1 # SPDX-License-Identifier: GPL-2.0-only
4 menu "printk and dmesg options"
7 bool "Show timing information on printks"
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.
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.
18 The behavior is also controlled by the kernel command line
19 parameter printk.time=1. See Documentation/admin-guide/kernel-parameters.rst
22 bool "Show caller information on printks"
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)
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.
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
38 config STACKTRACE_BUILD_ID
39 bool "Show build ID information in stacktraces"
42 Selecting this option adds build ID information for symbols in
43 stacktraces printed with the printk format '%p[SR]b'.
45 This option is intended for distros where debuginfo is not easily
46 accessible but can be downloaded given the build ID of the vmlinux or
47 kernel module where the function is located.
49 config CONSOLE_LOGLEVEL_DEFAULT
50 int "Default console loglevel (1-15)"
54 Default loglevel to determine what will be printed on the console.
56 Setting a default here is equivalent to passing in loglevel=<x> in
57 the kernel bootargs. loglevel=<x> continues to override whatever
58 value is specified here as well.
60 Note: This does not affect the log level of un-prefixed printk()
61 usage in the kernel. That is controlled by the MESSAGE_LOGLEVEL_DEFAULT
64 config CONSOLE_LOGLEVEL_QUIET
65 int "quiet console loglevel (1-15)"
69 loglevel to use when "quiet" is passed on the kernel commandline.
71 When "quiet" is passed on the kernel commandline this loglevel
72 will be used as the loglevel. IOW passing "quiet" will be the
73 equivalent of passing "loglevel=<CONSOLE_LOGLEVEL_QUIET>"
75 config MESSAGE_LOGLEVEL_DEFAULT
76 int "Default message log level (1-7)"
80 Default log level for printk statements with no specified priority.
82 This was hard-coded to KERN_WARNING since at least 2.6.10 but folks
83 that are auditing their logs closely may want to set it to a lower
86 Note: This does not affect what message level gets printed on the console
87 by default. To change that, use loglevel=<x> in the kernel bootargs,
88 or pick a different CONSOLE_LOGLEVEL_DEFAULT configuration value.
90 config BOOT_PRINTK_DELAY
91 bool "Delay each boot printk message by N milliseconds"
92 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY
94 This build option allows you to read kernel boot messages
95 by inserting a short delay after each one. The delay is
96 specified in milliseconds on the kernel command line,
99 It is likely that you would also need to use "lpj=M" to preset
100 the "loops per jiffie" value.
101 See a previous boot log for the "lpj" value to use for your
102 system, and then set "lpj=M" before setting "boot_delay=N".
103 NOTE: Using this option may adversely affect SMP systems.
104 I.e., processors other than the first one may not boot up.
105 BOOT_PRINTK_DELAY also may cause LOCKUP_DETECTOR to detect
106 what it believes to be lockup conditions.
109 bool "Enable dynamic printk() support"
112 depends on (DEBUG_FS || PROC_FS)
113 select DYNAMIC_DEBUG_CORE
116 Compiles debug level messages into the kernel, which would not
117 otherwise be available at runtime. These messages can then be
118 enabled/disabled based on various levels of scope - per source file,
119 function, module, format string, and line number. This mechanism
120 implicitly compiles in all pr_debug() and dev_dbg() calls, which
121 enlarges the kernel text size by about 2%.
123 If a source file is compiled with DEBUG flag set, any
124 pr_debug() calls in it are enabled by default, but can be
125 disabled at runtime as below. Note that DEBUG flag is
126 turned on by many CONFIG_*DEBUG* options.
130 Dynamic debugging is controlled via the 'dynamic_debug/control' file,
131 which is contained in the 'debugfs' filesystem or procfs.
132 Thus, the debugfs or procfs filesystem must first be mounted before
133 making use of this feature.
134 We refer the control file as: <debugfs>/dynamic_debug/control. This
135 file contains a list of the debug statements that can be enabled. The
136 format for each line of the file is:
138 filename:lineno [module]function flags format
140 filename : source file of the debug statement
141 lineno : line number of the debug statement
142 module : module that contains the debug statement
143 function : function that contains the debug statement
144 flags : '=p' means the line is turned 'on' for printing
145 format : the format used for the debug statement
149 nullarbor:~ # cat <debugfs>/dynamic_debug/control
150 # filename:lineno [module]function flags format
151 fs/aio.c:222 [aio]__put_ioctx =_ "__put_ioctx:\040freeing\040%p\012"
152 fs/aio.c:248 [aio]ioctx_alloc =_ "ENOMEM:\040nr_events\040too\040high\012"
153 fs/aio.c:1770 [aio]sys_io_cancel =_ "calling\040cancel\012"
157 // enable the message at line 1603 of file svcsock.c
158 nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' >
159 <debugfs>/dynamic_debug/control
161 // enable all the messages in file svcsock.c
162 nullarbor:~ # echo -n 'file svcsock.c +p' >
163 <debugfs>/dynamic_debug/control
165 // enable all the messages in the NFS server module
166 nullarbor:~ # echo -n 'module nfsd +p' >
167 <debugfs>/dynamic_debug/control
169 // enable all 12 messages in the function svc_process()
170 nullarbor:~ # echo -n 'func svc_process +p' >
171 <debugfs>/dynamic_debug/control
173 // disable all 12 messages in the function svc_process()
174 nullarbor:~ # echo -n 'func svc_process -p' >
175 <debugfs>/dynamic_debug/control
177 See Documentation/admin-guide/dynamic-debug-howto.rst for additional
180 config DYNAMIC_DEBUG_CORE
181 bool "Enable core function of dynamic debug support"
183 depends on (DEBUG_FS || PROC_FS)
185 Enable core functional support of dynamic debug. It is useful
186 when you want to tie dynamic debug to your kernel modules with
187 DYNAMIC_DEBUG_MODULE defined for each of them, especially for
188 the case of embedded system where the kernel image size is
189 sensitive for people.
191 config SYMBOLIC_ERRNAME
192 bool "Support symbolic error names in printf"
195 If you say Y here, the kernel's printf implementation will
196 be able to print symbolic error names such as ENOSPC instead
197 of the number 28. It makes the kernel image slightly larger
198 (about 3KB), but can make the kernel logs easier to read.
200 config DEBUG_BUGVERBOSE
201 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT
202 depends on BUG && (GENERIC_BUG || HAVE_DEBUG_BUGVERBOSE)
205 Say Y here to make BUG() panics output the file name and line number
206 of the BUG call as well as the EIP and oops trace. This aids
207 debugging but costs about 70-100K of memory.
209 endmenu # "printk and dmesg options"
212 bool "Kernel debugging"
214 Say Y here if you are developing drivers or trying to debug and
215 identify kernel problems.
218 bool "Miscellaneous debug code"
220 depends on DEBUG_KERNEL
222 Say Y here if you need to enable miscellaneous debug code that should
223 be under a more specific debug option but isn't.
225 menu "Compile-time checks and compiler options"
230 A kernel debug info option other than "None" has been selected
231 in the "Debug information" choice below, indicating that debug
232 information will be generated for build targets.
234 # Clang generates .uleb128 with label differences for DWARF v5, a feature that
235 # older binutils ports do not support when utilizing RISC-V style linker
236 # relaxation: https://sourceware.org/bugzilla/show_bug.cgi?id=27215
237 config AS_HAS_NON_CONST_ULEB128
238 def_bool $(as-instr,.uleb128 .Lexpr_end4 - .Lexpr_start3\n.Lexpr_start3:\n.Lexpr_end4:)
241 prompt "Debug information"
242 depends on DEBUG_KERNEL
244 Selecting something other than "None" results in a kernel image
245 that will include debugging info resulting in a larger kernel image.
246 This adds debug symbols to the kernel and modules (gcc -g), and
247 is needed if you intend to use kernel crashdump or binary object
248 tools like crash, kgdb, LKCD, gdb, etc on the kernel.
250 Choose which version of DWARF debug info to emit. If unsure,
251 select "Toolchain default".
253 config DEBUG_INFO_NONE
254 bool "Disable debug information"
256 Do not build the kernel with debugging information, which will
257 result in a faster and smaller build.
259 config DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT
260 bool "Rely on the toolchain's implicit default DWARF version"
262 depends on !CC_IS_CLANG || AS_IS_LLVM || CLANG_VERSION < 140000 || (AS_IS_GNU && AS_VERSION >= 23502 && AS_HAS_NON_CONST_ULEB128)
264 The implicit default version of DWARF debug info produced by a
265 toolchain changes over time.
267 This can break consumers of the debug info that haven't upgraded to
268 support newer revisions, and prevent testing newer versions, but
269 those should be less common scenarios.
271 config DEBUG_INFO_DWARF4
272 bool "Generate DWARF Version 4 debuginfo"
274 depends on !CC_IS_CLANG || AS_IS_LLVM || (AS_IS_GNU && AS_VERSION >= 23502)
276 Generate DWARF v4 debug info. This requires gcc 4.5+, binutils 2.35.2
277 if using clang without clang's integrated assembler, and gdb 7.0+.
279 If you have consumers of DWARF debug info that are not ready for
280 newer revisions of DWARF, you may wish to choose this or have your
283 config DEBUG_INFO_DWARF5
284 bool "Generate DWARF Version 5 debuginfo"
286 depends on !ARCH_HAS_BROKEN_DWARF5
287 depends on !CC_IS_CLANG || AS_IS_LLVM || (AS_IS_GNU && AS_VERSION >= 23502 && AS_HAS_NON_CONST_ULEB128)
289 Generate DWARF v5 debug info. Requires binutils 2.35.2, gcc 5.0+ (gcc
290 5.0+ accepts the -gdwarf-5 flag but only had partial support for some
291 draft features until 7.0), and gdb 8.0+.
293 Changes to the structure of debug info in Version 5 allow for around
294 15-18% savings in resulting image and debug info section sizes as
295 compared to DWARF Version 4. DWARF Version 5 standardizes previous
296 extensions such as accelerators for symbol indexing and the format
297 for fission (.dwo/.dwp) files. Users may not want to select this
298 config if they rely on tooling that has not yet been updated to
299 support DWARF Version 5.
301 endchoice # "Debug information"
305 config DEBUG_INFO_REDUCED
306 bool "Reduce debugging information"
308 If you say Y here gcc is instructed to generate less debugging
309 information for structure types. This means that tools that
310 need full debugging information (like kgdb or systemtap) won't
311 be happy. But if you merely need debugging information to
312 resolve line numbers there is no loss. Advantage is that
313 build directory object sizes shrink dramatically over a full
314 DEBUG_INFO build and compile times are reduced too.
315 Only works with newer gcc versions.
318 prompt "Compressed Debug information"
320 Compress the resulting debug info. Results in smaller debug info sections,
321 but requires that consumers are able to decompress the results.
323 If unsure, choose DEBUG_INFO_COMPRESSED_NONE.
325 config DEBUG_INFO_COMPRESSED_NONE
326 bool "Don't compress debug information"
328 Don't compress debug info sections.
330 config DEBUG_INFO_COMPRESSED_ZLIB
331 bool "Compress debugging information with zlib"
332 depends on $(cc-option,-gz=zlib)
333 depends on $(ld-option,--compress-debug-sections=zlib)
335 Compress the debug information using zlib. Requires GCC 5.0+ or Clang
336 5.0+, binutils 2.26+, and zlib.
338 Users of dpkg-deb via scripts/package/builddeb may find an increase in
339 size of their debug .deb packages with this config set, due to the
340 debug info being compressed with zlib, then the object files being
341 recompressed with a different compression scheme. But this is still
342 preferable to setting $KDEB_COMPRESS to "none" which would be even
345 config DEBUG_INFO_COMPRESSED_ZSTD
346 bool "Compress debugging information with zstd"
347 depends on $(cc-option,-gz=zstd)
348 depends on $(ld-option,--compress-debug-sections=zstd)
350 Compress the debug information using zstd. This may provide better
351 compression than zlib, for about the same time costs, but requires newer
352 toolchain support. Requires GCC 13.0+ or Clang 16.0+, binutils 2.40+, and
355 endchoice # "Compressed Debug information"
357 config DEBUG_INFO_SPLIT
358 bool "Produce split debuginfo in .dwo files"
359 depends on $(cc-option,-gsplit-dwarf)
360 # RISC-V linker relaxation + -gsplit-dwarf has issues with LLVM and GCC
362 # https://github.com/llvm/llvm-project/issues/56642
363 # https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99090
364 depends on !RISCV || GCC_VERSION >= 120000
366 Generate debug info into separate .dwo files. This significantly
367 reduces the build directory size for builds with DEBUG_INFO,
368 because it stores the information only once on disk in .dwo
369 files instead of multiple times in object files and executables.
370 In addition the debug information is also compressed.
372 Requires recent gcc (4.7+) and recent gdb/binutils.
373 Any tool that packages or reads debug information would need
374 to know about the .dwo files and include them.
375 Incompatible with older versions of ccache.
377 config DEBUG_INFO_BTF
378 bool "Generate BTF type information"
379 depends on !DEBUG_INFO_SPLIT && !DEBUG_INFO_REDUCED
380 depends on !GCC_PLUGIN_RANDSTRUCT || COMPILE_TEST
381 depends on BPF_SYSCALL
382 depends on !DEBUG_INFO_DWARF5 || PAHOLE_VERSION >= 121
383 # pahole uses elfutils, which does not have support for Hexagon relocations
386 Generate deduplicated BTF type information from DWARF debug info.
387 Turning this on expects presence of pahole tool, which will convert
388 DWARF type info into equivalent deduplicated BTF type info.
390 config PAHOLE_HAS_SPLIT_BTF
391 def_bool PAHOLE_VERSION >= 119
393 config PAHOLE_HAS_BTF_TAG
394 def_bool PAHOLE_VERSION >= 123
395 depends on CC_IS_CLANG
397 Decide whether pahole emits btf_tag attributes (btf_type_tag and
398 btf_decl_tag) or not. Currently only clang compiler implements
399 these attributes, so make the config depend on CC_IS_CLANG.
401 config PAHOLE_HAS_LANG_EXCLUDE
402 def_bool PAHOLE_VERSION >= 124
404 Support for the --lang_exclude flag which makes pahole exclude
405 compilation units from the supplied language. Used in Kbuild to
406 omit Rust CUs which are not supported in version 1.24 of pahole,
407 otherwise it would emit malformed kernel and module binaries when
408 using DEBUG_INFO_BTF_MODULES.
410 config DEBUG_INFO_BTF_MODULES
411 bool "Generate BTF type information for kernel modules"
413 depends on DEBUG_INFO_BTF && MODULES && PAHOLE_HAS_SPLIT_BTF
415 Generate compact split BTF type information for kernel modules.
417 config MODULE_ALLOW_BTF_MISMATCH
418 bool "Allow loading modules with non-matching BTF type info"
419 depends on DEBUG_INFO_BTF_MODULES
421 For modules whose split BTF does not match vmlinux, load without
422 BTF rather than refusing to load. The default behavior with
423 module BTF enabled is to reject modules with such mismatches;
424 this option will still load module BTF where possible but ignore
425 it when a mismatch is found.
428 bool "Provide GDB scripts for kernel debugging"
430 This creates the required links to GDB helper scripts in the
431 build directory. If you load vmlinux into gdb, the helper
432 scripts will be automatically imported by gdb as well, and
433 additional functions are available to analyze a Linux kernel
434 instance. See Documentation/dev-tools/gdb-kernel-debugging.rst
440 int "Warn for stack frames larger than"
443 default 2048 if GCC_PLUGIN_LATENT_ENTROPY
444 default 2048 if PARISC
445 default 1536 if (!64BIT && XTENSA)
446 default 1280 if KASAN && !64BIT
447 default 1024 if !64BIT
448 default 2048 if 64BIT
450 Tell the compiler to warn at build time for stack frames larger than this.
451 Setting this too low will cause a lot of warnings.
452 Setting it to 0 disables the warning.
454 config STRIP_ASM_SYMS
455 bool "Strip assembler-generated symbols during link"
458 Strip internal assembler-generated symbols during a link (symbols
459 that look like '.Lxxx') so they don't pollute the output of
460 get_wchan() and suchlike.
463 bool "Generate readable assembler code"
464 depends on DEBUG_KERNEL
467 Disable some compiler optimizations that tend to generate human unreadable
468 assembler output. This may make the kernel slightly slower, but it helps
469 to keep kernel developers who have to stare a lot at assembler listings
472 config HEADERS_INSTALL
473 bool "Install uapi headers to usr/include"
476 This option will install uapi headers (headers exported to user-space)
477 into the usr/include directory for use during the kernel build.
478 This is unneeded for building the kernel itself, but needed for some
479 user-space program samples. It is also needed by some features such
480 as uapi header sanity checks.
482 config DEBUG_SECTION_MISMATCH
483 bool "Enable full Section mismatch analysis"
486 The section mismatch analysis checks if there are illegal
487 references from one section to another section.
488 During linktime or runtime, some sections are dropped;
489 any use of code/data previously in these sections would
490 most likely result in an oops.
491 In the code, functions and variables are annotated with
492 __init,, etc. (see the full list in include/linux/init.h),
493 which results in the code/data being placed in specific sections.
494 The section mismatch analysis is always performed after a full
495 kernel build, and enabling this option causes the following
496 additional step to occur:
497 - Add the option -fno-inline-functions-called-once to gcc commands.
498 When inlining a function annotated with __init in a non-init
499 function, we would lose the section information and thus
500 the analysis would not catch the illegal reference.
501 This option tells gcc to inline less (but it does result in
504 config SECTION_MISMATCH_WARN_ONLY
505 bool "Make section mismatch errors non-fatal"
508 If you say N here, the build process will fail if there are any
509 section mismatch, instead of just throwing warnings.
513 config DEBUG_FORCE_FUNCTION_ALIGN_64B
514 bool "Force all function address 64B aligned"
515 depends on EXPERT && (X86_64 || ARM64 || PPC32 || PPC64 || ARC || RISCV || S390)
516 select FUNCTION_ALIGNMENT_64B
518 There are cases that a commit from one domain changes the function
519 address alignment of other domains, and cause magic performance
520 bump (regression or improvement). Enable this option will help to
521 verify if the bump is caused by function alignment changes, while
522 it will slightly increase the kernel size and affect icache usage.
524 It is mainly for debug and performance tuning use.
527 # Select this config option from the architecture Kconfig, if it
528 # is preferred to always offer frame pointers as a config
529 # option on the architecture (regardless of KERNEL_DEBUG):
531 config ARCH_WANT_FRAME_POINTERS
535 bool "Compile the kernel with frame pointers"
536 depends on DEBUG_KERNEL && (M68K || UML || SUPERH) || ARCH_WANT_FRAME_POINTERS
537 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS
539 If you say Y here the resulting kernel image will be slightly
540 larger and slower, but it gives very useful debugging information
541 in case of kernel bugs. (precise oopses/stacktraces/warnings)
546 config STACK_VALIDATION
547 bool "Compile-time stack metadata validation"
548 depends on HAVE_STACK_VALIDATION && UNWINDER_FRAME_POINTER
552 Validate frame pointer rules at compile-time. This helps ensure that
553 runtime stack traces are more reliable.
555 For more information, see
556 tools/objtool/Documentation/objtool.txt.
558 config NOINSTR_VALIDATION
560 depends on HAVE_NOINSTR_VALIDATION && DEBUG_ENTRY
565 bool "Generate vmlinux.map file when linking"
568 Selecting this option will pass "-Map=vmlinux.map" to ld
569 when linking vmlinux. That file can be useful for verifying
570 and debugging magic section games, and for seeing which
571 pieces of code get eliminated with
572 CONFIG_LD_DEAD_CODE_DATA_ELIMINATION.
574 config DEBUG_FORCE_WEAK_PER_CPU
575 bool "Force weak per-cpu definitions"
576 depends on DEBUG_KERNEL
578 s390 and alpha require percpu variables in modules to be
579 defined weak to work around addressing range issue which
580 puts the following two restrictions on percpu variable
583 1. percpu symbols must be unique whether static or not
584 2. percpu variables can't be defined inside a function
586 To ensure that generic code follows the above rules, this
587 option forces all percpu variables to be defined as weak.
589 endmenu # "Compiler options"
591 menu "Generic Kernel Debugging Instruments"
594 bool "Magic SysRq key"
597 If you say Y here, you will have some control over the system even
598 if the system crashes for example during kernel debugging (e.g., you
599 will be able to flush the buffer cache to disk, reboot the system
600 immediately or dump some status information). This is accomplished
601 by pressing various keys while holding SysRq (Alt+PrintScreen). It
602 also works on a serial console (on PC hardware at least), if you
603 send a BREAK and then within 5 seconds a command keypress. The
604 keys are documented in <file:Documentation/admin-guide/sysrq.rst>.
605 Don't say Y unless you really know what this hack does.
607 config MAGIC_SYSRQ_DEFAULT_ENABLE
608 hex "Enable magic SysRq key functions by default"
609 depends on MAGIC_SYSRQ
612 Specifies which SysRq key functions are enabled by default.
613 This may be set to 1 or 0 to enable or disable them all, or
614 to a bitmask as described in Documentation/admin-guide/sysrq.rst.
616 config MAGIC_SYSRQ_SERIAL
617 bool "Enable magic SysRq key over serial"
618 depends on MAGIC_SYSRQ
621 Many embedded boards have a disconnected TTL level serial which can
622 generate some garbage that can lead to spurious false sysrq detects.
623 This option allows you to decide whether you want to enable the
626 config MAGIC_SYSRQ_SERIAL_SEQUENCE
627 string "Char sequence that enables magic SysRq over serial"
628 depends on MAGIC_SYSRQ_SERIAL
631 Specifies a sequence of characters that can follow BREAK to enable
632 SysRq on a serial console.
634 If unsure, leave an empty string and the option will not be enabled.
637 bool "Debug Filesystem"
639 debugfs is a virtual file system that kernel developers use to put
640 debugging files into. Enable this option to be able to read and
641 write to these files.
643 For detailed documentation on the debugfs API, see
644 Documentation/filesystems/.
649 prompt "Debugfs default access"
651 default DEBUG_FS_ALLOW_ALL
653 This selects the default access restrictions for debugfs.
654 It can be overridden with kernel command line option
655 debugfs=[on,no-mount,off]. The restrictions apply for API access
656 and filesystem registration.
658 config DEBUG_FS_ALLOW_ALL
661 No restrictions apply. Both API and filesystem registration
662 is on. This is the normal default operation.
664 config DEBUG_FS_DISALLOW_MOUNT
665 bool "Do not register debugfs as filesystem"
667 The API is open but filesystem is not loaded. Clients can still do
668 their work and read with debug tools that do not need
671 config DEBUG_FS_ALLOW_NONE
674 Access is off. Clients get -PERM when trying to create nodes in
675 debugfs tree and debugfs is not registered as a filesystem.
676 Client can then back-off or continue without debugfs access.
680 source "lib/Kconfig.kgdb"
681 source "lib/Kconfig.ubsan"
682 source "lib/Kconfig.kcsan"
686 menu "Networking Debugging"
688 source "net/Kconfig.debug"
690 endmenu # "Networking Debugging"
692 menu "Memory Debugging"
694 source "mm/Kconfig.debug"
697 bool "Debug object operations"
698 depends on DEBUG_KERNEL
700 If you say Y here, additional code will be inserted into the
701 kernel to track the life time of various objects and validate
702 the operations on those objects.
704 config DEBUG_OBJECTS_SELFTEST
705 bool "Debug objects selftest"
706 depends on DEBUG_OBJECTS
708 This enables the selftest of the object debug code.
710 config DEBUG_OBJECTS_FREE
711 bool "Debug objects in freed memory"
712 depends on DEBUG_OBJECTS
714 This enables checks whether a k/v free operation frees an area
715 which contains an object which has not been deactivated
716 properly. This can make kmalloc/kfree-intensive workloads
719 config DEBUG_OBJECTS_TIMERS
720 bool "Debug timer objects"
721 depends on DEBUG_OBJECTS
723 If you say Y here, additional code will be inserted into the
724 timer routines to track the life time of timer objects and
725 validate the timer operations.
727 config DEBUG_OBJECTS_WORK
728 bool "Debug work objects"
729 depends on DEBUG_OBJECTS
731 If you say Y here, additional code will be inserted into the
732 work queue routines to track the life time of work objects and
733 validate the work operations.
735 config DEBUG_OBJECTS_RCU_HEAD
736 bool "Debug RCU callbacks objects"
737 depends on DEBUG_OBJECTS
739 Enable this to turn on debugging of RCU list heads (call_rcu() usage).
741 config DEBUG_OBJECTS_PERCPU_COUNTER
742 bool "Debug percpu counter objects"
743 depends on DEBUG_OBJECTS
745 If you say Y here, additional code will be inserted into the
746 percpu counter routines to track the life time of percpu counter
747 objects and validate the percpu counter operations.
749 config DEBUG_OBJECTS_ENABLE_DEFAULT
750 int "debug_objects bootup default value (0-1)"
753 depends on DEBUG_OBJECTS
755 Debug objects boot parameter default value
757 config SHRINKER_DEBUG
758 bool "Enable shrinker debugging support"
761 Say Y to enable the shrinker debugfs interface which provides
762 visibility into the kernel memory shrinkers subsystem.
763 Disable it to avoid an extra memory footprint.
765 config DEBUG_STACK_USAGE
766 bool "Stack utilization instrumentation"
767 depends on DEBUG_KERNEL
769 Enables the display of the minimum amount of free stack which each
770 task has ever had available in the sysrq-T and sysrq-P debug output.
771 Also emits a message to dmesg when a process exits if that process
772 used more stack space than previously exiting processes.
774 This option will slow down process creation somewhat.
776 config SCHED_STACK_END_CHECK
777 bool "Detect stack corruption on calls to schedule()"
778 depends on DEBUG_KERNEL
781 This option checks for a stack overrun on calls to schedule().
782 If the stack end location is found to be over written always panic as
783 the content of the corrupted region can no longer be trusted.
784 This is to ensure no erroneous behaviour occurs which could result in
785 data corruption or a sporadic crash at a later stage once the region
786 is examined. The runtime overhead introduced is minimal.
788 config ARCH_HAS_DEBUG_VM_PGTABLE
791 An architecture should select this when it can successfully
792 build and run DEBUG_VM_PGTABLE.
794 config DEBUG_VM_IRQSOFF
795 def_bool DEBUG_VM && !PREEMPT_RT
799 depends on DEBUG_KERNEL
801 Enable this to turn on extended checks in the virtual-memory system
802 that may impact performance.
806 config DEBUG_VM_SHOOT_LAZIES
807 bool "Debug MMU_LAZY_TLB_SHOOTDOWN implementation"
809 depends on MMU_LAZY_TLB_SHOOTDOWN
811 Enable additional IPIs that ensure lazy tlb mm references are removed
812 before the mm is freed.
816 config DEBUG_VM_MAPLE_TREE
817 bool "Debug VM maple trees"
819 select DEBUG_MAPLE_TREE
821 Enable VM maple tree debugging information and extra validations.
826 bool "Debug VM red-black trees"
829 Enable VM red-black tree debugging information and extra validations.
833 config DEBUG_VM_PGFLAGS
834 bool "Debug page-flags operations"
837 Enables extra validation on page flags operations.
841 config DEBUG_VM_PGTABLE
842 bool "Debug arch page table for semantics compliance"
844 depends on ARCH_HAS_DEBUG_VM_PGTABLE
845 default y if DEBUG_VM
847 This option provides a debug method which can be used to test
848 architecture page table helper functions on various platforms in
849 verifying if they comply with expected generic MM semantics. This
850 will help architecture code in making sure that any changes or
851 new additions of these helpers still conform to expected
852 semantics of the generic MM. Platforms will have to opt in for
853 this through ARCH_HAS_DEBUG_VM_PGTABLE.
857 config ARCH_HAS_DEBUG_VIRTUAL
861 bool "Debug VM translations"
862 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL
864 Enable some costly sanity checks in virtual to page code. This can
865 catch mistakes with virt_to_page() and friends.
869 config DEBUG_NOMMU_REGIONS
870 bool "Debug the global anon/private NOMMU mapping region tree"
871 depends on DEBUG_KERNEL && !MMU
873 This option causes the global tree of anonymous and private mapping
874 regions to be regularly checked for invalid topology.
876 config DEBUG_MEMORY_INIT
877 bool "Debug memory initialisation" if EXPERT
880 Enable this for additional checks during memory initialisation.
881 The sanity checks verify aspects of the VM such as the memory model
882 and other information provided by the architecture. Verbose
883 information will be printed at KERN_DEBUG loglevel depending
884 on the mminit_loglevel= command-line option.
888 config MEMORY_NOTIFIER_ERROR_INJECT
889 tristate "Memory hotplug notifier error injection module"
890 depends on MEMORY_HOTPLUG && NOTIFIER_ERROR_INJECTION
892 This option provides the ability to inject artificial errors to
893 memory hotplug notifier chain callbacks. It is controlled through
894 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory
896 If the notifier call chain should be failed with some events
897 notified, write the error code to "actions/<notifier event>/error".
899 Example: Inject memory hotplug offline error (-12 == -ENOMEM)
901 # cd /sys/kernel/debug/notifier-error-inject/memory
902 # echo -12 > actions/MEM_GOING_OFFLINE/error
903 # echo offline > /sys/devices/system/memory/memoryXXX/state
904 bash: echo: write error: Cannot allocate memory
906 To compile this code as a module, choose M here: the module will
907 be called memory-notifier-error-inject.
911 config DEBUG_PER_CPU_MAPS
912 bool "Debug access to per_cpu maps"
913 depends on DEBUG_KERNEL
916 Say Y to verify that the per_cpu map being accessed has
917 been set up. This adds a fair amount of code to kernel memory
918 and decreases performance.
922 config DEBUG_KMAP_LOCAL
923 bool "Debug kmap_local temporary mappings"
924 depends on DEBUG_KERNEL && KMAP_LOCAL
926 This option enables additional error checking for the kmap_local
927 infrastructure. Disable for production use.
929 config ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP
932 config DEBUG_KMAP_LOCAL_FORCE_MAP
933 bool "Enforce kmap_local temporary mappings"
934 depends on DEBUG_KERNEL && ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP
936 select DEBUG_KMAP_LOCAL
938 This option enforces temporary mappings through the kmap_local
939 mechanism for non-highmem pages and on non-highmem systems.
940 Disable this for production systems!
943 bool "Highmem debugging"
944 depends on DEBUG_KERNEL && HIGHMEM
945 select DEBUG_KMAP_LOCAL_FORCE_MAP if ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP
946 select DEBUG_KMAP_LOCAL
948 This option enables additional error checking for high memory
949 systems. Disable for production systems.
951 config HAVE_DEBUG_STACKOVERFLOW
954 config DEBUG_STACKOVERFLOW
955 bool "Check for stack overflows"
956 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW
958 Say Y here if you want to check for overflows of kernel, IRQ
959 and exception stacks (if your architecture uses them). This
960 option will show detailed messages if free stack space drops
961 below a certain limit.
963 These kinds of bugs usually occur when call-chains in the
964 kernel get too deep, especially when interrupts are
967 Use this in cases where you see apparently random memory
968 corruption, especially if it appears in 'struct thread_info'
970 If in doubt, say "N".
972 source "lib/Kconfig.kasan"
973 source "lib/Kconfig.kfence"
974 source "lib/Kconfig.kmsan"
976 endmenu # "Memory Debugging"
979 bool "Debug shared IRQ handlers"
980 depends on DEBUG_KERNEL
982 Enable this to generate a spurious interrupt just before a shared
983 interrupt handler is deregistered (generating one when registering
984 is currently disabled). Drivers need to handle this correctly. Some
985 don't and need to be caught.
987 menu "Debug Oops, Lockups and Hangs"
992 Say Y here to enable the kernel to panic when it oopses. This
993 has the same effect as setting oops=panic on the kernel command
996 This feature is useful to ensure that the kernel does not do
997 anything erroneous after an oops which could result in data
998 corruption or other issues.
1002 config PANIC_ON_OOPS_VALUE
1005 default 0 if !PANIC_ON_OOPS
1006 default 1 if PANIC_ON_OOPS
1008 config PANIC_TIMEOUT
1012 Set the timeout value (in seconds) until a reboot occurs when
1013 the kernel panics. If n = 0, then we wait forever. A timeout
1014 value n > 0 will wait n seconds before rebooting, while a timeout
1015 value n < 0 will reboot immediately.
1017 config LOCKUP_DETECTOR
1020 config SOFTLOCKUP_DETECTOR
1021 bool "Detect Soft Lockups"
1022 depends on DEBUG_KERNEL && !S390
1023 select LOCKUP_DETECTOR
1025 Say Y here to enable the kernel to act as a watchdog to detect
1028 Softlockups are bugs that cause the kernel to loop in kernel
1029 mode for more than 20 seconds, without giving other tasks a
1030 chance to run. The current stack trace is displayed upon
1031 detection and the system will stay locked up.
1033 config BOOTPARAM_SOFTLOCKUP_PANIC
1034 bool "Panic (Reboot) On Soft Lockups"
1035 depends on SOFTLOCKUP_DETECTOR
1037 Say Y here to enable the kernel to panic on "soft lockups",
1038 which are bugs that cause the kernel to loop in kernel
1039 mode for more than 20 seconds (configurable using the watchdog_thresh
1040 sysctl), without giving other tasks a chance to run.
1042 The panic can be used in combination with panic_timeout,
1043 to cause the system to reboot automatically after a
1044 lockup has been detected. This feature is useful for
1045 high-availability systems that have uptime guarantees and
1046 where a lockup must be resolved ASAP.
1050 config HAVE_HARDLOCKUP_DETECTOR_BUDDY
1056 # Global switch whether to build a hardlockup detector at all. It is available
1057 # only when the architecture supports at least one implementation. There are
1058 # two exceptions. The hardlockup detector is never enabled on:
1060 # s390: it reported many false positives there
1062 # sparc64: has a custom implementation which is not using the common
1063 # hardlockup command line options and sysctl interface.
1065 config HARDLOCKUP_DETECTOR
1066 bool "Detect Hard Lockups"
1067 depends on DEBUG_KERNEL && !S390 && !HARDLOCKUP_DETECTOR_SPARC64
1068 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_BUDDY || HAVE_HARDLOCKUP_DETECTOR_ARCH
1069 imply HARDLOCKUP_DETECTOR_PERF
1070 imply HARDLOCKUP_DETECTOR_BUDDY
1071 imply HARDLOCKUP_DETECTOR_ARCH
1072 select LOCKUP_DETECTOR
1075 Say Y here to enable the kernel to act as a watchdog to detect
1078 Hardlockups are bugs that cause the CPU to loop in kernel mode
1079 for more than 10 seconds, without letting other interrupts have a
1080 chance to run. The current stack trace is displayed upon detection
1081 and the system will stay locked up.
1084 # Note that arch-specific variants are always preferred.
1086 config HARDLOCKUP_DETECTOR_PREFER_BUDDY
1087 bool "Prefer the buddy CPU hardlockup detector"
1088 depends on HARDLOCKUP_DETECTOR
1089 depends on HAVE_HARDLOCKUP_DETECTOR_PERF && HAVE_HARDLOCKUP_DETECTOR_BUDDY
1090 depends on !HAVE_HARDLOCKUP_DETECTOR_ARCH
1092 Say Y here to prefer the buddy hardlockup detector over the perf one.
1094 With the buddy detector, each CPU uses its softlockup hrtimer
1095 to check that the next CPU is processing hrtimer interrupts by
1096 verifying that a counter is increasing.
1098 This hardlockup detector is useful on systems that don't have
1099 an arch-specific hardlockup detector or if resources needed
1100 for the hardlockup detector are better used for other things.
1102 config HARDLOCKUP_DETECTOR_PERF
1104 depends on HARDLOCKUP_DETECTOR
1105 depends on HAVE_HARDLOCKUP_DETECTOR_PERF && !HARDLOCKUP_DETECTOR_PREFER_BUDDY
1106 depends on !HAVE_HARDLOCKUP_DETECTOR_ARCH
1107 select HARDLOCKUP_DETECTOR_COUNTS_HRTIMER
1109 config HARDLOCKUP_DETECTOR_BUDDY
1111 depends on HARDLOCKUP_DETECTOR
1112 depends on HAVE_HARDLOCKUP_DETECTOR_BUDDY
1113 depends on !HAVE_HARDLOCKUP_DETECTOR_PERF || HARDLOCKUP_DETECTOR_PREFER_BUDDY
1114 depends on !HAVE_HARDLOCKUP_DETECTOR_ARCH
1115 select HARDLOCKUP_DETECTOR_COUNTS_HRTIMER
1117 config HARDLOCKUP_DETECTOR_ARCH
1119 depends on HARDLOCKUP_DETECTOR
1120 depends on HAVE_HARDLOCKUP_DETECTOR_ARCH
1122 The arch-specific implementation of the hardlockup detector will
1126 # Both the "perf" and "buddy" hardlockup detectors count hrtimer
1127 # interrupts. This config enables functions managing this common code.
1129 config HARDLOCKUP_DETECTOR_COUNTS_HRTIMER
1131 select SOFTLOCKUP_DETECTOR
1134 # Enables a timestamp based low pass filter to compensate for perf based
1135 # hard lockup detection which runs too fast due to turbo modes.
1137 config HARDLOCKUP_CHECK_TIMESTAMP
1140 config BOOTPARAM_HARDLOCKUP_PANIC
1141 bool "Panic (Reboot) On Hard Lockups"
1142 depends on HARDLOCKUP_DETECTOR
1144 Say Y here to enable the kernel to panic on "hard lockups",
1145 which are bugs that cause the kernel to loop in kernel
1146 mode with interrupts disabled for more than 10 seconds (configurable
1147 using the watchdog_thresh sysctl).
1151 config DETECT_HUNG_TASK
1152 bool "Detect Hung Tasks"
1153 depends on DEBUG_KERNEL
1154 default SOFTLOCKUP_DETECTOR
1156 Say Y here to enable the kernel to detect "hung tasks",
1157 which are bugs that cause the task to be stuck in
1158 uninterruptible "D" state indefinitely.
1160 When a hung task is detected, the kernel will print the
1161 current stack trace (which you should report), but the
1162 task will stay in uninterruptible state. If lockdep is
1163 enabled then all held locks will also be reported. This
1164 feature has negligible overhead.
1166 config DEFAULT_HUNG_TASK_TIMEOUT
1167 int "Default timeout for hung task detection (in seconds)"
1168 depends on DETECT_HUNG_TASK
1171 This option controls the default timeout (in seconds) used
1172 to determine when a task has become non-responsive and should
1175 It can be adjusted at runtime via the kernel.hung_task_timeout_secs
1176 sysctl or by writing a value to
1177 /proc/sys/kernel/hung_task_timeout_secs.
1179 A timeout of 0 disables the check. The default is two minutes.
1180 Keeping the default should be fine in most cases.
1182 config BOOTPARAM_HUNG_TASK_PANIC
1183 bool "Panic (Reboot) On Hung Tasks"
1184 depends on DETECT_HUNG_TASK
1186 Say Y here to enable the kernel to panic on "hung tasks",
1187 which are bugs that cause the kernel to leave a task stuck
1188 in uninterruptible "D" state.
1190 The panic can be used in combination with panic_timeout,
1191 to cause the system to reboot automatically after a
1192 hung task has been detected. This feature is useful for
1193 high-availability systems that have uptime guarantees and
1194 where a hung tasks must be resolved ASAP.
1199 bool "Detect Workqueue Stalls"
1200 depends on DEBUG_KERNEL
1202 Say Y here to enable stall detection on workqueues. If a
1203 worker pool doesn't make forward progress on a pending work
1204 item for over a given amount of time, 30s by default, a
1205 warning message is printed along with dump of workqueue
1206 state. This can be configured through kernel parameter
1207 "workqueue.watchdog_thresh" and its sysfs counterpart.
1209 config WQ_CPU_INTENSIVE_REPORT
1210 bool "Report per-cpu work items which hog CPU for too long"
1211 depends on DEBUG_KERNEL
1213 Say Y here to enable reporting of concurrency-managed per-cpu work
1214 items that hog CPUs for longer than
1215 workqueue.cpu_intensive_thresh_us. Workqueue automatically
1216 detects and excludes them from concurrency management to prevent
1217 them from stalling other per-cpu work items. Occassional
1218 triggering may not necessarily indicate a problem. Repeated
1219 triggering likely indicates that the work item should be switched
1220 to use an unbound workqueue.
1223 tristate "Test module to generate lockups"
1226 This builds the "test_lockup" module that helps to make sure
1227 that watchdogs and lockup detectors are working properly.
1229 Depending on module parameters it could emulate soft or hard
1230 lockup, "hung task", or locking arbitrary lock for a long time.
1231 Also it could generate series of lockups with cooling-down periods.
1235 endmenu # "Debug lockups and hangs"
1237 menu "Scheduler Debugging"
1240 bool "Collect scheduler debugging info"
1241 depends on DEBUG_KERNEL && DEBUG_FS
1244 If you say Y here, the /sys/kernel/debug/sched file will be provided
1245 that can help debug the scheduler. The runtime overhead of this
1253 bool "Collect scheduler statistics"
1254 depends on DEBUG_KERNEL && PROC_FS
1257 If you say Y here, additional code will be inserted into the
1258 scheduler and related routines to collect statistics about
1259 scheduler behavior and provide them in /proc/schedstat. These
1260 stats may be useful for both tuning and debugging the scheduler
1261 If you aren't debugging the scheduler or trying to tune a specific
1262 application, you can say N to avoid the very slight overhead
1267 config DEBUG_TIMEKEEPING
1268 bool "Enable extra timekeeping sanity checking"
1270 This option will enable additional timekeeping sanity checks
1271 which may be helpful when diagnosing issues where timekeeping
1272 problems are suspected.
1274 This may include checks in the timekeeping hotpaths, so this
1275 option may have a (very small) performance impact to some
1280 config DEBUG_PREEMPT
1281 bool "Debug preemptible kernel"
1282 depends on DEBUG_KERNEL && PREEMPTION && TRACE_IRQFLAGS_SUPPORT
1284 If you say Y here then the kernel will use a debug variant of the
1285 commonly used smp_processor_id() function and will print warnings
1286 if kernel code uses it in a preemption-unsafe way. Also, the kernel
1287 will detect preemption count underflows.
1289 This option has potential to introduce high runtime overhead,
1290 depending on workload as it triggers debugging routines for each
1291 this_cpu operation. It should only be used for debugging purposes.
1293 menu "Lock Debugging (spinlocks, mutexes, etc...)"
1295 config LOCK_DEBUGGING_SUPPORT
1297 depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1300 config PROVE_LOCKING
1301 bool "Lock debugging: prove locking correctness"
1302 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1304 select DEBUG_SPINLOCK
1305 select DEBUG_MUTEXES if !PREEMPT_RT
1306 select DEBUG_RT_MUTEXES if RT_MUTEXES
1307 select DEBUG_RWSEMS if !PREEMPT_RT
1308 select DEBUG_WW_MUTEX_SLOWPATH
1309 select DEBUG_LOCK_ALLOC
1310 select PREEMPT_COUNT if !ARCH_NO_PREEMPT
1311 select TRACE_IRQFLAGS
1314 This feature enables the kernel to prove that all locking
1315 that occurs in the kernel runtime is mathematically
1316 correct: that under no circumstance could an arbitrary (and
1317 not yet triggered) combination of observed locking
1318 sequences (on an arbitrary number of CPUs, running an
1319 arbitrary number of tasks and interrupt contexts) cause a
1322 In short, this feature enables the kernel to report locking
1323 related deadlocks before they actually occur.
1325 The proof does not depend on how hard and complex a
1326 deadlock scenario would be to trigger: how many
1327 participant CPUs, tasks and irq-contexts would be needed
1328 for it to trigger. The proof also does not depend on
1329 timing: if a race and a resulting deadlock is possible
1330 theoretically (no matter how unlikely the race scenario
1331 is), it will be proven so and will immediately be
1332 reported by the kernel (once the event is observed that
1333 makes the deadlock theoretically possible).
1335 If a deadlock is impossible (i.e. the locking rules, as
1336 observed by the kernel, are mathematically correct), the
1337 kernel reports nothing.
1339 NOTE: this feature can also be enabled for rwlocks, mutexes
1340 and rwsems - in which case all dependencies between these
1341 different locking variants are observed and mapped too, and
1342 the proof of observed correctness is also maintained for an
1343 arbitrary combination of these separate locking variants.
1345 For more details, see Documentation/locking/lockdep-design.rst.
1347 config PROVE_RAW_LOCK_NESTING
1348 bool "Enable raw_spinlock - spinlock nesting checks"
1349 depends on PROVE_LOCKING
1352 Enable the raw_spinlock vs. spinlock nesting checks which ensure
1353 that the lock nesting rules for PREEMPT_RT enabled kernels are
1356 NOTE: There are known nesting problems. So if you enable this
1357 option expect lockdep splats until these problems have been fully
1358 addressed which is work in progress. This config switch allows to
1359 identify and analyze these problems. It will be removed and the
1360 check permanently enabled once the main issues have been fixed.
1362 If unsure, select N.
1365 bool "Lock usage statistics"
1366 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1368 select DEBUG_SPINLOCK
1369 select DEBUG_MUTEXES if !PREEMPT_RT
1370 select DEBUG_RT_MUTEXES if RT_MUTEXES
1371 select DEBUG_LOCK_ALLOC
1374 This feature enables tracking lock contention points
1376 For more details, see Documentation/locking/lockstat.rst
1378 This also enables lock events required by "perf lock",
1380 If you want to use "perf lock", you also need to turn on
1381 CONFIG_EVENT_TRACING.
1383 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events.
1384 (CONFIG_LOCKDEP defines "acquire" and "release" events.)
1386 config DEBUG_RT_MUTEXES
1387 bool "RT Mutex debugging, deadlock detection"
1388 depends on DEBUG_KERNEL && RT_MUTEXES
1390 This allows rt mutex semantics violations and rt mutex related
1391 deadlocks (lockups) to be detected and reported automatically.
1393 config DEBUG_SPINLOCK
1394 bool "Spinlock and rw-lock debugging: basic checks"
1395 depends on DEBUG_KERNEL
1396 select UNINLINE_SPIN_UNLOCK
1398 Say Y here and build SMP to catch missing spinlock initialization
1399 and certain other kinds of spinlock errors commonly made. This is
1400 best used in conjunction with the NMI watchdog so that spinlock
1401 deadlocks are also debuggable.
1403 config DEBUG_MUTEXES
1404 bool "Mutex debugging: basic checks"
1405 depends on DEBUG_KERNEL && !PREEMPT_RT
1407 This feature allows mutex semantics violations to be detected and
1410 config DEBUG_WW_MUTEX_SLOWPATH
1411 bool "Wait/wound mutex debugging: Slowpath testing"
1412 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1413 select DEBUG_LOCK_ALLOC
1414 select DEBUG_SPINLOCK
1415 select DEBUG_MUTEXES if !PREEMPT_RT
1416 select DEBUG_RT_MUTEXES if PREEMPT_RT
1418 This feature enables slowpath testing for w/w mutex users by
1419 injecting additional -EDEADLK wound/backoff cases. Together with
1420 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this
1421 will test all possible w/w mutex interface abuse with the
1422 exception of simply not acquiring all the required locks.
1423 Note that this feature can introduce significant overhead, so
1424 it really should not be enabled in a production or distro kernel,
1425 even a debug kernel. If you are a driver writer, enable it. If
1426 you are a distro, do not.
1429 bool "RW Semaphore debugging: basic checks"
1430 depends on DEBUG_KERNEL && !PREEMPT_RT
1432 This debugging feature allows mismatched rw semaphore locks
1433 and unlocks to be detected and reported.
1435 config DEBUG_LOCK_ALLOC
1436 bool "Lock debugging: detect incorrect freeing of live locks"
1437 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1438 select DEBUG_SPINLOCK
1439 select DEBUG_MUTEXES if !PREEMPT_RT
1440 select DEBUG_RT_MUTEXES if RT_MUTEXES
1443 This feature will check whether any held lock (spinlock, rwlock,
1444 mutex or rwsem) is incorrectly freed by the kernel, via any of the
1445 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(),
1446 vfree(), etc.), whether a live lock is incorrectly reinitialized via
1447 spin_lock_init()/mutex_init()/etc., or whether there is any lock
1448 held during task exit.
1452 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1457 config LOCKDEP_SMALL
1461 int "Bitsize for MAX_LOCKDEP_ENTRIES"
1462 depends on LOCKDEP && !LOCKDEP_SMALL
1466 Try increasing this value if you hit "BUG: MAX_LOCKDEP_ENTRIES too low!" message.
1468 config LOCKDEP_CHAINS_BITS
1469 int "Bitsize for MAX_LOCKDEP_CHAINS"
1470 depends on LOCKDEP && !LOCKDEP_SMALL
1474 Try increasing this value if you hit "BUG: MAX_LOCKDEP_CHAINS too low!" message.
1476 config LOCKDEP_STACK_TRACE_BITS
1477 int "Bitsize for MAX_STACK_TRACE_ENTRIES"
1478 depends on LOCKDEP && !LOCKDEP_SMALL
1482 Try increasing this value if you hit "BUG: MAX_STACK_TRACE_ENTRIES too low!" message.
1484 config LOCKDEP_STACK_TRACE_HASH_BITS
1485 int "Bitsize for STACK_TRACE_HASH_SIZE"
1486 depends on LOCKDEP && !LOCKDEP_SMALL
1490 Try increasing this value if you need large STACK_TRACE_HASH_SIZE.
1492 config LOCKDEP_CIRCULAR_QUEUE_BITS
1493 int "Bitsize for elements in circular_queue struct"
1498 Try increasing this value if you hit "lockdep bfs error:-1" warning due to __cq_enqueue() failure.
1500 config DEBUG_LOCKDEP
1501 bool "Lock dependency engine debugging"
1502 depends on DEBUG_KERNEL && LOCKDEP
1503 select DEBUG_IRQFLAGS
1505 If you say Y here, the lock dependency engine will do
1506 additional runtime checks to debug itself, at the price
1507 of more runtime overhead.
1509 config DEBUG_ATOMIC_SLEEP
1510 bool "Sleep inside atomic section checking"
1511 select PREEMPT_COUNT
1512 depends on DEBUG_KERNEL
1513 depends on !ARCH_NO_PREEMPT
1515 If you say Y here, various routines which may sleep will become very
1516 noisy if they are called inside atomic sections: when a spinlock is
1517 held, inside an rcu read side critical section, inside preempt disabled
1518 sections, inside an interrupt, etc...
1520 config DEBUG_LOCKING_API_SELFTESTS
1521 bool "Locking API boot-time self-tests"
1522 depends on DEBUG_KERNEL
1524 Say Y here if you want the kernel to run a short self-test during
1525 bootup. The self-test checks whether common types of locking bugs
1526 are detected by debugging mechanisms or not. (if you disable
1527 lock debugging then those bugs won't be detected of course.)
1528 The following locking APIs are covered: spinlocks, rwlocks,
1531 config LOCK_TORTURE_TEST
1532 tristate "torture tests for locking"
1533 depends on DEBUG_KERNEL
1536 This option provides a kernel module that runs torture tests
1537 on kernel locking primitives. The kernel module may be built
1538 after the fact on the running kernel to be tested, if desired.
1540 Say Y here if you want kernel locking-primitive torture tests
1541 to be built into the kernel.
1542 Say M if you want these torture tests to build as a module.
1543 Say N if you are unsure.
1545 config WW_MUTEX_SELFTEST
1546 tristate "Wait/wound mutex selftests"
1548 This option provides a kernel module that runs tests on the
1549 on the struct ww_mutex locking API.
1551 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction
1552 with this test harness.
1554 Say M if you want these self tests to build as a module.
1555 Say N if you are unsure.
1557 config SCF_TORTURE_TEST
1558 tristate "torture tests for smp_call_function*()"
1559 depends on DEBUG_KERNEL
1562 This option provides a kernel module that runs torture tests
1563 on the smp_call_function() family of primitives. The kernel
1564 module may be built after the fact on the running kernel to
1565 be tested, if desired.
1567 config CSD_LOCK_WAIT_DEBUG
1568 bool "Debugging for csd_lock_wait(), called from smp_call_function*()"
1569 depends on DEBUG_KERNEL
1573 This option enables debug prints when CPUs are slow to respond
1574 to the smp_call_function*() IPI wrappers. These debug prints
1575 include the IPI handler function currently executing (if any)
1576 and relevant stack traces.
1578 config CSD_LOCK_WAIT_DEBUG_DEFAULT
1579 bool "Default csd_lock_wait() debugging on at boot time"
1580 depends on CSD_LOCK_WAIT_DEBUG
1584 This option causes the csdlock_debug= kernel boot parameter to
1585 default to 1 (basic debugging) instead of 0 (no debugging).
1587 endmenu # lock debugging
1589 config TRACE_IRQFLAGS
1590 depends on TRACE_IRQFLAGS_SUPPORT
1593 Enables hooks to interrupt enabling and disabling for
1594 either tracing or lock debugging.
1596 config TRACE_IRQFLAGS_NMI
1598 depends on TRACE_IRQFLAGS
1599 depends on TRACE_IRQFLAGS_NMI_SUPPORT
1601 config NMI_CHECK_CPU
1602 bool "Debugging for CPUs failing to respond to backtrace requests"
1603 depends on DEBUG_KERNEL
1607 Enables debug prints when a CPU fails to respond to a given
1608 backtrace NMI. These prints provide some reasons why a CPU
1609 might legitimately be failing to respond, for example, if it
1610 is offline of if ignore_nmis is set.
1612 config DEBUG_IRQFLAGS
1613 bool "Debug IRQ flag manipulation"
1615 Enables checks for potentially unsafe enabling or disabling of
1616 interrupts, such as calling raw_local_irq_restore() when interrupts
1620 bool "Stack backtrace support"
1621 depends on STACKTRACE_SUPPORT
1623 This option causes the kernel to create a /proc/pid/stack for
1624 every process, showing its current stack trace.
1625 It is also used by various kernel debugging features that require
1626 stack trace generation.
1628 config WARN_ALL_UNSEEDED_RANDOM
1629 bool "Warn for all uses of unseeded randomness"
1632 Some parts of the kernel contain bugs relating to their use of
1633 cryptographically secure random numbers before it's actually possible
1634 to generate those numbers securely. This setting ensures that these
1635 flaws don't go unnoticed, by enabling a message, should this ever
1636 occur. This will allow people with obscure setups to know when things
1637 are going wrong, so that they might contact developers about fixing
1640 Unfortunately, on some models of some architectures getting
1641 a fully seeded CRNG is extremely difficult, and so this can
1642 result in dmesg getting spammed for a surprisingly long
1643 time. This is really bad from a security perspective, and
1644 so architecture maintainers really need to do what they can
1645 to get the CRNG seeded sooner after the system is booted.
1646 However, since users cannot do anything actionable to
1647 address this, by default this option is disabled.
1649 Say Y here if you want to receive warnings for all uses of
1650 unseeded randomness. This will be of use primarily for
1651 those developers interested in improving the security of
1652 Linux kernels running on their architecture (or
1655 config DEBUG_KOBJECT
1656 bool "kobject debugging"
1657 depends on DEBUG_KERNEL
1659 If you say Y here, some extra kobject debugging messages will be sent
1662 config DEBUG_KOBJECT_RELEASE
1663 bool "kobject release debugging"
1664 depends on DEBUG_OBJECTS_TIMERS
1666 kobjects are reference counted objects. This means that their
1667 last reference count put is not predictable, and the kobject can
1668 live on past the point at which a driver decides to drop its
1669 initial reference to the kobject gained on allocation. An
1670 example of this would be a struct device which has just been
1673 However, some buggy drivers assume that after such an operation,
1674 the memory backing the kobject can be immediately freed. This
1675 goes completely against the principles of a refcounted object.
1677 If you say Y here, the kernel will delay the release of kobjects
1678 on the last reference count to improve the visibility of this
1679 kind of kobject release bug.
1681 config HAVE_DEBUG_BUGVERBOSE
1684 menu "Debug kernel data structures"
1687 bool "Debug linked list manipulation"
1688 depends on DEBUG_KERNEL
1689 select LIST_HARDENED
1691 Enable this to turn on extended checks in the linked-list walking
1694 This option trades better quality error reports for performance, and
1695 is more suitable for kernel debugging. If you care about performance,
1696 you should only enable CONFIG_LIST_HARDENED instead.
1701 bool "Debug priority linked list manipulation"
1702 depends on DEBUG_KERNEL
1704 Enable this to turn on extended checks in the priority-ordered
1705 linked-list (plist) walking routines. This checks the entire
1706 list multiple times during each manipulation.
1711 bool "Debug SG table operations"
1712 depends on DEBUG_KERNEL
1714 Enable this to turn on checks on scatter-gather tables. This can
1715 help find problems with drivers that do not properly initialize
1720 config DEBUG_NOTIFIERS
1721 bool "Debug notifier call chains"
1722 depends on DEBUG_KERNEL
1724 Enable this to turn on sanity checking for notifier call chains.
1725 This is most useful for kernel developers to make sure that
1726 modules properly unregister themselves from notifier chains.
1727 This is a relatively cheap check but if you care about maximum
1730 config DEBUG_CLOSURES
1731 bool "Debug closures (bcache async widgits)"
1735 Keeps all active closures in a linked list and provides a debugfs
1736 interface to list them, which makes it possible to see asynchronous
1737 operations that get stuck.
1739 config DEBUG_MAPLE_TREE
1740 bool "Debug maple trees"
1741 depends on DEBUG_KERNEL
1743 Enable maple tree debugging information and extra validations.
1749 source "kernel/rcu/Kconfig.debug"
1751 config DEBUG_WQ_FORCE_RR_CPU
1752 bool "Force round-robin CPU selection for unbound work items"
1753 depends on DEBUG_KERNEL
1756 Workqueue used to implicitly guarantee that work items queued
1757 without explicit CPU specified are put on the local CPU. This
1758 guarantee is no longer true and while local CPU is still
1759 preferred work items may be put on foreign CPUs. Kernel
1760 parameter "workqueue.debug_force_rr_cpu" is added to force
1761 round-robin CPU selection to flush out usages which depend on the
1762 now broken guarantee. This config option enables the debug
1763 feature by default. When enabled, memory and cache locality will
1766 config CPU_HOTPLUG_STATE_CONTROL
1767 bool "Enable CPU hotplug state control"
1768 depends on DEBUG_KERNEL
1769 depends on HOTPLUG_CPU
1772 Allows to write steps between "offline" and "online" to the CPUs
1773 sysfs target file so states can be stepped granular. This is a debug
1774 option for now as the hotplug machinery cannot be stopped and
1775 restarted at arbitrary points yet.
1777 Say N if your are unsure.
1780 bool "Latency measuring infrastructure"
1781 depends on DEBUG_KERNEL
1782 depends on STACKTRACE_SUPPORT
1784 depends on FRAME_POINTER || MIPS || PPC || S390 || MICROBLAZE || ARM || ARC || X86
1790 Enable this option if you want to use the LatencyTOP tool
1791 to find out which userspace is blocking on what kernel operations.
1793 config DEBUG_CGROUP_REF
1794 bool "Disable inlining of cgroup css reference count functions"
1795 depends on DEBUG_KERNEL
1800 Force cgroup css reference count functions to not be inlined so
1801 that they can be kprobed for debugging.
1803 source "kernel/trace/Kconfig"
1805 config PROVIDE_OHCI1394_DMA_INIT
1806 bool "Remote debugging over FireWire early on boot"
1807 depends on PCI && X86
1809 If you want to debug problems which hang or crash the kernel early
1810 on boot and the crashing machine has a FireWire port, you can use
1811 this feature to remotely access the memory of the crashed machine
1812 over FireWire. This employs remote DMA as part of the OHCI1394
1813 specification which is now the standard for FireWire controllers.
1815 With remote DMA, you can monitor the printk buffer remotely using
1816 firescope and access all memory below 4GB using fireproxy from gdb.
1817 Even controlling a kernel debugger is possible using remote DMA.
1821 If ohci1394_dma=early is used as boot parameter, it will initialize
1822 all OHCI1394 controllers which are found in the PCI config space.
1824 As all changes to the FireWire bus such as enabling and disabling
1825 devices cause a bus reset and thereby disable remote DMA for all
1826 devices, be sure to have the cable plugged and FireWire enabled on
1827 the debugging host before booting the debug target for debugging.
1829 This code (~1k) is freed after boot. By then, the firewire stack
1830 in charge of the OHCI-1394 controllers should be used instead.
1832 See Documentation/core-api/debugging-via-ohci1394.rst for more information.
1834 source "samples/Kconfig"
1836 config ARCH_HAS_DEVMEM_IS_ALLOWED
1839 config STRICT_DEVMEM
1840 bool "Filter access to /dev/mem"
1841 depends on MMU && DEVMEM
1842 depends on ARCH_HAS_DEVMEM_IS_ALLOWED || GENERIC_LIB_DEVMEM_IS_ALLOWED
1843 default y if PPC || X86 || ARM64
1845 If this option is disabled, you allow userspace (root) access to all
1846 of memory, including kernel and userspace memory. Accidental
1847 access to this is obviously disastrous, but specific access can
1848 be used by people debugging the kernel. Note that with PAT support
1849 enabled, even in this case there are restrictions on /dev/mem
1850 use due to the cache aliasing requirements.
1852 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem
1853 file only allows userspace access to PCI space and the BIOS code and
1854 data regions. This is sufficient for dosemu and X and all common
1859 config IO_STRICT_DEVMEM
1860 bool "Filter I/O access to /dev/mem"
1861 depends on STRICT_DEVMEM
1863 If this option is disabled, you allow userspace (root) access to all
1864 io-memory regardless of whether a driver is actively using that
1865 range. Accidental access to this is obviously disastrous, but
1866 specific access can be used by people debugging kernel drivers.
1868 If this option is switched on, the /dev/mem file only allows
1869 userspace access to *idle* io-memory ranges (see /proc/iomem) This
1870 may break traditional users of /dev/mem (dosemu, legacy X, etc...)
1871 if the driver using a given range cannot be disabled.
1875 menu "$(SRCARCH) Debugging"
1877 source "arch/$(SRCARCH)/Kconfig.debug"
1881 menu "Kernel Testing and Coverage"
1883 source "lib/kunit/Kconfig"
1885 config NOTIFIER_ERROR_INJECTION
1886 tristate "Notifier error injection"
1887 depends on DEBUG_KERNEL
1890 This option provides the ability to inject artificial errors to
1891 specified notifier chain callbacks. It is useful to test the error
1892 handling of notifier call chain failures.
1896 config PM_NOTIFIER_ERROR_INJECT
1897 tristate "PM notifier error injection module"
1898 depends on PM && NOTIFIER_ERROR_INJECTION
1899 default m if PM_DEBUG
1901 This option provides the ability to inject artificial errors to
1902 PM notifier chain callbacks. It is controlled through debugfs
1903 interface /sys/kernel/debug/notifier-error-inject/pm
1905 If the notifier call chain should be failed with some events
1906 notified, write the error code to "actions/<notifier event>/error".
1908 Example: Inject PM suspend error (-12 = -ENOMEM)
1910 # cd /sys/kernel/debug/notifier-error-inject/pm/
1911 # echo -12 > actions/PM_SUSPEND_PREPARE/error
1912 # echo mem > /sys/power/state
1913 bash: echo: write error: Cannot allocate memory
1915 To compile this code as a module, choose M here: the module will
1916 be called pm-notifier-error-inject.
1920 config OF_RECONFIG_NOTIFIER_ERROR_INJECT
1921 tristate "OF reconfig notifier error injection module"
1922 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION
1924 This option provides the ability to inject artificial errors to
1925 OF reconfig notifier chain callbacks. It is controlled
1926 through debugfs interface under
1927 /sys/kernel/debug/notifier-error-inject/OF-reconfig/
1929 If the notifier call chain should be failed with some events
1930 notified, write the error code to "actions/<notifier event>/error".
1932 To compile this code as a module, choose M here: the module will
1933 be called of-reconfig-notifier-error-inject.
1937 config NETDEV_NOTIFIER_ERROR_INJECT
1938 tristate "Netdev notifier error injection module"
1939 depends on NET && NOTIFIER_ERROR_INJECTION
1941 This option provides the ability to inject artificial errors to
1942 netdevice notifier chain callbacks. It is controlled through debugfs
1943 interface /sys/kernel/debug/notifier-error-inject/netdev
1945 If the notifier call chain should be failed with some events
1946 notified, write the error code to "actions/<notifier event>/error".
1948 Example: Inject netdevice mtu change error (-22 = -EINVAL)
1950 # cd /sys/kernel/debug/notifier-error-inject/netdev
1951 # echo -22 > actions/NETDEV_CHANGEMTU/error
1952 # ip link set eth0 mtu 1024
1953 RTNETLINK answers: Invalid argument
1955 To compile this code as a module, choose M here: the module will
1956 be called netdev-notifier-error-inject.
1960 config FUNCTION_ERROR_INJECTION
1961 bool "Fault-injections of functions"
1962 depends on HAVE_FUNCTION_ERROR_INJECTION && KPROBES
1964 Add fault injections into various functions that are annotated with
1965 ALLOW_ERROR_INJECTION() in the kernel. BPF may also modify the return
1966 value of these functions. This is useful to test error paths of code.
1970 config FAULT_INJECTION
1971 bool "Fault-injection framework"
1972 depends on DEBUG_KERNEL
1974 Provide fault-injection framework.
1975 For more details, see Documentation/fault-injection/.
1978 bool "Fault-injection capability for kmalloc"
1979 depends on FAULT_INJECTION
1981 Provide fault-injection capability for kmalloc.
1983 config FAIL_PAGE_ALLOC
1984 bool "Fault-injection capability for alloc_pages()"
1985 depends on FAULT_INJECTION
1987 Provide fault-injection capability for alloc_pages().
1989 config FAULT_INJECTION_USERCOPY
1990 bool "Fault injection capability for usercopy functions"
1991 depends on FAULT_INJECTION
1993 Provides fault-injection capability to inject failures
1994 in usercopy functions (copy_from_user(), get_user(), ...).
1996 config FAIL_MAKE_REQUEST
1997 bool "Fault-injection capability for disk IO"
1998 depends on FAULT_INJECTION && BLOCK
2000 Provide fault-injection capability for disk IO.
2002 config FAIL_IO_TIMEOUT
2003 bool "Fault-injection capability for faking disk interrupts"
2004 depends on FAULT_INJECTION && BLOCK
2006 Provide fault-injection capability on end IO handling. This
2007 will make the block layer "forget" an interrupt as configured,
2008 thus exercising the error handling.
2010 Only works with drivers that use the generic timeout handling,
2011 for others it won't do anything.
2014 bool "Fault-injection capability for futexes"
2016 depends on FAULT_INJECTION && FUTEX
2018 Provide fault-injection capability for futexes.
2020 config FAULT_INJECTION_DEBUG_FS
2021 bool "Debugfs entries for fault-injection capabilities"
2022 depends on FAULT_INJECTION && SYSFS && DEBUG_FS
2024 Enable configuration of fault-injection capabilities via debugfs.
2026 config FAIL_FUNCTION
2027 bool "Fault-injection capability for functions"
2028 depends on FAULT_INJECTION_DEBUG_FS && FUNCTION_ERROR_INJECTION
2030 Provide function-based fault-injection capability.
2031 This will allow you to override a specific function with a return
2032 with given return value. As a result, function caller will see
2033 an error value and have to handle it. This is useful to test the
2034 error handling in various subsystems.
2036 config FAIL_MMC_REQUEST
2037 bool "Fault-injection capability for MMC IO"
2038 depends on FAULT_INJECTION_DEBUG_FS && MMC
2040 Provide fault-injection capability for MMC IO.
2041 This will make the mmc core return data errors. This is
2042 useful to test the error handling in the mmc block device
2043 and to test how the mmc host driver handles retries from
2047 bool "Fault-injection capability for SunRPC"
2048 depends on FAULT_INJECTION_DEBUG_FS && SUNRPC_DEBUG
2050 Provide fault-injection capability for SunRPC and
2053 config FAULT_INJECTION_CONFIGFS
2054 bool "Configfs interface for fault-injection capabilities"
2055 depends on FAULT_INJECTION
2058 This option allows configfs-based drivers to dynamically configure
2059 fault-injection via configfs. Each parameter for driver-specific
2060 fault-injection can be made visible as a configfs attribute in a
2064 config FAULT_INJECTION_STACKTRACE_FILTER
2065 bool "stacktrace filter for fault-injection capabilities"
2066 depends on FAULT_INJECTION
2067 depends on (FAULT_INJECTION_DEBUG_FS || FAULT_INJECTION_CONFIGFS) && STACKTRACE_SUPPORT
2069 depends on FRAME_POINTER || MIPS || PPC || S390 || MICROBLAZE || ARM || ARC || X86
2071 Provide stacktrace filter for fault-injection capabilities
2073 config ARCH_HAS_KCOV
2076 An architecture should select this when it can successfully
2077 build and run with CONFIG_KCOV. This typically requires
2078 disabling instrumentation for some early boot code.
2080 config CC_HAS_SANCOV_TRACE_PC
2081 def_bool $(cc-option,-fsanitize-coverage=trace-pc)
2085 bool "Code coverage for fuzzing"
2086 depends on ARCH_HAS_KCOV
2087 depends on CC_HAS_SANCOV_TRACE_PC || GCC_PLUGINS
2088 depends on !ARCH_WANTS_NO_INSTR || HAVE_NOINSTR_HACK || \
2089 GCC_VERSION >= 120000 || CC_IS_CLANG
2091 select GCC_PLUGIN_SANCOV if !CC_HAS_SANCOV_TRACE_PC
2092 select OBJTOOL if HAVE_NOINSTR_HACK
2094 KCOV exposes kernel code coverage information in a form suitable
2095 for coverage-guided fuzzing (randomized testing).
2097 For more details, see Documentation/dev-tools/kcov.rst.
2099 config KCOV_ENABLE_COMPARISONS
2100 bool "Enable comparison operands collection by KCOV"
2102 depends on $(cc-option,-fsanitize-coverage=trace-cmp)
2104 KCOV also exposes operands of every comparison in the instrumented
2105 code along with operand sizes and PCs of the comparison instructions.
2106 These operands can be used by fuzzing engines to improve the quality
2107 of fuzzing coverage.
2109 config KCOV_INSTRUMENT_ALL
2110 bool "Instrument all code by default"
2114 If you are doing generic system call fuzzing (like e.g. syzkaller),
2115 then you will want to instrument the whole kernel and you should
2116 say y here. If you are doing more targeted fuzzing (like e.g.
2117 filesystem fuzzing with AFL) then you will want to enable coverage
2118 for more specific subsets of files, and should say n here.
2120 config KCOV_IRQ_AREA_SIZE
2121 hex "Size of interrupt coverage collection area in words"
2125 KCOV uses preallocated per-cpu areas to collect coverage from
2126 soft interrupts. This specifies the size of those areas in the
2127 number of unsigned long words.
2129 menuconfig RUNTIME_TESTING_MENU
2130 bool "Runtime Testing"
2133 if RUNTIME_TESTING_MENU
2136 tristate "Dhrystone benchmark test"
2138 Enable this to include the Dhrystone 2.1 benchmark. This test
2139 calculates the number of Dhrystones per second, and the number of
2140 DMIPS (Dhrystone MIPS) obtained when the Dhrystone score is divided
2141 by 1757 (the number of Dhrystones per second obtained on the VAX
2142 11/780, nominally a 1 MIPS machine).
2144 To run the benchmark, it needs to be enabled explicitly, either from
2145 the kernel command line (when built-in), or from userspace (when
2146 built-in or modular).
2148 Run once during kernel boot:
2152 Set number of iterations from kernel command line:
2154 test_dhry.iterations=<n>
2156 Set number of iterations from userspace:
2158 echo <n> > /sys/module/test_dhry/parameters/iterations
2160 Trigger manual run from userspace:
2162 echo y > /sys/module/test_dhry/parameters/run
2164 If the number of iterations is <= 0, the test will devise a suitable
2165 number of iterations (test runs for at least 2s) automatically.
2166 This process takes ca. 4s.
2171 tristate "Linux Kernel Dump Test Tool Module"
2174 This module enables testing of the different dumping mechanisms by
2175 inducing system failures at predefined crash points.
2176 If you don't need it: say N
2177 Choose M here to compile this code as a module. The module will be
2180 Documentation on how to use the module can be found in
2181 Documentation/fault-injection/provoke-crashes.rst
2183 config CPUMASK_KUNIT_TEST
2184 tristate "KUnit test for cpumask" if !KUNIT_ALL_TESTS
2186 default KUNIT_ALL_TESTS
2188 Enable to turn on cpumask tests, running at boot or module load time.
2190 For more information on KUnit and unit tests in general, please refer
2191 to the KUnit documentation in Documentation/dev-tools/kunit/.
2195 config TEST_LIST_SORT
2196 tristate "Linked list sorting test" if !KUNIT_ALL_TESTS
2198 default KUNIT_ALL_TESTS
2200 Enable this to turn on 'list_sort()' function test. This test is
2201 executed only once during system boot (so affects only boot time),
2202 or at module load time.
2206 config TEST_MIN_HEAP
2207 tristate "Min heap test"
2208 depends on DEBUG_KERNEL || m
2210 Enable this to turn on min heap function tests. This test is
2211 executed only once during system boot (so affects only boot time),
2212 or at module load time.
2217 tristate "Array-based sort test" if !KUNIT_ALL_TESTS
2219 default KUNIT_ALL_TESTS
2221 This option enables the self-test function of 'sort()' at boot,
2222 or at module load time.
2227 tristate "64bit/32bit division and modulo test"
2228 depends on DEBUG_KERNEL || m
2230 Enable this to turn on 'do_div()' function test. This test is
2231 executed only once during system boot (so affects only boot time),
2232 or at module load time.
2236 config TEST_IOV_ITER
2237 tristate "Test iov_iter operation" if !KUNIT_ALL_TESTS
2240 default KUNIT_ALL_TESTS
2242 Enable this to turn on testing of the operation of the I/O iterator
2243 (iov_iter). This test is executed only once during system boot (so
2244 affects only boot time), or at module load time.
2248 config KPROBES_SANITY_TEST
2249 tristate "Kprobes sanity tests" if !KUNIT_ALL_TESTS
2250 depends on DEBUG_KERNEL
2253 select STACKTRACE if ARCH_CORRECT_STACKTRACE_ON_KRETPROBE
2254 default KUNIT_ALL_TESTS
2256 This option provides for testing basic kprobes functionality on
2257 boot. Samples of kprobe and kretprobe are inserted and
2258 verified for functionality.
2260 Say N if you are unsure.
2262 config FPROBE_SANITY_TEST
2263 bool "Self test for fprobe"
2264 depends on DEBUG_KERNEL
2268 This option will enable testing the fprobe when the system boot.
2269 A series of tests are made to verify that the fprobe is functioning
2272 Say N if you are unsure.
2274 config BACKTRACE_SELF_TEST
2275 tristate "Self test for the backtrace code"
2276 depends on DEBUG_KERNEL
2278 This option provides a kernel module that can be used to test
2279 the kernel stack backtrace code. This option is not useful
2280 for distributions or general kernels, but only for kernel
2281 developers working on architecture code.
2283 Note that if you want to also test saved backtraces, you will
2284 have to enable STACKTRACE as well.
2286 Say N if you are unsure.
2288 config TEST_REF_TRACKER
2289 tristate "Self test for reference tracker"
2290 depends on DEBUG_KERNEL && STACKTRACE_SUPPORT
2293 This option provides a kernel module performing tests
2294 using reference tracker infrastructure.
2296 Say N if you are unsure.
2299 tristate "Red-Black tree test"
2300 depends on DEBUG_KERNEL
2302 A benchmark measuring the performance of the rbtree library.
2303 Also includes rbtree invariant checks.
2305 config REED_SOLOMON_TEST
2306 tristate "Reed-Solomon library test"
2307 depends on DEBUG_KERNEL || m
2309 select REED_SOLOMON_ENC16
2310 select REED_SOLOMON_DEC16
2312 This option enables the self-test function of rslib at boot,
2313 or at module load time.
2317 config INTERVAL_TREE_TEST
2318 tristate "Interval tree test"
2319 depends on DEBUG_KERNEL
2320 select INTERVAL_TREE
2322 A benchmark measuring the performance of the interval tree library
2325 tristate "Per cpu operations test"
2326 depends on m && DEBUG_KERNEL
2328 Enable this option to build test module which validates per-cpu
2333 config ATOMIC64_SELFTEST
2334 tristate "Perform an atomic64_t self-test"
2336 Enable this option to test the atomic64_t functions at boot or
2337 at module load time.
2341 config ASYNC_RAID6_TEST
2342 tristate "Self test for hardware accelerated raid6 recovery"
2343 depends on ASYNC_RAID6_RECOV
2346 This is a one-shot self test that permutes through the
2347 recovery of all the possible two disk failure scenarios for a
2348 N-disk array. Recovery is performed with the asynchronous
2349 raid6 recovery routines, and will optionally use an offload
2350 engine if one is available.
2355 tristate "Test functions located in the hexdump module at runtime"
2357 config STRING_KUNIT_TEST
2358 tristate "KUnit test string functions at runtime" if !KUNIT_ALL_TESTS
2360 default KUNIT_ALL_TESTS
2362 config STRING_HELPERS_KUNIT_TEST
2363 tristate "KUnit test string helpers at runtime" if !KUNIT_ALL_TESTS
2365 default KUNIT_ALL_TESTS
2368 tristate "Test kstrto*() family of functions at runtime"
2371 tristate "Test printf() family of functions at runtime"
2374 tristate "Test scanf() family of functions at runtime"
2377 tristate "Test bitmap_*() family of functions at runtime"
2379 Enable this option to test the bitmap functions at boot.
2384 tristate "Test functions located in the uuid module at runtime"
2387 tristate "Test the XArray code at runtime"
2389 config TEST_MAPLE_TREE
2390 tristate "Test the Maple Tree code at runtime or module load"
2392 Enable this option to test the maple tree code functions at boot, or
2393 when the module is loaded. Enable "Debug Maple Trees" will enable
2394 more verbose output on failures.
2398 config TEST_RHASHTABLE
2399 tristate "Perform selftest on resizable hash table"
2401 Enable this option to test the rhashtable functions at boot.
2406 tristate "Perform selftest on IDA functions"
2409 tristate "Perform selftest on priority array manager"
2412 Enable this option to test priority array manager on boot
2417 config TEST_IRQ_TIMINGS
2418 bool "IRQ timings selftest"
2419 depends on IRQ_TIMINGS
2421 Enable this option to test the irq timings code on boot.
2426 tristate "Test module loading with 'hello world' module"
2429 This builds the "test_module" module that emits "Hello, world"
2430 on printk when loaded. It is designed to be used for basic
2431 evaluation of the module loading subsystem (for example when
2432 validating module verification). It lacks any extra dependencies,
2433 and will not normally be loaded by the system unless explicitly
2439 tristate "Test module for compilation of bitops operations"
2442 This builds the "test_bitops" module that is much like the
2443 TEST_LKM module except that it does a basic exercise of the
2444 set/clear_bit macros and get_count_order/long to make sure there are
2445 no compiler warnings from C=1 sparse checker or -Wextra
2446 compilations. It has no dependencies and doesn't run or load unless
2447 explicitly requested by name. for example: modprobe test_bitops.
2452 tristate "Test module for stress/performance analysis of vmalloc allocator"
2457 This builds the "test_vmalloc" module that should be used for
2458 stress and performance analysis. So, any new change for vmalloc
2459 subsystem can be evaluated from performance and stability point
2464 config TEST_USER_COPY
2465 tristate "Test user/kernel boundary protections"
2468 This builds the "test_user_copy" module that runs sanity checks
2469 on the copy_to/from_user infrastructure, making sure basic
2470 user/kernel boundary testing is working. If it fails to load,
2471 a regression has been detected in the user/kernel memory boundary
2477 tristate "Test BPF filter functionality"
2480 This builds the "test_bpf" module that runs various test vectors
2481 against the BPF interpreter or BPF JIT compiler depending on the
2482 current setting. This is in particular useful for BPF JIT compiler
2483 development, but also to run regression tests against changes in
2484 the interpreter code. It also enables test stubs for eBPF maps and
2485 verifier used by user space verifier testsuite.
2489 config TEST_BLACKHOLE_DEV
2490 tristate "Test blackhole netdev functionality"
2493 This builds the "test_blackhole_dev" module that validates the
2494 data path through this blackhole netdev.
2498 config FIND_BIT_BENCHMARK
2499 tristate "Test find_bit functions"
2501 This builds the "test_find_bit" module that measure find_*_bit()
2502 functions performance.
2506 config TEST_FIRMWARE
2507 tristate "Test firmware loading via userspace interface"
2508 depends on FW_LOADER
2510 This builds the "test_firmware" module that creates a userspace
2511 interface for testing firmware loading. This can be used to
2512 control the triggering of firmware loading without needing an
2513 actual firmware-using device. The contents can be rechecked by
2519 tristate "sysctl test driver"
2520 depends on PROC_SYSCTL
2522 This builds the "test_sysctl" module. This driver enables to test the
2523 proc sysctl interfaces available to drivers safely without affecting
2524 production knobs which might alter system functionality.
2528 config BITFIELD_KUNIT
2529 tristate "KUnit test bitfield functions at runtime" if !KUNIT_ALL_TESTS
2531 default KUNIT_ALL_TESTS
2533 Enable this option to test the bitfield functions at boot.
2535 KUnit tests run during boot and output the results to the debug log
2536 in TAP format (http://testanything.org/). Only useful for kernel devs
2537 running the KUnit test harness, and not intended for inclusion into a
2540 For more information on KUnit and unit tests in general please refer
2541 to the KUnit documentation in Documentation/dev-tools/kunit/.
2545 config CHECKSUM_KUNIT
2546 tristate "KUnit test checksum functions at runtime" if !KUNIT_ALL_TESTS
2548 default KUNIT_ALL_TESTS
2550 Enable this option to test the checksum functions at boot.
2552 KUnit tests run during boot and output the results to the debug log
2553 in TAP format (http://testanything.org/). Only useful for kernel devs
2554 running the KUnit test harness, and not intended for inclusion into a
2557 For more information on KUnit and unit tests in general please refer
2558 to the KUnit documentation in Documentation/dev-tools/kunit/.
2562 config HASH_KUNIT_TEST
2563 tristate "KUnit Test for integer hash functions" if !KUNIT_ALL_TESTS
2565 default KUNIT_ALL_TESTS
2567 Enable this option to test the kernel's string (<linux/stringhash.h>), and
2568 integer (<linux/hash.h>) hash functions on boot.
2570 KUnit tests run during boot and output the results to the debug log
2571 in TAP format (https://testanything.org/). Only useful for kernel devs
2572 running the KUnit test harness, and not intended for inclusion into a
2575 For more information on KUnit and unit tests in general please refer
2576 to the KUnit documentation in Documentation/dev-tools/kunit/.
2578 This is intended to help people writing architecture-specific
2579 optimized versions. If unsure, say N.
2581 config RESOURCE_KUNIT_TEST
2582 tristate "KUnit test for resource API" if !KUNIT_ALL_TESTS
2584 default KUNIT_ALL_TESTS
2586 This builds the resource API unit test.
2587 Tests the logic of API provided by resource.c and ioport.h.
2588 For more information on KUnit and unit tests in general please refer
2589 to the KUnit documentation in Documentation/dev-tools/kunit/.
2593 config SYSCTL_KUNIT_TEST
2594 tristate "KUnit test for sysctl" if !KUNIT_ALL_TESTS
2596 default KUNIT_ALL_TESTS
2598 This builds the proc sysctl unit test, which runs on boot.
2599 Tests the API contract and implementation correctness of sysctl.
2600 For more information on KUnit and unit tests in general please refer
2601 to the KUnit documentation in Documentation/dev-tools/kunit/.
2605 config LIST_KUNIT_TEST
2606 tristate "KUnit Test for Kernel Linked-list structures" if !KUNIT_ALL_TESTS
2608 default KUNIT_ALL_TESTS
2610 This builds the linked list KUnit test suite.
2611 It tests that the API and basic functionality of the list_head type
2612 and associated macros.
2614 KUnit tests run during boot and output the results to the debug log
2615 in TAP format (https://testanything.org/). Only useful for kernel devs
2616 running the KUnit test harness, and not intended for inclusion into a
2619 For more information on KUnit and unit tests in general please refer
2620 to the KUnit documentation in Documentation/dev-tools/kunit/.
2624 config HASHTABLE_KUNIT_TEST
2625 tristate "KUnit Test for Kernel Hashtable structures" if !KUNIT_ALL_TESTS
2627 default KUNIT_ALL_TESTS
2629 This builds the hashtable KUnit test suite.
2630 It tests the basic functionality of the API defined in
2631 include/linux/hashtable.h. For more information on KUnit and
2632 unit tests in general please refer to the KUnit documentation
2633 in Documentation/dev-tools/kunit/.
2637 config LINEAR_RANGES_TEST
2638 tristate "KUnit test for linear_ranges"
2640 select LINEAR_RANGES
2642 This builds the linear_ranges unit test, which runs on boot.
2643 Tests the linear_ranges logic correctness.
2644 For more information on KUnit and unit tests in general please refer
2645 to the KUnit documentation in Documentation/dev-tools/kunit/.
2649 config CMDLINE_KUNIT_TEST
2650 tristate "KUnit test for cmdline API" if !KUNIT_ALL_TESTS
2652 default KUNIT_ALL_TESTS
2654 This builds the cmdline API unit test.
2655 Tests the logic of API provided by cmdline.c.
2656 For more information on KUnit and unit tests in general please refer
2657 to the KUnit documentation in Documentation/dev-tools/kunit/.
2662 tristate "KUnit test for bits.h" if !KUNIT_ALL_TESTS
2664 default KUNIT_ALL_TESTS
2666 This builds the bits unit test.
2667 Tests the logic of macros defined in bits.h.
2668 For more information on KUnit and unit tests in general please refer
2669 to the KUnit documentation in Documentation/dev-tools/kunit/.
2673 config SLUB_KUNIT_TEST
2674 tristate "KUnit test for SLUB cache error detection" if !KUNIT_ALL_TESTS
2675 depends on SLUB_DEBUG && KUNIT
2676 default KUNIT_ALL_TESTS
2678 This builds SLUB allocator unit test.
2679 Tests SLUB cache debugging functionality.
2680 For more information on KUnit and unit tests in general please refer
2681 to the KUnit documentation in Documentation/dev-tools/kunit/.
2685 config RATIONAL_KUNIT_TEST
2686 tristate "KUnit test for rational.c" if !KUNIT_ALL_TESTS
2687 depends on KUNIT && RATIONAL
2688 default KUNIT_ALL_TESTS
2690 This builds the rational math unit test.
2691 For more information on KUnit and unit tests in general please refer
2692 to the KUnit documentation in Documentation/dev-tools/kunit/.
2696 config MEMCPY_KUNIT_TEST
2697 tristate "Test memcpy(), memmove(), and memset() functions at runtime" if !KUNIT_ALL_TESTS
2699 default KUNIT_ALL_TESTS
2701 Builds unit tests for memcpy(), memmove(), and memset() functions.
2702 For more information on KUnit and unit tests in general please refer
2703 to the KUnit documentation in Documentation/dev-tools/kunit/.
2707 config IS_SIGNED_TYPE_KUNIT_TEST
2708 tristate "Test is_signed_type() macro" if !KUNIT_ALL_TESTS
2710 default KUNIT_ALL_TESTS
2712 Builds unit tests for the is_signed_type() macro.
2714 For more information on KUnit and unit tests in general please refer
2715 to the KUnit documentation in Documentation/dev-tools/kunit/.
2719 config OVERFLOW_KUNIT_TEST
2720 tristate "Test check_*_overflow() functions at runtime" if !KUNIT_ALL_TESTS
2722 default KUNIT_ALL_TESTS
2724 Builds unit tests for the check_*_overflow(), size_*(), allocation, and
2727 For more information on KUnit and unit tests in general please refer
2728 to the KUnit documentation in Documentation/dev-tools/kunit/.
2732 config STACKINIT_KUNIT_TEST
2733 tristate "Test level of stack variable initialization" if !KUNIT_ALL_TESTS
2735 default KUNIT_ALL_TESTS
2737 Test if the kernel is zero-initializing stack variables and
2738 padding. Coverage is controlled by compiler flags,
2739 CONFIG_INIT_STACK_ALL_PATTERN, CONFIG_INIT_STACK_ALL_ZERO,
2740 CONFIG_GCC_PLUGIN_STRUCTLEAK, CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF,
2741 or CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL.
2743 config FORTIFY_KUNIT_TEST
2744 tristate "Test fortified str*() and mem*() function internals at runtime" if !KUNIT_ALL_TESTS
2746 default KUNIT_ALL_TESTS
2748 Builds unit tests for checking internals of FORTIFY_SOURCE as used
2749 by the str*() and mem*() family of functions. For testing runtime
2750 traps of FORTIFY_SOURCE, see LKDTM's "FORTIFY_*" tests.
2752 config HW_BREAKPOINT_KUNIT_TEST
2753 bool "Test hw_breakpoint constraints accounting" if !KUNIT_ALL_TESTS
2754 depends on HAVE_HW_BREAKPOINT
2756 default KUNIT_ALL_TESTS
2758 Tests for hw_breakpoint constraints accounting.
2762 config STRCAT_KUNIT_TEST
2763 tristate "Test strcat() family of functions at runtime" if !KUNIT_ALL_TESTS
2765 default KUNIT_ALL_TESTS
2767 config STRSCPY_KUNIT_TEST
2768 tristate "Test strscpy*() family of functions at runtime" if !KUNIT_ALL_TESTS
2770 default KUNIT_ALL_TESTS
2772 config SIPHASH_KUNIT_TEST
2773 tristate "Perform selftest on siphash functions" if !KUNIT_ALL_TESTS
2775 default KUNIT_ALL_TESTS
2777 Enable this option to test the kernel's siphash (<linux/siphash.h>) hash
2778 functions on boot (or module load).
2780 This is intended to help people writing architecture-specific
2781 optimized versions. If unsure, say N.
2784 tristate "udelay test driver"
2786 This builds the "udelay_test" module that helps to make sure
2787 that udelay() is working properly.
2791 config TEST_STATIC_KEYS
2792 tristate "Test static keys"
2795 Test the static key interfaces.
2799 config TEST_DYNAMIC_DEBUG
2800 tristate "Test DYNAMIC_DEBUG"
2801 depends on DYNAMIC_DEBUG
2803 This module registers a tracer callback to count enabled
2804 pr_debugs in a 'do_debugging' function, then alters their
2805 enablements, calls the function, and compares counts.
2810 tristate "kmod stress tester"
2812 depends on NETDEVICES && NET_CORE && INET # for TUN
2814 depends on PAGE_SIZE_LESS_THAN_256KB # for BTRFS
2820 Test the kernel's module loading mechanism: kmod. kmod implements
2821 support to load modules using the Linux kernel's usermode helper.
2822 This test provides a series of tests against kmod.
2824 Although technically you can either build test_kmod as a module or
2825 into the kernel we disallow building it into the kernel since
2826 it stress tests request_module() and this will very likely cause
2827 some issues by taking over precious threads available from other
2828 module load requests, ultimately this could be fatal.
2832 tools/testing/selftests/kmod/kmod.sh --help
2836 config TEST_DEBUG_VIRTUAL
2837 tristate "Test CONFIG_DEBUG_VIRTUAL feature"
2838 depends on DEBUG_VIRTUAL
2840 Test the kernel's ability to detect incorrect calls to
2841 virt_to_phys() done against the non-linear part of the
2842 kernel's virtual address map.
2846 config TEST_MEMCAT_P
2847 tristate "Test memcat_p() helper function"
2849 Test the memcat_p() helper for correctly merging two
2850 pointer arrays together.
2855 tristate "Perform selftest on object aggreration manager"
2859 Enable this option to test object aggregation manager on boot
2863 tristate "Test heap/page initialization"
2865 Test if the kernel is zero-initializing heap and page allocations.
2866 This can be useful to test init_on_alloc and init_on_free features.
2871 tristate "Test HMM (Heterogeneous Memory Management)"
2872 depends on TRANSPARENT_HUGEPAGE
2873 depends on DEVICE_PRIVATE
2877 This is a pseudo device driver solely for testing HMM.
2878 Say M here if you want to build the HMM test module.
2879 Doing so will allow you to run tools/testing/selftest/vm/hmm-tests.
2883 config TEST_FREE_PAGES
2884 tristate "Test freeing pages"
2886 Test that a memory leak does not occur due to a race between
2887 freeing a block of pages and a speculative page reference.
2888 Loading this module is safe if your kernel has the bug fixed.
2889 If the bug is not fixed, it will leak gigabytes of memory and
2890 probably OOM your system.
2893 tristate "Test floating point operations in kernel space"
2894 depends on X86 && !KCOV_INSTRUMENT_ALL
2896 Enable this option to add /sys/kernel/debug/selftest_helpers/test_fpu
2897 which will trigger a sequence of floating point operations. This is used
2898 for self-testing floating point control register setting in
2903 config TEST_CLOCKSOURCE_WATCHDOG
2904 tristate "Test clocksource watchdog in kernel space"
2905 depends on CLOCKSOURCE_WATCHDOG
2907 Enable this option to create a kernel module that will trigger
2908 a test of the clocksource watchdog. This module may be loaded
2909 via modprobe or insmod in which case it will run upon being
2910 loaded, or it may be built in, in which case it will run
2916 tristate "Test module for correctness and stress of objpool"
2918 depends on m && DEBUG_KERNEL
2920 This builds the "test_objpool" module that should be used for
2921 correctness verification and concurrent testings of objects
2922 allocation and reclamation.
2926 endif # RUNTIME_TESTING_MENU
2928 config ARCH_USE_MEMTEST
2931 An architecture should select this when it uses early_memtest()
2932 during boot process.
2936 depends on ARCH_USE_MEMTEST
2938 This option adds a kernel parameter 'memtest', which allows memtest
2939 to be set and executed.
2940 memtest=0, mean disabled; -- default
2941 memtest=1, mean do 1 test pattern;
2943 memtest=17, mean do 17 test patterns.
2944 If you are unsure how to answer this question, answer N.
2948 config HYPERV_TESTING
2949 bool "Microsoft Hyper-V driver testing"
2951 depends on HYPERV && DEBUG_FS
2953 Select this option to enable Hyper-V vmbus testing.
2955 endmenu # "Kernel Testing and Coverage"
2959 config RUST_DEBUG_ASSERTIONS
2960 bool "Debug assertions"
2963 Enables rustc's `-Cdebug-assertions` codegen option.
2965 This flag lets you turn `cfg(debug_assertions)` conditional
2966 compilation on or off. This can be used to enable extra debugging
2967 code in development but not in production. For example, it controls
2968 the behavior of the standard library's `debug_assert!` macro.
2970 Note that this will apply to all Rust code, including `core`.
2974 config RUST_OVERFLOW_CHECKS
2975 bool "Overflow checks"
2979 Enables rustc's `-Coverflow-checks` codegen option.
2981 This flag allows you to control the behavior of runtime integer
2982 overflow. When overflow-checks are enabled, a Rust panic will occur
2985 Note that this will apply to all Rust code, including `core`.
2989 config RUST_BUILD_ASSERT_ALLOW
2990 bool "Allow unoptimized build-time assertions"
2993 Controls how are `build_error!` and `build_assert!` handled during build.
2995 If calls to them exist in the binary, it may indicate a violated invariant
2996 or that the optimizer failed to verify the invariant during compilation.
2998 This should not happen, thus by default the build is aborted. However,
2999 as an escape hatch, you can choose Y here to ignore them during build
3000 and let the check be carried at runtime (with `panic!` being called if
3005 config RUST_KERNEL_DOCTESTS
3006 bool "Doctests for the `kernel` crate" if !KUNIT_ALL_TESTS
3007 depends on RUST && KUNIT=y
3008 default KUNIT_ALL_TESTS
3010 This builds the documentation tests of the `kernel` crate
3013 For more information on KUnit and unit tests in general,
3014 please refer to the KUnit documentation in Documentation/dev-tools/kunit/.
3020 endmenu # Kernel hacking