_SUPPORT 1425 1426config STACKTRACE 1427 bool "Stack backtrace support" 1428 depends on STACKTRACE_SUPPORT 1429 help 1430 This option causes the kernel to create a /proc/pid/stack for 1431 every process, showing its current stack trace. 1432 It is also used by various kernel debugging features that require 1433 stack trace generation. 1434 1435config WARN_ALL_UNSEEDED_RANDOM 1436 bool "Warn for all uses of unseeded randomness" 1437 default n 1438 help 1439 Some parts of the kernel contain bugs relating to their use of 1440 cryptographically secure random numbers before it's actually possible 1441 to generate those numbers securely. This setting ensures that these 1442 flaws don't go unnoticed, by enabling a message, should this ever 1443 occur. This will allow people with obscure setups to know when things 1444 are going wrong, so that they might contact developers about fixing 1445 it. 1446 1447 Unfortunately, on some models of some architectures getting 1448 a fully seeded CRNG is extremely difficult, and so this can 1449 result in dmesg getting spammed for a surprisingly long 1450 time. This is really bad from a security perspective, and 1451 so architecture maintainers really need to do what they can 1452 to get the CRNG seeded sooner after the system is booted. 1453 However, since users cannot do anything actionable to 1454 address this, by default the kernel will issue only a single 1455 warning for the first use of unseeded randomness. 1456 1457 Say Y here if you want to receive warnings for all uses of 1458 unseeded randomness. This will be of use primarily for 1459 those developers interested in improving the security of 1460 Linux kernels running on their architecture (or 1461 subarchitecture). 1462 1463config DEBUG_KOBJECT 1464 bool "kobject debugging" 1465 depends on DEBUG_KERNEL 1466 help 1467 If you say Y here, some extra kobject debugging messages will be sent 1468 to the syslog. 1469 1470config DEBUG_KOBJECT_RELEASE 1471 bool "kobject release debugging" 1472 depends on DEBUG_OBJECTS_TIMERS 1473 help 1474 kobjects are reference counted objects. This means that their 1475 last reference count put is not predictable, and the kobject can 1476 live on past the point at which a driver decides to drop it's 1477 initial reference to the kobject gained on allocation. An 1478 example of this would be a struct device which has just been 1479 unregistered. 1480 1481 However, some buggy drivers assume that after such an operation, 1482 the memory backing the kobject can be immediately freed. This 1483 goes completely against the principles of a refcounted object. 1484 1485 If you say Y here, the kernel will delay the release of kobjects 1486 on the last reference count to improve the visibility of this 1487 kind of kobject release bug. 1488 1489config HAVE_DEBUG_BUGVERBOSE 1490 bool 1491 1492menu "Debug kernel data structures" 1493 1494config DEBUG_LIST 1495 bool "Debug linked list manipulation" 1496 depends on DEBUG_KERNEL || BUG_ON_DATA_CORRUPTION 1497 help 1498 Enable this to turn on extended checks in the linked-list 1499 walking routines. 1500 1501 If unsure, say N. 1502 1503config DEBUG_PLIST 1504 bool "Debug priority linked list manipulation" 1505 depends on DEBUG_KERNEL 1506 help 1507 Enable this to turn on extended checks in the priority-ordered 1508 linked-list (plist) walking routines. This checks the entire 1509 list multiple times during each manipulation. 1510 1511 If unsure, say N. 1512 1513config DEBUG_SG 1514 bool "Debug SG table operations" 1515 depends on DEBUG_KERNEL 1516 help 1517 Enable this to turn on checks on scatter-gather tables. This can 1518 help find problems with drivers that do not properly initialize 1519 their sg tables. 1520 1521 If unsure, say N. 1522 1523config DEBUG_NOTIFIERS 1524 bool "Debug notifier call chains" 1525 depends on DEBUG_KERNEL 1526 help 1527 Enable this to turn on sanity checking for notifier call chains. 1528 This is most useful for kernel developers to make sure that 1529 modules properly unregister themselves from notifier chains. 1530 This is a relatively cheap check but if you care about maximum 1531 performance, say N. 1532 1533config BUG_ON_DATA_CORRUPTION 1534 bool "Trigger a BUG when data corruption is detected" 1535 select DEBUG_LIST 1536 help 1537 Select this option if the kernel should BUG when it encounters 1538 data corruption in kernel memory structures when they get checked 1539 for validity. 1540 1541 If unsure, say N. 1542 1543endmenu 1544 1545config DEBUG_CREDENTIALS 1546 bool "Debug credential management" 1547 depends on DEBUG_KERNEL 1548 help 1549 Enable this to turn on some debug checking for credential 1550 management. The additional code keeps track of the number of 1551 pointers from task_structs to any given cred struct, and checks to 1552 see that this number never exceeds the usage count of the cred 1553 struct. 1554 1555 Furthermore, if SELinux is enabled, this also checks that the 1556 security pointer in the cred struct is never seen to be invalid. 1557 1558 If unsure, say N. 1559 1560source "kernel/rcu/Kconfig.debug" 1561 1562config DEBUG_WQ_FORCE_RR_CPU 1563 bool "Force round-robin CPU selection for unbound work items" 1564 depends on DEBUG_KERNEL 1565 default n 1566 help 1567 Workqueue used to implicitly guarantee that work items queued 1568 without explicit CPU specified are put on the local CPU. This 1569 guarantee is no longer true and while local CPU is still 1570 preferred work items may be put on foreign CPUs. Kernel 1571 parameter "workqueue.debug_force_rr_cpu" is added to force 1572 round-robin CPU selection to flush out usages which depend on the 1573 now broken guarantee. This config option enables the debug 1574 feature by default. When enabled, memory and cache locality will 1575 be impacted. 1576 1577config DEBUG_BLOCK_EXT_DEVT 1578 bool "Force extended block device numbers and spread them" 1579 depends on DEBUG_KERNEL 1580 depends on BLOCK 1581 default n 1582 help 1583 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON 1584 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT 1585 YOU ARE DOING. Distros, please enable this and fix whatever 1586 is broken. 1587 1588 Conventionally, block device numbers are allocated from 1589 predetermined contiguous area. However, extended block area 1590 may introduce non-contiguous block device numbers. This 1591 option forces most block device numbers to be allocated from 1592 the extended space and spreads them to discover kernel or 1593 userland code paths which assume predetermined contiguous 1594 device number allocation. 1595 1596 Note that turning on this debug option shuffles all the 1597 device numbers for all IDE and SCSI devices including libata 1598 ones, so root partition specified using device number 1599 directly (via rdev or root=MAJ:MIN) won't work anymore. 1600 Textual device names (root=/dev/sdXn) will continue to work. 1601 1602 Say N if you are unsure. 1603 1604config CPU_HOTPLUG_STATE_CONTROL 1605 bool "Enable CPU hotplug state control" 1606 depends on DEBUG_KERNEL 1607 depends on HOTPLUG_CPU 1608 default n 1609 help 1610 Allows to write steps between "offline" and "online" to the CPUs 1611 sysfs target file so states can be stepped granular. This is a debug 1612 option for now as the hotplug machinery cannot be stopped and 1613 restarted at arbitrary points yet. 1614 1615 Say N if your are unsure. 1616 1617config LATENCYTOP 1618 bool "Latency measuring infrastructure" 1619 depends on DEBUG_KERNEL 1620 depends on STACKTRACE_SUPPORT 1621 depends on PROC_FS 1622 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM && !ARC && !X86 1623 select KALLSYMS 1624 select KALLSYMS_ALL 1625 select STACKTRACE 1626 select SCHEDSTATS 1627 select SCHED_DEBUG 1628 help 1629 Enable this option if you want to use the LatencyTOP tool 1630 to find out which userspace is blocking on what kernel operations. 1631 1632source "kernel/trace/Kconfig" 1633 1634config PROVIDE_OHCI1394_DMA_INIT 1635 bool "Remote debugging over FireWire early on boot" 1636 depends on PCI && X86 1637 help 1638 If you want to debug problems which hang or crash the kernel early 1639 on boot and the crashing machine has a FireWire port, you can use 1640 this feature to remotely access the memory of the crashed machine 1641 over FireWire. This employs remote DMA as part of the OHCI1394 1642 specification which is now the standard for FireWire controllers. 1643 1644 With remote DMA, you can monitor the printk buffer remotely using 1645 firescope and access all memory below 4GB using fireproxy from gdb. 1646 Even controlling a kernel debugger is possible using remote DMA. 1647 1648 Usage: 1649 1650 If ohci1394_dma=early is used as boot parameter, it will initialize 1651 all OHCI1394 controllers which are found in the PCI config space. 1652 1653 As all changes to the FireWire bus such as enabling and disabling 1654 devices cause a bus reset and thereby disable remote DMA for all 1655 devices, be sure to have the cable plugged and FireWire enabled on 1656 the debugging host before booting the debug target for debugging. 1657 1658 This code (~1k) is freed after boot. By then, the firewire stack 1659 in charge of the OHCI-1394 controllers should be used instead. 1660 1661 See Documentation/core-api/debugging-via-ohci1394.rst for more information. 1662 1663source "samples/Kconfig" 1664 1665config ARCH_HAS_DEVMEM_IS_ALLOWED 1666 bool 1667 1668config STRICT_DEVMEM 1669 bool "Filter access to /dev/mem" 1670 depends on MMU && DEVMEM 1671 depends on ARCH_HAS_DEVMEM_IS_ALLOWED || GENERIC_LIB_DEVMEM_IS_ALLOWED 1672 default y if PPC || X86 || ARM64 1673 help 1674 If this option is disabled, you allow userspace (root) access to all 1675 of memory, including kernel and userspace memory. Accidental 1676 access to this is obviously disastrous, but specific access can 1677 be used by people debugging the kernel. Note that with PAT support 1678 enabled, even in this case there are restrictions on /dev/mem 1679 use due to the cache aliasing requirements. 1680 1681 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem 1682 file only allows userspace access to PCI space and the BIOS code and 1683 data regions. This is sufficient for dosemu and X and all common 1684 users of /dev/mem. 1685 1686 If in doubt, say Y. 1687 1688config IO_STRICT_DEVMEM 1689 bool "Filter I/O access to /dev/mem" 1690 depends on STRICT_DEVMEM 1691 help 1692 If this option is disabled, you allow userspace (root) access to all 1693 io-memory regardless of whether a driver is actively using that 1694 range. Accidental access to this is obviously disastrous, but 1695 specific access can be used by people debugging kernel drivers. 1696 1697 If this option is switched on, the /dev/mem file only allows 1698 userspace access to *idle* io-memory ranges (see /proc/iomem) This 1699 may break traditional users of /dev/mem (dosemu, legacy X, etc...) 1700 if the driver using a given range cannot be disabled. 1701 1702 If in doubt, say Y. 1703 1704menu "$(SRCARCH) Debugging" 1705 1706source "arch/$(SRCARCH)/Kconfig.debug" 1707 1708endmenu 1709 1710menu "Kernel Testing and Coverage" 1711 1712source "lib/kunit/Kconfig" 1713 1714config NOTIFIER_ERROR_INJECTION 1715 tristate "Notifier error injection" 1716 depends on DEBUG_KERNEL 1717 select DEBUG_FS 1718 help 1719 This option provides the ability to inject artificial errors to 1720 specified notifier chain callbacks. It is useful to test the error 1721 handling of notifier call chain failures. 1722 1723 Say N if unsure. 1724 1725config PM_NOTIFIER_ERROR_INJECT 1726 tristate "PM notifier error injection module" 1727 depends on PM && NOTIFIER_ERROR_INJECTION 1728 default m if PM_DEBUG 1729 help 1730 This option provides the ability to inject artificial errors to 1731 PM notifier chain callbacks. It is controlled through debugfs 1732 interface /sys/kernel/debug/notifier-error-inject/pm 1733 1734 If the notifier call chain should be failed with some events 1735 notified, write the error code to "actions/<notifier event>/error". 1736 1737 Example: Inject PM suspend error (-12 = -ENOMEM) 1738 1739 # cd /sys/kernel/debug/notifier-error-inject/pm/ 1740 # echo -12 > actions/PM_SUSPEND_PREPARE/error 1741 # echo mem > /sys/power/state 1742 bash: echo: write error: Cannot allocate memory 1743 1744 To compile this code as a module, choose M here: the module will 1745 be called pm-notifier-error-inject. 1746 1747 If unsure, say N. 1748 1749config OF_RECONFIG_NOTIFIER_ERROR_INJECT 1750 tristate "OF reconfig notifier error injection module" 1751 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION 1752 help 1753 This option provides the ability to inject artificial errors to 1754 OF reconfig notifier chain callbacks. It is controlled 1755 through debugfs interface under 1756 /sys/kernel/debug/notifier-error-inject/OF-reconfig/ 1757 1758 If the notifier call chain should be failed with some events 1759 notified, write the error code to "actions/<notifier event>/error". 1760 1761 To compile this code as a module, choose M here: the module will 1762 be called of-reconfig-notifier-error-inject. 1763 1764 If unsure, say N. 1765 1766config NETDEV_NOTIFIER_ERROR_INJECT 1767 tristate "Netdev notifier error injection module" 1768 depends on NET && NOTIFIER_ERROR_INJECTION 1769 help 1770 This option provides the ability to inject artificial errors to 1771 netdevice notifier chain callbacks. It is controlled through debugfs 1772 interface /sys/kernel/debug/notifier-error-inject/netdev 1773 1774 If the notifier call chain should be failed with some events 1775 notified, write the error code to "actions/<notifier event>/error". 1776 1777 Example: Inject netdevice mtu change error (-22 = -EINVAL) 1778 1779 # cd /sys/kernel/debug/notifier-error-inject/netdev 1780 # echo -22 > actions/NETDEV_CHANGEMTU/error 1781 # ip link set eth0 mtu 1024 1782 RTNETLINK answers: Invalid argument 1783 1784 To compile this code as a module, choose M here: the module will 1785 be called netdev-notifier-error-inject. 1786 1787 If unsure, say N. 1788 1789config FUNCTION_ERROR_INJECTION 1790 def_bool y 1791 depends on HAVE_FUNCTION_ERROR_INJECTION && KPROBES 1792 1793config FAULT_INJECTION 1794 bool "Fault-injection framework" 1795 depends on DEBUG_KERNEL 1796 help 1797 Provide fault-injection framework. 1798 For more details, see Documentation/fault-injection/. 1799 1800config FAILSLAB 1801 bool "Fault-injection capability for kmalloc" 1802 depends on FAULT_INJECTION 1803 depends on SLAB || SLUB 1804 help 1805 Provide fault-injection capability for kmalloc. 1806 1807config FAIL_PAGE_ALLOC 1808 bool "Fault-injection capability for alloc_pages()" 1809 depends on FAULT_INJECTION 1810 help 1811 Provide fault-injection capability for alloc_pages(). 1812 1813config FAULT_INJECTION_USERCOPY 1814 bool "Fault injection capability for usercopy functions" 1815 depends on FAULT_INJECTION 1816 help 1817 Provides fault-injection capability to inject failures 1818 in usercopy functions (copy_from_user(), get_user(), ...). 1819 1820config FAIL_MAKE_REQUEST 1821 bool "Fault-injection capability for disk IO" 1822 depends on FAULT_INJECTION && BLOCK 1823 help 1824 Provide fault-injection capability for disk IO. 1825 1826config FAIL_IO_TIMEOUT 1827 bool "Fault-injection capability for faking disk interrupts" 1828 depends on FAULT_INJECTION && BLOCK 1829 help 1830 Provide fault-injection capability on end IO handling. This 1831 will make the block layer "forget" an interrupt as configured, 1832 thus exercising the error handling. 1833 1834 Only works with drivers that use the generic timeout handling, 1835 for others it wont do anything. 1836 1837config FAIL_FUTEX 1838 bool "Fault-injection capability for futexes" 1839 select DEBUG_FS 1840 depends on FAULT_INJECTION && FUTEX 1841 help 1842 Provide fault-injection capability for futexes. 1843 1844config FAULT_INJECTION_DEBUG_FS 1845 bool "Debugfs entries for fault-injection capabilities" 1846 depends on FAULT_INJECTION && SYSFS && DEBUG_FS 1847 help 1848 Enable configuration of fault-injection capabilities via debugfs. 1849 1850config FAIL_FUNCTION 1851 bool "Fault-injection capability for functions" 1852 depends on FAULT_INJECTION_DEBUG_FS && FUNCTION_ERROR_INJECTION 1853 help 1854 Provide function-based fault-injection capability. 1855 This will allow you to override a specific function with a return 1856 with given return value. As a result, function caller will see 1857 an error value and have to handle it. This is useful to test the 1858 error handling in various subsystems. 1859 1860config FAIL_MMC_REQUEST 1861 bool "Fault-injection capability for MMC IO" 1862 depends on FAULT_INJECTION_DEBUG_FS && MMC 1863 help 1864 Provide fault-injection capability for MMC IO. 1865 This will make the mmc core return data errors. This is 1866 useful to test the error handling in the mmc block device 1867 and to test how the mmc host driver handles retries from 1868 the block device. 1869 1870config FAULT_INJECTION_STACKTRACE_FILTER 1871 bool "stacktrace filter for fault-injection capabilities" 1872 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT 1873 depends on !X86_64 1874 select STACKTRACE 1875 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM && !ARC && !X86 1876 help 1877 Provide stacktrace filter for fault-injection capabilities 1878 1879config ARCH_HAS_KCOV 1880 bool 1881 help 1882 An architecture should select this when it can successfully 1883 build and run with CONFIG_KCOV. This typically requires 1884 disabling instrumentation for some early boot code. 1885 1886config CC_HAS_SANCOV_TRACE_PC 1887 def_bool $(cc-option,-fsanitize-coverage=trace-pc) 1888 1889 1890config KCOV 1891 bool "Code coverage for fuzzing" 1892 depends on ARCH_HAS_KCOV 1893 depends on CC_HAS_SANCOV_TRACE_PC || GCC_PLUGINS 1894 select DEBUG_FS 1895 select GCC_PLUGIN_SANCOV if !CC_HAS_SANCOV_TRACE_PC 1896 help 1897 KCOV exposes kernel code coverage information in a form suitable 1898 for coverage-guided fuzzing (randomized testing). 1899 1900 If RANDOMIZE_BASE is enabled, PC values will not be stable across 1901 different machines and across reboots. If you need stable PC values, 1902 disable RANDOMIZE_BASE. 1903 1904 For more details, see Documentation/dev-tools/kcov.rst. 1905 1906config KCOV_ENABLE_COMPARISONS 1907 bool "Enable comparison operands collection by KCOV" 1908 depends on KCOV 1909 depends on $(cc-option,-fsanitize-coverage=trace-cmp) 1910 help 1911 KCOV also exposes operands of every comparison in the instrumented 1912 code along with operand sizes and PCs of the comparison instructions. 1913 These operands can be used by fuzzing engines to improve the quality 1914 of fuzzing coverage. 1915 1916config KCOV_INSTRUMENT_ALL 1917 bool "Instrument all code by default" 1918 depends on KCOV 1919 default y 1920 help 1921 If you are doing generic system call fuzzing (like e.g. syzkaller), 1922 then you will want to instrument the whole kernel and you should 1923 say y here. If you are doing more targeted fuzzing (like e.g. 1924 filesystem fuzzing with AFL) then you will want to enable coverage 1925 for more specific subsets of files, and should say n here. 1926 1927config KCOV_IRQ_AREA_SIZE 1928 hex "Size of interrupt coverage collection area in words" 1929 depends on KCOV 1930 default 0x40000 1931 help 1932 KCOV uses preallocated per-cpu areas to collect coverage from 1933 soft interrupts. This specifies the size of those areas in the 1934 number of unsigned long words. 1935 1936menuconfig RUNTIME_TESTING_MENU 1937 bool "Runtime Testing" 1938 def_bool y 1939 1940if RUNTIME_TESTING_MENU 1941 1942config LKDTM 1943 tristate "Linux Kernel Dump Test Tool Module" 1944 depends on DEBUG_FS 1945 help 1946 This module enables testing of the different dumping mechanisms by 1947 inducing system failures at predefined crash points. 1948 If you don't need it: say N 1949 Choose M here to compile this code as a module. The module will be 1950 called lkdtm. 1951 1952 Documentation on how to use the module can be found in 1953 Documentation/fault-injection/provoke-crashes.rst 1954 1955config TEST_LIST_SORT 1956 tristate "Linked list sorting test" 1957 depends on DEBUG_KERNEL || m 1958 help 1959 Enable this to turn on 'list_sort()' function test. This test is 1960 executed only once during system boot (so affects only boot time), 1961 or at module load time. 1962 1963 If unsure, say N. 1964 1965config TEST_MIN_HEAP 1966 tristate "Min heap test" 1967 depends on DEBUG_KERNEL || m 1968 help 1969 Enable this to turn on min heap function tests. This test is 1970 executed only once during system boot (so affects only boot time), 1971 or at module load time. 1972 1973 If unsure, say N. 1974 1975config TEST_SORT 1976 tristate "Array-based sort test" 1977 depends on DEBUG_KERNEL || m 1978 help 1979 This option enables the self-test function of 'sort()' at boot, 1980 or at module load time. 1981 1982 If unsure, say N. 1983 1984config KPROBES_SANITY_TEST 1985 bool "Kprobes sanity tests" 1986 depends on DEBUG_KERNEL 1987 depends on KPROBES 1988 help 1989 This option provides for testing basic kprobes functionality on 1990 boot. Samples of kprobe and kretprobe are inserted and 1991 verified for functionality. 1992 1993 Say N if you are unsure. 1994 1995config BACKTRACE_SELF_TEST 1996 tristate "Self test for the backtrace code" 1997 depends on DEBUG_KERNEL 1998 help 1999 This option provides a kernel module that can be used to test 2000 the kernel stack backtrace code. This option is not useful 2001 for distributions or general kernels, but only for kernel 2002 developers working on architecture code. 2003 2004 Note that if you want to also test saved backtraces, you will 2005 have to enable STACKTRACE as well. 2006 2007 Say N if you are unsure. 2008 2009config RBTREE_TEST 2010 tristate "Red-Black tree test" 2011 depends on DEBUG_KERNEL 2012 help 2013 A benchmark measuring the performance of the rbtree library. 2014 Also includes rbtree invariant checks. 2015 2016config REED_SOLOMON_TEST 2017 tristate "Reed-Solomon library test" 2018 depends on DEBUG_KERNEL || m 2019 select REED_SOLOMON 2020 select REED_SOLOMON_ENC16 2021 select REED_SOLOMON_DEC16 2022 help 2023 This option enables the self-test function of rslib at boot, 2024 or at module load time. 2025 2026 If unsure, say N. 2027 2028config INTERVAL_TREE_TEST 2029 tristate "Interval tree test" 2030 depends on DEBUG_KERNEL 2031 select INTERVAL_TREE 2032 help 2033 A benchmark measuring the performance of the interval tree library 2034 2035config PERCPU_TEST 2036 tristate "Per cpu operations test" 2037 depends on m && DEBUG_KERNEL 2038 help 2039 Enable this option to build test module which validates per-cpu 2040 operations. 2041 2042 If unsure, say N. 2043 2044config ATOMIC64_SELFTEST 2045 tristate "Perform an atomic64_t self-test" 2046 help 2047 Enable this option to test the atomic64_t functions at boot or 2048 at module load time. 2049 2050 If unsure, say N. 2051 2052config ASYNC_RAID6_TEST 2053 tristate "Self test for hardware accelerated raid6 recovery" 2054 depends on ASYNC_RAID6_RECOV 2055 select ASYNC_MEMCPY 2056 help 2057 This is a one-shot self test that permutes through the 2058 recovery of all the possible two disk failure scenarios for a 2059 N-disk array. Recovery is performed with the asynchronous 2060 raid6 recovery routines, and will optionally use an offload 2061 engine if one is available. 2062 2063 If unsure, say N. 2064 2065config TEST_HEXDUMP 2066 tristate "Test functions located in the hexdump module at runtime" 2067 2068config TEST_STRING_HELPERS 2069 tristate "Test functions located in the string_helpers module at runtime" 2070 2071config TEST_STRSCPY 2072 tristate "Test strscpy*() family of functions at runtime" 2073 2074config TEST_KSTRTOX 2075 tristate "Test kstrto*() family of functions at runtime" 2076 2077config TEST_PRINTF 2078 tristate "Test printf() family of functions at runtime" 2079 2080config TEST_BITMAP 2081 tristate "Test bitmap_*() family of functions at runtime" 2082 help 2083 Enable this option to test the bitmap functions at boot. 2084 2085 If unsure, say N. 2086 2087config TEST_UUID 2088 tristate "Test functions located in the uuid module at runtime" 2089 2090config TEST_XARRAY 2091 tristate "Test the XArray code at runtime" 2092 2093config TEST_OVERFLOW 2094 tristate "Test check_*_overflow() functions at runtime" 2095 2096config TEST_RHASHTABLE 2097 tristate "Perform selftest on resizable hash table" 2098 help 2099 Enable this option to test the rhashtable functions at boot. 2100 2101 If unsure, say N. 2102 2103config TEST_HASH 2104 tristate "Perform selftest on hash functions" 2105 help 2106 Enable this option to test the kernel's integer (<linux/hash.h>), 2107 string (<linux/stringhash.h>), and siphash (<linux/siphash.h>) 2108 hash functions on boot (or module load). 2109 2110 This is intended to help people writing architecture-specific 2111 optimized versions. If unsure, say N. 2112 2113config TEST_IDA 2114 tristate "Perform selftest on IDA functions" 2115 2116config TEST_PARMAN 2117 tristate "Perform selftest on priority array manager" 2118 depends on PARMAN 2119 help 2120 Enable this option to test priority array manager on boot 2121 (or module load). 2122 2123 If unsure, say N. 2124 2125config TEST_IRQ_TIMINGS 2126 bool "IRQ timings selftest" 2127 depends on IRQ_TIMINGS 2128 help 2129 Enable this option to test the irq timings code on boot. 2130 2131 If unsure, say N. 2132 2133config TEST_LKM 2134 tristate "Test module loading with 'hello world' module" 2135 depends on m 2136 help 2137 This builds the "test_module" module that emits "Hello, world" 2138 on printk when loaded. It is designed to be used for basic 2139 evaluation of the module loading subsystem (for example when 2140 validating module verification). It lacks any extra dependencies, 2141 and will not normally be loaded by the system unless explicitly 2142 requested by name. 2143 2144 If unsure, say N. 2145 2146config TEST_BITOPS 2147 tristate "Test module for compilation of bitops operations" 2148 depends on m 2149 help 2150 This builds the "test_bitops" module that is much like the 2151 TEST_LKM module except that it does a basic exercise of the 2152 set/clear_bit macros and get_count_order/long to make sure there are 2153 no compiler warnings from C=1 sparse checker or -Wextra 2154 compilations. It has no dependencies and doesn't run or load unless 2155 explicitly requested by name. for example: modprobe test_bitops. 2156 2157 If unsure, say N. 2158 2159config TEST_VMALLOC 2160 tristate "Test module for stress/performance analysis of vmalloc allocator" 2161 default n 2162 depends on MMU 2163 depends on m 2164 help 2165 This builds the "test_vmalloc" module that should be used for 2166 stress and performance analysis. So, any new change for vmalloc 2167 subsystem can be evaluated from performance and stability point 2168 of view. 2169 2170 If unsure, say N. 2171 2172config TEST_USER_COPY 2173 tristate "Test user/kernel boundary protections" 2174 depends on m 2175 help 2176 This builds the "test_user_copy" module that runs sanity checks 2177 on the copy_to/from_user infrastructure, making sure basic 2178 user/kernel boundary testing is working. If it fails to load, 2179 a regression has been detected in the user/kernel memory boundary 2180 protections. 2181 2182 If unsure, say N. 2183 2184config TEST_BPF 2185 tristate "Test BPF filter functionality" 2186 depends on m && NET 2187 help 2188 This builds the "test_bpf" module that runs various test vectors 2189 against the BPF interpreter or BPF JIT compiler depending on the 2190 current setting. This is in particular useful for BPF JIT compiler 2191 development, but also to run regression tests against changes in 2192 the interpreter code. It also enables test stubs for eBPF maps and 2193 verifier used by user space verifier testsuite. 2194 2195 If unsure, say N. 2196 2197config TEST_BLACKHOLE_DEV 2198 tristate "Test blackhole netdev functionality" 2199 depends on m && NET 2200 help 2201 This builds the "test_blackhole_dev" module that validates the 2202 data path through this blackhole netdev. 2203 2204 If unsure, say N. 2205 2206config FIND_BIT_BENCHMARK 2207 tristate "Test find_bit functions" 2208 help 2209 This builds the "test_find_bit" module that measure find_*_bit() 2210 functions performance. 2211 2212 If unsure, say N. 2213 2214config TEST_FIRMWARE 2215 tristate "Test firmware loading via userspace interface" 2216 depends on FW_LOADER 2217 help 2218 This builds the "test_firmware" module that creates a userspace 2219 interface for testing firmware loading. This can be used to 2220 control the triggering of firmware loading without needing an 2221 actual firmware-using device. The contents can be rechecked by 2222 userspace. 2223 2224 If unsure, say N. 2225 2226config TEST_SYSCTL 2227 tristate "sysctl test driver" 2228 depends on PROC_SYSCTL 2229 help 2230 This builds the "test_sysctl" module. This driver enables to test the 2231 proc sysctl interfaces available to drivers safely without affecting 2232 production knobs which might alter system functionality. 2233 2234 If unsure, say N. 2235 2236config BITFIELD_KUNIT 2237 tristate "KUnit test bitfield functions at runtime" 2238 depends on KUNIT 2239 help 2240 Enable this option to test the bitfield functions at boot. 2241 2242 KUnit tests run during boot and output the results to the debug log 2243 in TAP format (http://testanything.org/). Only useful for kernel devs 2244 running the KUnit test harness, and not intended for inclusion into a 2245 production build. 2246 2247 For more information on KUnit and unit tests in general please refer 2248 to the KUnit documentation in Documentation/dev-tools/kunit/. 2249 2250 If unsure, say N. 2251 2252config RESOURCE_KUNIT_TEST 2253 tristate "KUnit test for resource API" 2254 depends on KUNIT 2255 help 2256 This builds the resource API unit test. 2257 Tests the logic of API provided by resource.c and ioport.h. 2258 For more information on KUnit and unit tests in general please refer 2259 to the KUnit documentation in Documentation/dev-tools/kunit/. 2260 2261 If unsure, say N. 2262 2263config SYSCTL_KUNIT_TEST 2264 tristate "KUnit test for sysctl" if !KUNIT_ALL_TESTS 2265 depends on KUNIT 2266 default KUNIT_ALL_TESTS 2267 help 2268 This builds the proc sysctl unit test, which runs on boot. 2269 Tests the API contract and implementation correctness of sysctl. 2270 For more information on KUnit and unit tests in general please refer 2271 to the KUnit documentation in Documentation/dev-tools/kunit/. 2272 2273 If unsure, say N. 2274 2275config LIST_KUNIT_TEST 2276 tristate "KUnit Test for Kernel Linked-list structures" if !KUNIT_ALL_TESTS 2277 depends on KUNIT 2278 default KUNIT_ALL_TESTS 2279 help 2280 This builds the linked list KUnit test suite. 2281 It tests that the API and basic functionality of the list_head type 2282 and associated macros. 2283 2284 KUnit tests run during boot and output the results to the debug log 2285 in TAP format (https://testanything.org/). Only useful for kernel devs 2286 running the KUnit test harness, and not intended for inclusion into a 2287 production build. 2288 2289 For more information on KUnit and unit tests in general please refer 2290 to the KUnit documentation in Documentation/dev-tools/kunit/. 2291 2292 If unsure, say N. 2293 2294config LINEAR_RANGES_TEST 2295 tristate "KUnit test for linear_ranges" 2296 depends on KUNIT 2297 select LINEAR_RANGES 2298 help 2299 This builds the linear_ranges unit test, which runs on boot. 2300 Tests the linear_ranges logic correctness. 2301 For more information on KUnit and unit tests in general please refer 2302 to the KUnit documentation in Documentation/dev-tools/kunit/. 2303 2304 If unsure, say N. 2305 2306config CMDLINE_KUNIT_TEST 2307 tristate "KUnit test for cmdline API" 2308 depends on KUNIT 2309 help 2310 This builds the cmdline API unit test. 2311 Tests the logic of API provided by cmdline.c. 2312 For more information on KUnit and unit tests in general please refer 2313 to the KUnit documentation in Documentation/dev-tools/kunit/. 2314 2315 If unsure, say N. 2316 2317config BITS_TEST 2318 tristate "KUnit test for bits.h" 2319 depends on KUNIT 2320 help 2321 This builds the bits unit test. 2322 Tests the logic of macros defined in bits.h. 2323 For more information on KUnit and unit tests in general please refer 2324 to the KUnit documentation in Documentation/dev-tools/kunit/. 2325 2326 If unsure, say N. 2327 2328config TEST_UDELAY 2329 tristate "udelay test driver" 2330 help 2331 This builds the "udelay_test" module that helps to make sure 2332 that udelay() is working properly. 2333 2334 If unsure, say N. 2335 2336config TEST_STATIC_KEYS 2337 tristate "Test static keys" 2338 depends on m 2339 help 2340 Test the static key interfaces. 2341 2342 If unsure, say N. 2343 2344config TEST_KMOD 2345 tristate "kmod stress tester" 2346 depends on m 2347 depends on NETDEVICES && NET_CORE && INET # for TUN 2348 depends on BLOCK 2349 select TEST_LKM 2350 select XFS_FS 2351 select TUN 2352 select BTRFS_FS 2353 help 2354 Test the kernel's module loading mechanism: kmod. kmod implements 2355 support to load modules using the Linux kernel's usermode helper. 2356 This test provides a series of tests against kmod. 2357 2358 Although technically you can either build test_kmod as a module or 2359 into the kernel we disallow building it into the kernel since 2360 it stress tests request_module() and this will very likely cause 2361 some issues by taking over precious threads available from other 2362 module load requests, ultimately this could be fatal. 2363 2364 To run tests run: 2365 2366 tools/testing/selftests/kmod/kmod.sh --help 2367 2368 If unsure, say N. 2369 2370config TEST_DEBUG_VIRTUAL 2371 tristate "Test CONFIG_DEBUG_VIRTUAL feature" 2372 depends on DEBUG_VIRTUAL 2373 help 2374 Test the kernel's ability to detect incorrect calls to 2375 virt_to_phys() done against the non-linear part of the 2376 kernel's virtual address map. 2377 2378 If unsure, say N. 2379 2380config TEST_MEMCAT_P 2381 tristate "Test memcat_p() helper function" 2382 help 2383 Test the memcat_p() helper for correctly merging two 2384 pointer arrays together. 2385 2386 If unsure, say N. 2387 2388config TEST_LIVEPATCH 2389 tristate "Test livepatching" 2390 default n 2391 depends on DYNAMIC_DEBUG 2392 depends on LIVEPATCH 2393 depends on m 2394 help 2395 Test kernel livepatching features for correctness. The tests will 2396 load test modules that will be livepatched in various scenarios. 2397 2398 To run all the livepatching tests: 2399 2400 make -C tools/testing/selftests TARGETS=livepatch run_tests 2401 2402 Alternatively, individual tests may be invoked: 2403 2404 tools/testing/selftests/livepatch/test-callbacks.sh 2405 tools/testing/selftests/livepatch/test-livepatch.sh 2406 tools/testing/selftests/livepatch/test-shadow-vars.sh 2407 2408 If unsure, say N. 2409 2410config TEST_OBJAGG 2411 tristate "Perform selftest on object aggreration manager" 2412 default n 2413 depends on OBJAGG 2414 help 2415 Enable this option to test object aggregation manager on boot 2416 (or module load). 2417 2418 2419config TEST_STACKINIT 2420 tristate "Test level of stack variable initialization" 2421 help 2422 Test if the kernel is zero-initializing stack variables and 2423 padding. Coverage is controlled by compiler flags, 2424 CONFIG_GCC_PLUGIN_STRUCTLEAK, CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF, 2425 or CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL. 2426 2427 If unsure, say N. 2428 2429config TEST_MEMINIT 2430 tristate "Test heap/page initialization" 2431 help 2432 Test if the kernel is zero-initializing heap and page allocations. 2433 This can be useful to test init_on_alloc and init_on_free features. 2434 2435 If unsure, say N. 2436 2437config TEST_HMM 2438 tristate "Test HMM (Heterogeneous Memory Management)" 2439 depends on TRANSPARENT_HUGEPAGE 2440 depends on DEVICE_PRIVATE 2441 select HMM_MIRROR 2442 select MMU_NOTIFIER 2443 help 2444 This is a pseudo device driver solely for testing HMM. 2445 Say M here if you want to build the HMM test module. 2446 Doing so will allow you to run tools/testing/selftest/vm/hmm-tests. 2447 2448 If unsure, say N. 2449 2450config TEST_FREE_PAGES 2451 tristate "Test freeing pages" 2452 help 2453 Test that a memory leak does not occur due to a race between 2454 freeing a block of pages and a speculative page reference. 2455 Loading this module is safe if your kernel has the bug fixed. 2456 If the bug is not fixed, it will leak gigabytes of memory and 2457 probably OOM your system. 2458 2459config TEST_FPU 2460 tristate "Test floating point operations in kernel space" 2461 depends on X86 && !KCOV_INSTRUMENT_ALL 2462 help 2463 Enable this option to add /sys/kernel/debug/selftest_helpers/test_fpu 2464 which will trigger a sequence of floating point operations. This is used 2465 for self-testing floating point control register setting in 2466 kernel_fpu_begin(). 2467 2468 If unsure, say N. 2469 2470endif # RUNTIME_TESTING_MENU 2471 2472config MEMTEST 2473 bool "Memtest" 2474 help 2475 This option adds a kernel parameter 'memtest', which allows memtest 2476 to be set. 2477 memtest=0, mean disabled; -- default 2478 memtest=1, mean do 1 test pattern; 2479 ... 2480 memtest=17, mean do 17 test patterns. 2481 If you are unsure how to answer this question, answer N. 2482 2483 2484 2485config HYPERV_TESTING 2486 bool "Microsoft Hyper-V driver testing" 2487 default n 2488 depends on HYPERV && DEBUG_FS 2489 help 2490 Select this option to enable Hyper-V vmbus testing. 2491 2492endmenu # "Kernel Testing and Coverage" 2493 2494source "Documentation/Kconfig" 2495 2496endmenu # Kernel hacking 2497